ArticlePDF Available

Finding Patterns with a Rotten Core: Data Mining for Crime Series with Cores


Abstract and Figures

One of the most challenging problems facing crime analysts is that of identifying crime series, which are sets of crimes committed by the same individual or group. Detecting crime series can be an important step in predictive policing, as knowledge of a pattern can be of paramount importance toward finding the offenders or stopping the pattern. Currently, crime analysts detect crime series manually; our goal is to assist them by providing automated tools for discovering crime series from within a database of crimes. Our approach relies on a key hypothesis that each crime series possesses at least one core of crimes that are very similar to each other, which can be used to characterize the modus operandi (M.O.) of the criminal. Based on this assumption, as long as we find all of the cores in the database, we have found a piece of each crime series. We propose a subspace clustering method, where the subspace is the M.O. of the series. The method has three steps: We first construct a similarity graph to link crimes that are generally similar, second we find cores of crime using an integer linear programming approach, and third we construct the rest of the crime series by merging cores to form the full crime series. To judge whether a set of crimes is indeed a core, we consider both pattern-general similarity, which can be learned from past crime series, and pattern-specific similarity, which is specific to the M.O. of the series and cannot be learned. Our method can be used for general pattern detection beyond crime series detection, as cores exist for patterns in many domains.
Content may be subject to copyright.
Finding Patterns with a Rotten Core:
Data Mining for Crime Series with Cores
Tong Wang,
Cynthia Rudin,
*Daniel Wagner,
and Rich Sevieri
One of the most challenging problems facing crime analysts is that of identifying crime series,whicharesetsof
crimes committed by the same individual or group. Detecting crime series can be an important step in predictive
policing, as knowledge of a pattern can be of paramount importance toward finding the offenders or stopping the
pattern. Currently, crime analysts detect crime series manually; our goal is to assist them by providing automated
tools for discovering crime series from within a database of crimes. Our approach relies on a key hypothesis that
each crime series possesses at least one core of crimes that are very similar to each other, which can be used to
characterize the modus operandi (M.O.) of the criminal. Based on this assumption, as long as we find all of the
cores in the database, we have found a piece of each crime series. We propose a subspace clustering method,
where the subspace is the M.O. of the series. The method has three steps: We first construct a similarity graph to
link crimes that are generally similar, second we find cores of crime using an integer linear programming approach,
and third we construct the rest of the crime series by merging cores to form the full crime series. To judge whether a
set of crimes is indeed a core, we consider both pattern-general similarity,whichcanbelearnedfrompastcrime
series, and pattern-specific similarity, which is specific to the M.O. of the series and cannot be learned. Our method
can be used for general pattern detection beyond crime series detection, as cores exist for patterns in many domains.
Key words: crime series detection; subspace clustering; clustering with feature selection; pattern mining; dense
core sets; similarity graph
One of the most important problems in crime analysis
is that of crime series detection, or the detection of a set
of crimes committed by the same individual or group.
Criminals follow a modus operandi (M.O.) that char-
acterizes their crime series; for instance, some crimi-
nals operate exclusively during the day, others work
at night, some criminals target apartments for house-
breaks, while others target single family houses. Crime an-
alysts need to identify these crime series within police
databases at the same time as they identify the M.O.
Currently, analysts identify crime series by hand:
They manually search through the data using database
queries trying to locate patterns, which (as Ref. notes)
can be very challenging and time-consuming. From a
computational perspective, crime series detection is a
very difficult problem: It is a clustering problem with
cluster-specific feature selection, where the set of fea-
tures for the cluster is the M.O. of the criminal(s).
One cannot know the M.O. of the series without deter-
mining which crimes were involved, and one cannot
locate the set of crimes without knowing the M.O.—
the M.O. and the set of crimes need to be determined
simultaneously. Pattern analysis for crime has existed
at least as far back as the 1840s,
but recently, big
data has changed the whole landscape for crime analy-
sis. To locate crime patterns, analysts now rely on large
amounts of very detailed data about large numbers of
past crimes. The computational problem of finding
crime series grows exponentially with the numbers of
Massachusetts Institute of Technology, Cambridge, Massachusetts.
Cambridge Police Department, Cambridge, Massachusetts.
*Address correspondence to: Cynthia Rudin, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, E-mail:
ªTong Wang et al. 2015; Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons
License (, which permits unrestricted use, distribution, and reproduction in any medium, provided
the original work is properly credited.
Big Data
Volume 3 Number 1, 2015
Mary Ann Liebert, Inc.
DOI: 10.1089/big.2014.0021
crimes and features of crimes. Even if a crime analyst
could identify crime series within his/her own depart-
ment database (which is already difficult), these manual
efforts cannot scale, for instance, when neighboring po-
lice departments start to combine databases.
Despite its critical importance for public safety, there
is little in the way of tools to help police. Most predic-
tive policing software has the capability to detect only
general background levels of crime density in time
and space, which are much easier to predict than spe-
cific patterns of crime. This is called hotspot prediction,
and it requires only time and location data, and a density
estimation algorithm: Crimes are predicted to occur
based on where and when they occurred in the past.
No detailed information about the M.O. of the crimes
is used for hotspot prediction, and hotspots can involve
crimes committed by different offenders (as opposed
to crime series, which have the same offenders). In
fact, at least in the case of Cambridge, Massachusetts,
crime series generally do not take place in hotspots
(see also Ref.
for a study of crime series in two other
U.S. cities). Hotspot prediction software is not what
we need for the problem of crime series detection.
The problem of crime series detection is much more
data intensive and computationally harder than hotspot
prediction. To determine the M.O., we require very fine-
grained detailed information about past crimes: where
the offender(s) entered (front door, window, etc.); how
they entered (pried doors, forced doors, unlocked doors,
pushed in air conditioner, etc.), whether they ransacked
mensional structured data, leading to a computationally
challenging data mining problem of finding all crimes
that are similar to each other in several ways.
Crime series detection results can be used for many
purposes. First, if a pattern is identified, investigative re-
sources can be prioritized to focus on the crime series,
to gather and assemble evidence. For instance, video re-
cordings from nearby streets and stores or banks can be
examined for evidence of the offender’s location with
respect to all crimes in the series, and combined with
suspect descriptions from witnesses (if any). Call-detail
records from cell phones can also be used for this pur-
pose, as well as latent fingerprints and tracking infor-
mation from stolen property. Second, if a current
crime series is localized enough to have predictable
times and locations (for instance, a pickpocket targeting
´at regular intervals throughout the week),
actions can be taken to stop the pattern. Third, if a cur-
rent crime series has the same M.O. as a past crime
series for which the offender is known, then the suspect
from the older series could be a potential offender for
the current crime series. Fourth, crime series detection
results can be used to study criminal behavior gener-
ally. There is strong evidence that a majority of crimes
are committed by a small number of serial criminals,
which underscores the importance of identifying and
studying the patterns of serial offenders.
Conversely, we remark that nothing can be done if
police do not know that a pattern is occurring at all.
Without the capability of automatically detecting a spe-
cific series of crime, it is possible that crime series may
take much longer to identify, or may never be identi-
fied. This is especially problematic for certain types of
crime—for instance, for housebreaks (burglaries) there
is often no suspect information since the crimes take
place when residents are not present. Housebreaks
can be extremely difficult crimes to solve, and nation-
wide only 14% of housebreaks are solved.
In this arti-
cle, we aim directly at identifying series in housebreaks
in an automated way.
The main hypothesis in this work is that most crime
patterns have a core of crimes that exemplify the M.O.
of the series. A core might be approximately three or
four crimes that are very similar to each other in
many different ways. According to our hypothesis that
each crime series has a core, if we can locate all small
cores of crime within the dataset, we have thus located
pieces of most crime series. This hypothesis is based on
the intuition of analysts and has the dual purpose of
assisting with computation: We can indeed consider
all reasonable small subsets to calculate whether they
are plausible cores.
Though we cannot determine the M.O. of a series
before we see it, we can characterize parts of the
M.O.s we expect generally—for instance, crimes in a
series are often close in time and space. We define a
pattern-general similarity that encodes factors that are
generally common to most crime series. It is learned
from past crime series; for instance, proximity in time
and space highly contribute to the pattern-general sim-
ilarity. The pattern-general similarity induces a similar-
ity graph over the set of crimes, where there is an edge
between two crimes when they are similar according to
the pattern-general similarity.
Crimes in a series also have pattern-specific similar-
ity, where crimes are similar to each other if they all
share the same M.O. The pattern-specific similarity
and pattern-general similarity (the similarity graph in
particular) are used for detecting cores: A core must
have both sufficiently large pattern-general similarity
and pattern-specific similarity. The cores are found
using an integer linear programming (ILP) formula-
tion. Once the cores are found, we construct the full
crime series by merging overlapping cores. We also
prove that merging cores preserves desired properties.
Our method is a general subspace clustering method
that can be used beyond crime series, and for other com-
pletely different application domains. It is novel in that it
considers both pattern-general and pattern-specific as-
pects of patterns, where ‘‘pattern-general’’ means that
it is common to many crime series (supervised from
past clusters), and ‘‘pattern-specific’’ meaning aspects
of a particular crime series (unsupervised). Our method
does not force all examples to be part of a cluster and
can thus accommodate background (non-series) crime.
Our method also can find subspace clusters whose sub-
space morphs dynamically over the cluster; an M.O. can
change when an offender becomes more sophisticated,
adapts his M.O. toavoid detection, or when his preferred
method is not available (e.g., he prefers entry through
unlocked rear doors but will push in the air conditioner
if his preferred means of entry is not available).
This project highlights several important aspects of
big data analytics. It highlights an emphasis on human
collaboration within the analysis pipeline, which is a
key challenge for big data (see, for instance, the CCC
white papers
). This project also exemplifies a separate
challenge, namely that of increased complexity. Once
our problem is formulated, solving it requires a combi-
natorially hard optimization problem and an explosion
of variables. We could, for instance, add a ‘‘V’’ for ‘‘Var-
iables’’ to the traditional V’s of big data in order to
capture problems that require massive computation
to solve a problem of increased complexity. The big
data challenge of problems with increased complexity
has arisen in other places (see, for instance, the Amer-
ican Statistical Association’s white paper on big data
Our three methodology sections follow the three
main components of our method: learn a similarity
graph,mine cores,andmerge cores.Wethenshowex-
periments where our method was tested on the
full housebreak database from the Cambridge Police
Department containing detailed information from
thousands of crimes from over a decade. This method
has been able to provide new insights into true patterns
of crime committed in Cambridge. One observation
that has been revealed in our experiments is that com-
puters do not have the same biases that humans do.
Computers search through the database in a different
way than a human would, and thus can work symbiot-
ically with analysts, showing them avenues to consider
where they would not have normally ventured.
Related work
There are at least three main types of recent approaches
to identifying crimes committed by the same individual
or group (according to Ref.
The first approach, sometimes known as pairwise
case linkage, involves identifying whether a pair of
crimes was committed by the same group, where
each pair is considered separately. Some of these works
use weights determined by experts
to weight the
various types of similarities between crimes, and other
works learn the similarity weights from data.
problem with considering only pairwise linkages is
that only one similarity measure between crimes is
used. This has a fundamental flaw that it does not
consider M.O.’s of individual crime series. Consider
two crime series with different M.O.’s: one has an
M.O. in which the means of entry is to push in an
air conditioner on a ground floor apartment (in build-
ings that are not concentrated geographically), and
the M.O. of the other crime series involves breaking
into a single family home at night (not using a consis-
tent entry method) while the residents are present in a
small geographical area. If we are investigating whether
a new crime fits into the first series, logically we should
consider whether the suspect entered by pushing in an
air conditioner of an apartment and we should not
consider geographical area. When we are investigating
the second series, we should consider geographical
area, time of day, and whether residents were present;
we should not consider means or location of entry. If
we use a single measure to judge similarity between
any two crimes (as in pairwise case linkage ap-
proaches), it is not possible to make this distinction.
In pairwise linkage approaches, it is not possible to ig-
nore certain types of information about similarity and
pay attention to others, as we claim is necessary for
understanding multiple different M.O.’s. As a result,
these approaches generally find that time and space
are the only relevant dimensions for this type of anal-
ysis (see for instance Ref.
), and end up ignoring the
detailed behavioral information.
The second type of approach is called reactive linkage,
where a crime series is discovered one crime at a time,
starting from a seed of one or more crimes (as defined
by Refs.
). Reactive linkage has a similar problem to
pairwise linkage in that when we start to grow a set
of crimes greedily from a small seed of one or two
crimes, we cannot yet know the M.O. of the crime se-
ries. This means that yet again we start from a common
similarity metric between crimes. The greediness of this
approach, coupled with the fact that the distance metric
does not take into account the M.O., could lead to
problematic results. Our past approach to this
use a theoretically motivated
greedy method, but
adapted the distance metric to be closer to the observed
M.O. as more crimes were added to the set. This still
does not solve the problem of greediness, and the result
could depend on which crimes were used as seed crimes.
On the other hand, these greedy methods are very com-
putationally tractable.
In the last type of approach, crime series clustering,all
the clusters are found simultaneously.
One of the
earliest approaches we know of for clustering crimes is
that of Dahbur and Muscarello,
who used a neural net-
work approach. (This method had some serious flaws
that required extensive heuristic post-processing after
the clusters were created, but aimed at solving the
more general problem of crime clustering.) Most of
these approaches use a form of hierarchical clustering,
which again has the disadvantage that the distance met-
ric between crimes is static and does not reflect the M.O.
of any particular crime series. (On the other hand, hier-
archical clustering is very computationally tractable
and could be made to reflect pattern-general similar-
ity, though not pattern-specific similarity.) Some of
these approaches reduce features one at a time through
hypothesis tests, or use basic dimensionality reduction
(multidimensional scaling) techniques before cluster-
ing. This still does not handle pattern-specific aspects,
and thus, cannot capture the M.O.
A main point of the present work is that in order to
model the M.O., we need to use some form of subspace
clustering. The M.O. for the pattern is precisely the sub-
space; it is the set of dimensions for which crimes in a
particular series should be considered to be similar.
For the first crime series in the example above, we
entry’’ involving the entry by pushing in air condition-
ers, and we would not heavily consider the dimension
for geographical area. The algorithm must determine
which observations go into which cluster, and which
subspace is relevant for each cluster. Our work relates
to various subfields of clustering (see, for instance,
), including pattern-based clustering,
is a semi-supervised approach (unlike ours—we do not
use test data at training time); general subspace cluster-
ing (e.g., Refs.
), which detects all clusters in all sub-
spaces; and work at the intersection of dense subgraph
mining and pattern mining in feature graphs.
These methods would not be able to take into account
the complexities we handle, for instance, learning the
pattern-general weights, or finding cores with the char-
acteristics we require. Other work on space-time event
detection is relevant,
where the goal is to detect pat-
terns such that the frequency of records is higher than
an expected frequency. A relevant method for cluster-
ing with simultaneous feature selection is that of
Guan et al.,
which is similar to our core detector in
flavor, although feature selection is controlled differ-
ently, and there are no pattern-general aspects.
Reviews of other data mining applications in crime
analysis are those of Chen et al.
and Thongtae and
There are several other works that use machine
learning for various criminology applications.
Model Formulation
Our data consist of entities (crimes) V, each of which
has a vector of features. We define D
as the set of pos-
sible values for feature j,j=1...J. In our specific da-
tabase we have features including time, space, whether
the crime occurred on a weekday, location of entry
(front door, back door, ground window, etc.), means
of entry (shoved, pried, forced, etc.), type of premise
(apartment, single-family house, etc.), indicator vari-
able for ‘‘ransacked,’’ suspect information (which is
rarely present), and so on. Although the approach we
introduce below is general and can be applied to prob-
lems beyond crime data mining, we use terminology
specific to our problem in our exposition.
Define s
to be a symmetric similarity function on the
jth feature s
/[0, 1], where s
) measures
the similarity between crimes v
and v
in the jth fea-
ture. In our case most features are categorical, some
are ranges, and some are spatial coordinates. For in-
stance, if two crimes have location of entry as ‘‘win-
dow,’’ then they would get a high location-of-entry
similarity. The similarity measures are discussed in
depth in our earlier work,
so we will not go into
detail here. The average similarity of a set of crimes
in feature jis defined as the cohesion of the set:
Definition 1. (Cohesion
) For a set of crimes V, the
cohesion in the jth feature is the mean of pairwise
The defining features of a pattern are those with suf-
ficiently high cohesion.
Definition 2. (Defining feature) Defining features
for V are those that satisfy Cohesion
. The set
of defining features for set V is denoted by L(V).
The defining features characterize the M.O. of the
crime series. If several housebreaks happen in the
same neighborhood, around the same time of day,
within the same month, and the location of entry is al-
ways a window, regardless of the differences in other
features, these similarities would indicate that the crimes
could have been committed by the same offender. The
features ‘‘geographic location,’’ ‘‘time of day,’’ ‘‘time be-
tween crimes,’’ and ‘‘location of entry’’ characterize the
M.O. for this particular crime series. When we later de-
fine cores, the pattern-specific statistic of interest is the
number of defining features of V.
Let us switch from pattern-specific definitions to
pattern-general definitions. We will learn a pattern-
general similarity function from past crime series.
Pattern-general similarity allows us to weight impor-
tant features highly; for instance, crimes that are
spread very far apart in time and space are unlikely
to be a pattern, and thus we will learn from past
tures. We will learn a set of weights [k1,...,kj,
...kJ]from past crime data that will provide the im-
portance of each feature within a linear combination.
We will search for cores that have high pattern-gen-
eral cohesion, defined as follows:
Definition 3. (Pattern-general cohesion) The pattern-
general cohesion of V is the weighted sum of cohesions
over the features: +J
Our cores will also be connected in the similarity
graph. To construct edges for the graph, we first define
a metric to measure the similarity between two crimes,
as follows:
Definition 4. (Pattern-general similarity) The
pattern-general similarity is a weighted sum of similar-
ity measures for each feature, using the pattern-general
weights [k1,k2,...kJ].
We define the similarity graph to contain edges be-
tween crimes that are sufficiently close in the pattern-
general sense.
Definition 5. (Similarity graph) A similarity graph is an
undirected graph G=(V,E),where V=fv1,v2,...vngis
thenodeset,Eis the edge set, E=ffv,vkgjc(v,vk)q
For us to even consider whether Vshould be a core
of a series, the core needs to be connected in the graph
theoretic sense (and is not composed of two separate
clusters for instance), must have sufficient connectivity,
and must have sufficiently high pattern general cohe-
sion. The requirement that a core must be a connected
set means that each crime in a core must be similar in
pattern-general similarity to at least one other crime in
the core.
We learn both the weights {k
and cut-off threshold
Dfrom data, as discussed in Section 3 below, in order to
perform the first piece of the method, which is to con-
struct the similarity graph. Note that we could eliminate
the first piece of the method, and eliminate the pattern-
general similarity all together by setting the threshold D
to be very low, and that way we consider only pattern-
specific similarity; this, however, would not ease compu-
tation or restrict the cores to be similar to those from
other patterns. Imposing a graph structure on the crimes
eases computation, in the sense that we are now only
looking for connected sets of the graph when we mine
for cores in the second part of the method.
The second part of the method is to mine for cores.
When we mine for cores, we cannot simply maximize
the number of defining features and/or the pattern-
general cohesion over all subsets of crime, as it would
favor choosing very small sets of crime as cores. To al-
leviate this problem, we specify the size of the core jVj
and the number of defining features d, and find cores
that maximize the pattern-general cohesion. Once we
find the cores, we merge them to find the rest of the
pattern. We call patterns formed by merging other pat-
terns together serieslike patterns. An illustration of a
serieslike pattern is shown in Figure 1.
Learning the Similarity Graph
As we discussed, the similarity graph is constructed by
connecting pairs of nodes with pattern-general cohesion
above a threshold D. The pattern-general cohesion cis
defined in (1) as a weighted sum of pairwise similarities
in different features, with pattern-general weights
J. The set of coefficients kand Dare parameters
that we learn from data. These data consist of 51 histor-
ical crime series that have been identified as true crime
series by crime analysts.
To learn the weights, we optimize over the historical
training patterns to make them as close as possible to
being connected subgraphs. This means we want each
crime in a historical pattern to be close to at least one
other crime in the same pattern. At the same time,
we want crimes within a historical pattern to be distant
from crimes not in the same pattern. The condition for
crime to be close to at least one other crime within its
pattern (with some slack e
) is:
Conversely, crimes that do not belong to the same
pattern should not be very similar:
kjsj(,k)pDþnk, (3)
true for all and ksuch that k=2pattern(). Our goal is
to minimize the total weighted slack. Informally, we
crimes in the
same pattern
slack þC1+
crimes not in the
same pattern
slack þC0kkk0,
where kkk
is the
semi-norm of k, which encourages
sparsity in k. The constant C
is set very small, so most
edges that should be present will be present, and we
consider removal of unnecessary edges as a secondary
goal. Removal of these unnecessary edges is where
we will gain a computational benefit later on. The
more edges we eliminate, the fewer connected subsets
we need to evaluate as being possible cores.
We propose a mixed-integer programming (MIP)
formulation for solving this. In what follows, decision
variables y
are binary, and they select a crime kthat
is most similar to a given crime within the same pat-
tern. That is, they encode the max from constraint (2).
The formulation is:
such that
kjsj(,k)q(D)þM(yk1) 88ks:t:k2pattern()
yk2f0, 1g8,k(6)
kjsj(,k)pDþnk8,8ks:t:k=2pattern() (7)
kj=1 (8)
bj2f0, 1g8j:(11)
Constraint (4) comes from (2). It forces y
to be cho-
sen correctly so that if kis the crime closest to ,theny
will be 1. This is because e
is minimized within the ob-
jective, so y
is necessarily going to correspond to the
index where e
is minimized, and where the similarity
is maximized within (4). Constraints (5) and (6) further
define y
’s by stating that a crime in the pattern needs
only to be connected to one closest neighbor in the
pattern (this is the requirement of connectivity), and
its entries are binary. Constraint (7) comes from (3).
The value +
is the
norm of k.Theb
’s are decision
variables where if b
=1, k
is nonzero. This formulation
is linear, and thus using MIP technology there is a guar-
antee on the optimality of the solution. In particular, the
solver will provide the duality gap, and when it is zero,
we know that the optimal solution to the optimization
problem has been attained.
FIG. 1. Adserieslike pattern consisting of three
dcores of various sizes.
Ad-core is a set of crimes that exhibit similarity in a
feature subspace of ddimensions. A d-core has dde-
fining features that are not predetermined. Further,
crimes in a d-core need to be well connected in the
similarity graph. Formally, the definition is as follows.
Definition 6. (d-core) A similarity graph G =(V,E)
with density threshold ais called a d-core if it satisfies
the core constraints:
Pattern specific constraint: the size of the defining
feature set of the graph is equal to d, jL(G)j=d.
Pattern general constraints: G is connected, and G
is dense, jEj
jVj(jVj1) qa. That is, the fraction of pos-
sible edges in the graph exceeds a.
Two parameters, dand a, control the property of
the core in a pattern-specific and pattern-general way,
The pattern-general constraints should be thought
of as being much looser than the pattern-specific con-
straints, as we often include unnecessary edges in the
similarity graph. For the set of crimes to be feasible
in the pattern-specific sense is much more difficult, as
crimes in the pattern need to be similar to each other
in dseparate ways.
We find cores G=(V,E) using an optimization method
to maximize the pattern-general cohesion while satisfy-
ing the core constraints.
subject to jVj=n
core constraints :
Gis connected,
jVj(jVj1) qa:
We propose a binary integer linear formulation for the
optimization problem (12). Let m=jVj, the number of
crimes in the database. Let nbe the size of the pattern
we want to discover, and we loop through possible val-
ues of n, re-solving each time. We define an m·msim-
ilarity matrix for each feature S1,...,SJwith elements
), which are precomputed. Let Xbe the
matrix of binary decision variables defining the core.
X(,k) is 1 if a pair of crimes and kare in core G,
which is the same as X(k,), so matrix Xis symmetric.
On the diagonal, X(,) represents whether crime
is in core G. We use d1,...,dJ2f0, 1gto indicate
whether feature jis a defining feature, or equivalently,
whether Cohesion
. Let Ebe the adjacency ma-
trix for the (pattern-general) similarity graph, where
E(,k)=1iffv,vkg2E, and E(,k)=0 otherwise.
Since the graph is undirected, Eis symmetric. We
set E(,)=1 for computational simplicity. (EsX)is
the Hadamard product of matrix Eand X, where
(EsX)(,k)=E(,k)$X(,k). Mis a large auxiliary
parameter for formulating the problem with a big-M
formulation, and eis small. With this notation, the
optimization problem (12) can be reformulated:
n(n1) +
n(n1) +
X(,k), dj2f0, 1g8,k,j(21)
where matrix ^
Eis defined just below.
Let us derive the objective. Since X(,k)=1 if and only
if both crimes and kare in the core (that is, they are in
graph Gthat we discover), we have the following:
n(n1) +
The objective is the pattern-general cohesion. Equation
(13) ensures that the cores we discover are of size n.
Constraints (14), (15), and (16) are the pattern-
specific constraints. Constraint (14) forces d
=1 when
, where the strict inequality is enforced
by e. Constraint (15) forces d
=0 when Cohesion
. Constraint (16) specifies the number of defining
features as d. The symmetry of Xis enforced by (17).
Constraints (18) and (19) imply X(,k)=1 iff both
X(,) and X(k,k) are 1 and 0 otherwise.
Expression (20) is a pattern-general constraint. Our
formulation does not enforce the core to be connected,
but it does enforce something weaker, namely that each
node in a core of size nis at most n1 steps along the
similarity graph from any other node in the core. We
handle the connectivity afterward by examining the re-
sult to ensure that it is connected and labeling it as in-
feasible if not. To handle the constraint that each node
in the core is at most distance n1 from every other
node, we recall that in graph theory, if node v
is reach-
able from node v
in exactly qsteps, E
(k,)>0. If node
is reachable from node v
in at most n1 steps, it
means that at least one of E
(k,)>0 for q£n1.
We define the following matrix ^
En1, where an ele-
ment ^
En1(k,)indicates if node kand node are dis-
tance at most n– 1 steps along the graph.
1(EþE2þ  þEn1)(,k)>0
Thus, (20) forces that if v
and v
are both in the pattern,
they must be at most distance n– 1 along the graph.
The pattern-general density constraint is handled
similarly to the connectivity constraint, where feasibil-
ity is checked for each solution, and infeasible solutions
are removed.
The integer program finds one solution at a time. In
order to avoid finding a solution that was found previ-
ously, we introduce a constraint for each previous solu-
tion found. Suppose in the t-th run, the crimes in the
solution are Qt, that is, X(k,k)=1ifk2Q
k)=0 otherwise. The constraint we add before running
the t+1-th time is
This constraint will exclude the current solution
from the feasible region, and we will obtain a different
solution in the next run if any feasible solutions remain.
Since all of the matrices are symmetric, in practice we
keep only the upper (or lower) triangle, including the
diagonal, to compute all of the sums.
Merging Cores
By our main hypothesis, the vast majority of crime se-
ries contain a core. This means that by finding all cores,
we would have located the vast majority of all crime se-
ries. We now grow the rest of the crime series from the
cores by merging them together. One advantage of
merging is that it allows the pattern to dynamically
change, as the defining features from the merged
cores are not always equal to each other. Consider a
burglar’s M.O. with a shifting means of entry. At first
he enters through unlocked doors, then he starts to
use bodily force to open doors, and later he learns to
use a screwdriver to pry the door open. His full pattern
thus consists of several smaller d-cores. This suggests a
more flexible definition of pattern than a simple core.
We provide such a definition below.
Definition 7. (d-serieslike pattern) A graph G =(V,
E)is called a d-serieslike pattern with defining feature
set P(G)of size d if it satisfies:
Pattern general constraint: G is connected.
Pattern specific constraint: Each node u in G is con-
tained in at least one subgraph G¢4G that is a d¢-
core with defining features that include set P(G),
that is,P(G)4L(G¢),d£d¢.
Note that if a graph is a d-serieslike pattern, it is also a
(d– 1)-serieslike pattern, a (d– 2)-serieslike pattern, and
so on. The pattern specific constraints for d-serieslike
patterns are looser than those for d-cores. A d-serieslike
pattern may not be a d-core. On the other hand, a d-
core is a special case of a d-serieslike pattern. Before
we proceed, we must ensure that merging is justified.
Theorem 1. (The set of serieslike patterns is closed
under merging.) Suppose G
is a d
-serieslike pattern
and G
is a d
-serieslike pattern. If G
s;, then
G=G1[G2is a d-serieslike pattern, with defining fea-
tures P(G
Proof. First, since G
and G
are connected and
s;, the union of them is also connected, that
is, ^
Gsatisfies the pattern general constraint. Then for
all nodes u2G1,dad
-core G1
usuch that u2G1
and P(G1)\P(G2)P(G1)L(G1
u), and for all
nodes u2G2,dad
-core G2
usuch that u2G2
u). So, either way, the
defining feature set includes P(G
). This means
that for any node ^
G,da core ^
Gusuch that
This leads directly to the following:
Corollary 1. Suppose G
is a d
-serieslike pattern, G
is a d
-serieslike pattern, .,G
is a d
-serieslike pattern,
and G1[...[Gnis connected,
Then G1[...[Gnis a d-serieslike pattern.
These properties lead to the following breadth first
search algorithm for mining d-serieslike patterns. We
start with the cores that we found using the integer
program. These cores are candidates for merging. We
also maintain an active pattern set that contains the d-
serieslike patterns that we are not done constructing.
We keep the d-serieslike patterns that we are done con-
structing in a maximal pattern set. To start, the active
pattern set contains all of the d-cores we found. For
each active pattern, we iterate through the candidates
to see if they meet the merging criteria provided just
below. If a merge is possible, and if the merged set
had not been previously created, we append the merged
pattern to the active pattern set and continue iterating
through the candidates. If there are no candidates that
can be merged with the active pattern at all, then the
active pattern is maximal, and it is placed in the max-
imal pattern list. The merging criteria for G
form a d-serieslike pattern ^
The merging algorithm is formulated in Algorithm 1.
Algorithm 1: Merging Cores
INPUT: d, cores, each with ddefining features
candidate list)cores
active set)cores, each with defining features
maximal set);
while active sets;do
)any element in active set
)defining features from G
for all G
2candidates, Gj6 Gcurrent do
if G
G)Gcurrent [Gj
if ^
Gdoes not exist in active set or maximal set then
append ^
Gto active set
end if
end if
end for
if isMaximal ==TRUE then
remove G
from active set, put into maximal set
end if
end while
OUTPUT: maximal set
Our data set was provided by the Crime Analysis Unit
of the Cambridge Police Department in Massachusetts.
It has 7,067 housebreaks that happened in Cambridge
between 1997 and 2011, containing 51 hand-curated
patterns contained within the 4,864 crimes between
1997 and 2006. (Patterns from 2007 to 2012 were not
assembled at the time of writing.) Crime attributes
include geographic location, date, day of week, time
frame, location of entry, means of entry, an indicator
for ‘‘ransacked,’’ type of premise, an indicator for
whether residents were present, and suspect and victim
information. The 51 crime series identified by police
contain an average of 12.1 crimes each, with the largest
series containing 59 crimes, and the smallest series con-
taining 2 crimes. These crimes span an average period
of 42 days, with the shortest series taking place within 1
day and the longest series taking 451 days. Data were
processed using the similarity functions {s
in our previous work,
where each pairwise feature
is mapped into a number between 0 and 1. These sim-
ilarity measures are p-values, and they consider the
baseline frequency of each possible outcome for the
categorical variables. For instance, most crimes are
committed when residents are not present. If two crimes
were committed where one had residents present and
the other had residents that were not present, then
the similarity score for ‘‘residents present’’ is zero. If
both crimes were committed when the residents were
not present, the similarity score for ‘‘residents present’’
would not be high (in particular, it is 1p2
not present
where p
not present
is the proportion of crimes in the
database where residents were not present), whereas
if two crimes were committed with residents present,
the similarity would be much higher (1p2
The similarity score for time frames is complicated be-
cause it takes into account the distribution of times
when crimes are more frequently committed. We took
the 51 hand-curated patterns and divided them ran-
domly into four subsets (folds) with sizes 12 or 13 pat-
terns each. We used three of the four folds to learn the
pattern-general weights and tested on the remaining
fold for the experiments discussed below.
As this problem is fundamentally a clustering prob-
lem, we compare with several varieties of hierarchi-
cal agglomerative clustering and incremental nearest
neighbor approaches. For these baselines, we use sev-
eral different schemes to iteratively add discovered
crimes, starting from pairs of nodes with high sim-
ilarity c, which is a weighted sum of the attribute
Unlike our method where the weights are learned, the
weights ^
kfor the baselines were provided by crime an-
alysts based on domain expertise, similar to several
other works.
Hierarchical agglomerative clustering (HAC) begins
with each crime as a singleton cluster, and iteratively
merges the clusters based on the similarity measure be-
tween clusters. Nearest neighbor classification (NN)
first selects pairs of crimes with high similarity and
then iteratively grows a cluster by adding the nearest
neighbor crime to the cluster.
HAC and NN were used with three different crite-
ria for cluster–cluster or cluster–crime similarity:
single linkage (SL), which considers the most similar
pair of crimes; complete linkage (CL), which consid-
ers the most dissimilar pair of crimes; and group
average (GA), which uses the averaged pairwise simi-
When the nearest neighbor algorithm is
used with the S
measure defined below with
weights provided by crime analysts, it is similar to
the Bayesian Sets algorithm and how it is used for
set expansion.
Evaluation metrics
There are two levels of performance we evaluate -
pattern-level and object-level.
Pattern-level precision and recall
We evaluate the quality of the core detector using
pattern-level precision and recall. The d-cores are
smaller as dbecomes larger. The cores are used for
discovering larger merged patterns. Thus we evaluate
the accuracy of the core finder in its detection ability;
if a real pattern is missed completely by our core de-
tector, there is no way to recover from this in order
to detect it. If a core covers more than one pattern,
this is also a bad seed for further mining, since it
would generate misleading defining features that do
not characterize any real patterns. Thus, we call
cores that cover one and only one real pattern good
cores. The pattern-level precision and recall are both
defined using good cores. Ndenotes the number of
cores we discover.
P-Precision (cores) =+N
i1(core iis good)
P-Recall (cores) =+N
i1(core iis good)
jPj :(26)
Note that pattern-level precision should be large, as
each real pattern should contain many cores, inflating
the reported precision values.
Object-level precision and recall
We evaluate the full pipeline for generating serieslike
patterns using object-level precision and recall. To do
this, for each pattern discovered, we determine how
close it is to one of the real patterns. If the discovered
pattern overlaps only one real pattern, then we call
this the dominating pattern and evaluate precision
and recall with respect to crimes in that pattern. If
the serieslike pattern overlaps more than one real pat-
tern, we assign the dominating pattern to be the real
pattern possessing the most crimes that overlap with
our discovered pattern. Note that it is possible for the
recall not to grow with the size of the discovered pat-
tern, as the dominating real pattern could change as
the discovered pattern grows larger. The definitions
of object-level precision and recall for a d-serieslike
pattern G=(V,E) are as follows:
O-Precision(G) =+jVj
=11(2dominating pattern)
O-Recall(G) =+jVj
=11(2dominating pattern)
jVdominating patternj(28)
where jV
dominating pattern
jis the number of crimes in the
dominating real pattern.
Computational gain from similarity graph
The first step in our method is to learn the similarity
graph. The similarity graph provides a computational
gain in that it creates constraints on possible cores, re-
ducing the feasibility region of the ILP. For this similar-
ity graph, recall that we desire crimes in the same real
pattern to be connected to each other. We call edges
connecting crimes that belong to the same pattern
good edges. If we have constructed the similarity graph
well, the similarity graph should have a higher percent-
age of good edges than if we had simply used the full
graph consisting of all possible edges. If we remove a
few good edges in the process, this is not problematic
as long as the true patterns are still connected in the sim-
ilarity graph—this will be assessed when we assess the
quality of the cores and the full pipeline next.
Table 1 shows the percentages of good edges in both
the similarity graph and the full graph for four test
folds (the data were divided into four folds, and each
was used in turn as the test fold). The learning method
tends to reduce the number of unnecessary edges by a
factor of 7 or 8 in each of the test folds, as shown in the
third column (which is the first column divided by the
second column). This reduction substantially reduces
computation for the core finder.
Pattern-general weights
The pattern-general weights come from the learning
step for the similarity graph. In Figure 2 we report
the mean over the test folds of the pattern-general
weights we discovered. The highest weights are similar-
ity in distance, number of days apart, suspect informa-
tion, whether residents are present, and means of entry
(e.g., pried, forced, cut screen).
Time windowing
Consider solving the ILP for finding cores on data from
4,864 housebreaks. Note that if we were to search for
patterns of size 10 among 1,000 crimes, this would
mean investigating 1000
2:634 ·1023 possible sub-
sets. Further, CPLEX would need to handle 1,000
straints (18) and (19) of the optimization problem.
Luckily it is unlikely that a crime series would possess
a core of size 10, but still we need to find ways to reduce
Because the pattern-general weight on closeness in
time is so high, we determined that we would be un-
likely to miss true cores if we considered windows of
time that include at least 200 crimes. We thus indexed
the housebreak records in chronological order and
created overlapping windowed blocks of 200 crimes
each, where neighboring blocks have an overlap of
100 crimes. Therefore we solve ILPs among crime sub-
sets f1, , 200g,f100, , 300g,,f4700, ,
4864g. In each subset, we input the number of defining
features dand core size n, and then iteratively run the
ILP to get all feasible solutions by adding the constraint
(23) after each iteration to avoid returning repeated
Evaluation of mining cores
We chose performance evaluation metrics from informa-
tion retrieval, and for some of these metrics, we need to
rank the discovered cores by a scoring function. This scor-
ing function represents how certain we are that these
cores are real. We use a scoring function that is a weighted
version of pattern-general cohesion and the (pattern-
specific) number of defining features, as we desire cores
that are both tight in the pattern-general sense and in
the pattern-specific sense. Here is the score function:
Series usually have about six defining features, so the
choice of 1/6 tends to balance the two terms, weighing
the pattern-specific term slightly higher. We ordered
the discovered cores in decreasing order of the scores.
Note that the first evaluation below does not require
these scores, but the second and third do.
1) Cores with different dWe expect that discov-
ered patterns with more defining features are
more likely to be true crime series. Figure 3
shows how the precision increases with the num-
ber of defining features d. In this figure, we con-
sider only cores of the same size (three crimes).
There are no overlapping cores between the
three bars, as each core is used once with its
exact number of defining features d, which is 6,
Table 1. Test Results of Weights Learning Algorithm
Good edges
% in similarity
Good edges
% in complete
1 1.84 0.26 7.1 3.64 ·10
2 2.42 0.33 7.3 9.58 ·10
3 3.34 0.42 8.0 6.20 ·10
4 2.95 0.34 8.7 5.41 ·10
FIG. 2. Pattern-general weights.
7, or 8. The number of discovered cores for d=6is
1072, d=7 is 215, and for d=8 is 36. There were
too few cores with more than eight defining fea-
tures to reliably calculate precision. The reported
numbers of cores are totals from all test folds.
2) Cores with different size nWe used the scor-
ing function (29) as a filter to pick the best
1,000 cores, from each of the sizes 3, 4, and 5,
and discarded the other discovered cores. Larger
cores have higher chances of hitting a pattern,
since there are more crimes in the core; however,
they also have higher chances of hitting more
than one real pattern. As shown in Figure 4,
cores of size 4 have a much higher pattern-level
precision than cores of size 3, but cores of size
5 do not have noticeable gains over size 4. This
is because the increased probability of hitting a
real pattern cancels with the increased probabil-
ity of hitting more than one real pattern.
3) P-precision–p-recall curve We generated a full
list of cores of size 3 with dbetween 6 and 8, and
ranked the cores according to their scores. As we
moved down the list, we evaluated pattern-level
precision and recall at each step. We also did
the same procedure with the baseline iterative
nearest neighbor method used for generating
cores of size 3, using all of the similarity measures
in (24). (Note that for HAC we cannot control
the size of cores for evaluation.) The precision-
recall curves for all four methods averaged over
the test folds are plotted in Figure 5. It is clear
that our core finder is substantially better than
the baselines, though that is not surprising
given that it searches globally for the best cores.
Evaluation for mining serieslike patterns
We evaluated the quality of our full pipeline and the
baseline methods as follows. After all the serieslike pat-
terns were discovered, we evaluated the average object-
level precision and recall for all the patterns and over all
the test folds, plotted as a point on Figure 6. For HAC,
we simply iterated it, stopping at a threshold where re-
call was approximately equivalent to our method, and
again reported average object-level precision and recall
on Figure 6. For the nearest neighbor method, after
each element was added to a growing pattern, we eval-
uated precision and recall to trace out a precision-recall
curve. All three metrics in (24) were used for HAC and
nearest neighbors. We note that for the same level of
recall, the precision attained by our method was quite
a bit higher than that of other methods.
There is still a lot of room for improvement. Cur-
rently, with precision on the order of 53%, we capture
approximately 18% of the crimes identified by analysts.
That is, when our method returns a crime, it is a real
crime in a series about 53% of the time. Weare returning
about one-fifth of the crimes at these settings, so our
6 7 8
Number of definin
features d
Pattern level precision
FIG. 3. Pattern-level average precision of dcores
of different sizes.
3 4 5
Size of core sets
Precision for best 1000 core sets
FIG. 4. Pattern-level precision of cores with
different sizes.
0 0.2 0.4 0.6 0.8 1
Pattern level recall
Pattern level precision
Core Sets
FIG. 5. Average pattern-level precision vs. recall.
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Object level recall
Object level precision
d series−like patterns
FIG. 6. Object-level precision and recall.
FIG. 7. The locations of crimes in the first series.
method currently is conservative–it will not claim that a
crime is in a series unless it is reasonably certain. This
can help ensure that analysts do not overlook crimes
that are very likely to be in a particular series. Note
that our ground truth consists of crime series that are
hand-labeled by analysts, and these labels are not perfect.
For instance, it is entirely possible that we return crimes
that are in a series that the police had not previously con-
sidered. We might opt to change the settings so that the
method returns more crimes as being potentially part of
the series (giving lower precision but higher recall). This
might be changed by making the core finder less conser-
vative, finding a way to incorporate crimes that are not in
cores, or possibly by working with domain experts to im-
prove the database used for evaluation. We note that the
baseline methods work surprisingly well and are them-
selves reasonable options.
Case Studies
We performed a blind test, where we aimed to detect
crime patterns between 2007 to 2012, for which we
do not have pattern data. The results were analyzed
by hand by crime analysts.
Case study 1
One particularly interesting crime series includes 10
crimes from November 2006 to March 2007. Figure 7
shows geographically where these crimes were located.
Table 2 provides some of the details about the crimes
within the series.
From the (pattern-general) similarity graph, we iso-
lated the subgraph containing the 10 crimes, which is
diagrammed in Figure 8. Crimes 1 to 5 are well con-
nected as a subset, and crimes 6 to 10 are well con-
nected as another subset. From only the similarity
graph, the two subsets do not seem very related except
for a single edge between crimes {5, 6} connecting
them; however, this is only the pattern-general part
of the story.
We used the integer linear program (12) to discover
cores of size 3 with at least 6 defining features. Table 3
lists the cores and their defining features in this series,
where a check mark means the feature is a defining fea-
ture and a circle means it is not. These cores show how
the crimes are similar to each other in a pattern-specific
way. The next step is merging the cores. The defining
feature set P(G) was chosen to include six features,
which are geographic location, days apart, location of
entry, the ransacked indicator, time of day, and day of
the week. One of the cores, core 16 in Table 3, which
was not included as geographic location is not a defin-
ing feature for that core, and the rest of the cores were
merged. (The same set of crimes is in the merged pattern
regardless of whether core 16 was used for the merge.)
As these data were reconsidered by crime analysts,
we found out that when these crimes were analyzed
back in 2006–2007, they were viewed as two unrelated
patterns, one at the end of 2006, crimes 1 to 5, and one
at the beginning of 2007, crimes 6 to 10. The connec-
tion between these two subsets of crime is very sub-
tle, and there is over a month gap between the two
Table 2. Example 1 of a Serieslike Pattern with d=6
No. Date
of entry
of entry Premises Ransacked Residents Time of day Day Suspect Victim
1 11/8/06 Basement door Unknown Unknown No Not in 10:45–15:00 Wed Null 1 F
2 11/8/06 Front door Pried Unknown No Not in 8:00–18:30 Wed Null 1 M
3 11/16/06 Front door Shoved/forced Unknown No Not in 9:00–17:00 Thur Null 1 M
4 12/7/06 Front door Pried Unknown No Not in 9:00–17:00 Thur Null 1 F
5 12/22/06 Front door Pried Unknown No In 11:48 Fri Null 1 M
6 2/1/07 Front door Shoved/forced Unknown No In 14:45 Thur 3 Males 1 F
7 2/15/07 Front door Unknown Aptment No In 12:00–13:30 Thur Null 2 F
8 3/5/07 Front door Shoved/forced Aptment No Not in 12:22–14:56 Mon Null White M
9 3/5/07 Front door Broke Aptment No Not in 12:22–14:56 Mon Null 1 F & 1 M
10 3/8/07 Front door Pried Aptment No Not in 12:50–13:30 Thur Null 1 M
F, female; M, male.
FIG. 8. Similarity graph for the first crime series.
patterns, so it did not occur to the crime analysts to link
them. Their intuition agrees completely with the simi-
larity graph, as the two subsets are weakly connected
only by one edge; however, recall that this only de-
scribes the pattern-general similarity–what one would
expect from generic pattern without considering a spe-
cific M.O. On examination of the cores, not only are
they correlated in six features, but five of the cores
(core indices 11, 12, 14, 15, 16) contain crimes from
both of the subsets, which is strong evidence that the
two subsets should be merged together. It is particu-
larly interesting that the core consisting of crimes 5,
6, and 7 spanned the two subsets, where these crimes
share the unusual feature that residents were present
during the break-in.
We wondered why the offenders changed their M.O.
to move a few blocks north since they committed
crimes 1–6 in the same area. What may have happened
is that the criminals left near the end of December (just
before the holidays), and returned in February to com-
mit housebreak number 6 in the same area as 1–5;
however, they were witnessed committing crime 6 (sus-
pect information in crime 6 reads ‘‘3 males’’). This may
have spooked the offenders, causing them to alter their
M.O. by moving north to commit crimes 7–10. Ana-
lysts now believe that these two series were actually a
single series, and that the suspect information from
crime 6 can be carried through to all the crimes in
the discovered series.
This is a good example to show how crime patterns
can be composed of cores and exhibit similarity both in
a pattern-general way and a pattern-specific way. It
shows how we can use both aspects to mine patterns.
This is a pattern that would be very difficult for a
crime analyst to find: the M.O. changes over time,
there was a long break in the middle of the series,
and there was nothing deterministic (e.g., fingerprints)
linking these crimes.
Table 3. Cores and Their Defining Features for Example 1
Cores Crimes
Loc of
of entry Premises Ransacked Residents
of day Day Suspect Victim
10 2 4 5 XX X X OXOXX OO
11 2 4 6 XX X OO XOXX OO
12 2 5 6 XX X OO XOXX OO
13 3 4 5 XX X OO XOXX OO
14 3 5 6 XX X OO XOXX OO
15 4 5 6 XX X OO XOXX OO
16 5 6 7 OXX OO XXXXOO
17 6 8 9 XX X OO XOXX OO
18 6 8 10 XX X OO XOXX OO
19 6 9 10 XX X OO XOXX OO
20 7 8 9 XX X OXX OXX OO
21 7 8 10 XX X OXX OXX OO
22 7 9 10 XX X OXX OXX OO
23 8 9 10 XX X OXX OXX OO
Table 4. Data from the Second Crime Series
No. Date
of entry Means of entry Premises Ransacked Residents Time of day Day Suspect Victim
1 1/25/07 Front Door Punched/Popped Apartment No Not in 10:20–12:00 Thur null 1 F
2 1/25/07 Unknown Cut Screen Apartment No Not in 8:45–14:30 Thur null 1 M
3 1/25/07 Unknown Pried Apartment No Not in 9:10–21:00 Thur null 1 F
4 1/25/07 Front door Unlocked Apartment No In 13:00–13:30 Mon null 1 F
5 1/29/07 Front door Key Apartment No Not in 14:52–14:52 Mon null 2 M & 3 F
6 1/29/07 Unknown Unknown Apartment No Not in 12:00–12:00 Mon 1 M 1 M
7 1/29/07 Unknown Unknown Apartment No Not in 15:00–15:00 Mon null 2 M
Case study 2
Table 4 and Figure 9 show a more typical pattern
in 2007 discovered by our method. The crimes were
committed on two dates in late January 2007, most of
them in the same building. According to the Cam-
bridge police, they arrested a suspect while he was com-
mitting the last crime in the series and confirmed that
he did commit crimes 2–7. Note that the record for the
fourth crime records the means of entry as ‘‘key.’’ This
is based only on the claim of a witness that the offender
possesses a master key. If the offender did possess a
master key to the apartments in the building, it
would explain why the means of entry was unknown
for other crimes in the series—the means of entry
would thus have been very difficult to determine for
earlier crimes where residents were not present.
Case study 3
Our method has also been used to help ensure the fidel-
ity of the historical database of past crimes. We ran our
method using data collected prior to 2006 to see
whether we would be able to make discoveries.
FIG. 9. The locations of crimes in the second series.
Table 5. The Third Serieslike Pattern with d=5
No. Date Location of entry Means of entry Premises Ransacked Residents Time of day Day Suspect Victim
1 2/9/05 Ground window Shoved/forced Apartment No In 1:47 Wed 1 White M 1 F
2 2/9/05 Rear door Shoved/forced Single-family House No In 9:50 Wed 1 Black M 1 M & 1 F
3 2/15/05 Ground window Broke Apartment Yes Not in 7:00–13:30 Tue Null 2 M
4 2/21/05 Front door Key Unknown No Not in 7:10–10:00 Mon Null 1 F
5 2/23/05 Front door Pried Apartment No Not in 7:10–16:00 Wed Null 2 M
6 2/23/05 Front door Pried Apartment No Not in 7:00–14:00 Wed Null 2 M
7 2/23/05 Front door Pried Apartment No Not in 7:45–17:25 Wed Null 2 F
8 2/28/05 Rear door Unknown Apartment No Not in 20:55 Mon 1 White M 1 F
Table 5 shows details from crimes within a 2005 pat-
tern, discovered by both the police and our algorithm.
Our algorithm and the crime analysts agreed on six out
of the eight crimes in the series, but disagreed on two
crimes: crime 3 and crime 4. The crime analysts iden-
tified these crimes as part of the pattern, but our algo-
rithms did not identify these crimes as being part of the
pattern. Our algorithm provides reasons why these
crimes should be excluded from the pattern: They are
not close to other crimes in the pattern-general sense
and are not connected to the other crimes in the series
within the similarity graph, as depicted in Figure 10.
Neither of the crimes are contained in any cores, as
shown in Table 6. In particular, the map in Figure 11
shows that these two crimes are geographically far
away from the other crimes. Since geographic closeness
has a large contribution to pattern-general similarity,
crimes 3 and 4 are already not likely to be part of the
same series. Besides that, we also notice that other
aspects of crimes 3 and 4 differ from the rest of the
FIG. 10. Similarity graph for the third crime
FIG. 11. The locations of crimes in the third series.
pattern (see Table 5). In crime 4, the means of entry is
by key, which is different from that of other crimes
where entries were forced or pried. According to police
narratives, the fourth crime had a 91-year-old victim,
who reported that $35 was stolen from her purse by
someone who used a key to enter the apartment; this
indicates the crime was not part of the series, and fur-
ther indicates the possibility that this crime never actu-
ally occurred. In crime 3, the apartment was ransacked
while none of the other locations were. The narrative
for the third crime is very generic: someone broke in
through the living room window, stole jewelry and
loose change, and exited through the front door. It is
not clear that this crime possesses any distinguishing
characteristics to identify it as being part of the series.
When examining the crime series retrospectively,
the police agreed that crimes 3 and 4 likely should be
excluded from the pattern (note that this pattern has
not been officially verified through an arrest).
The problem of detecting crime series is both subtle and
difficult. In cases where crime series detection is trivial,
meaning that there is identifiable information about
the criminal (e.g., fingerprints or DNA) then sophisti-
cated modeling techniques are not needed—the solution
is direct in those cases. The more subtle cases where
crime series might otherwise become hidden in a sea
of data are where data mining methods can shine. Our
automated series detection methods are able to detect
crime patterns within big data sets where a human
could not. Detecting patterns among crime data is criti-
cal to effective policing. When a pattern is identified, po-
lice can implement effective strategies to prevent future
crimes, solve past crimes, identify and capture offenders,
and ensure offenders are investigated and prosecuted
fully for the crimes for which they are responsible. We
envision that methods such as ours could be used very
broadly for crime series detection, and would be partic-
ularly useful across regions or districts, where human
crime analysts would fail to be able to manually handle
vast amounts of data from several different regions or
data sources in order to detect patterns.
This work was supported by a grant from MIT Lincoln
Author Disclosure Statement
No competing financial interests exist.
1. Woodhams J, Bull R, Hollin CR. Chapter 6: Case linkage: identifying crimes
committed by the same offender. In: Criminal Profiling: International
Theory, Research, and Practice. New York: Humana Press Inc., 2007.
2. Gwinn SL, Bruce C, Cooper JP, Hick S. Exploring Crime Analysis. Readings
on Essential Skills, 2nd ed. Charleston, SC: BookSurge, LLC, 2008.
3. Hering AS, Bair S. Characterizing spatial and chronological target selec-
tion of serial offenders. J R Stat Soc Ser C. 2014;63:123–140.
4. Wolfgang ME. Delinquency in a Birth Cohort. Chicago: University of Chi-
cago Press, 1987.
5. Weisel DL. Burglary of single-family houses. Problem-Oriented Guides for
Police Series, No. 18, 2002.
6. Computing Community Consortium, Computing Research Association.
2012. Challenges and opportunities with big data, a community white
paper developed by leading researchers across the United States.
7. Rudin C, et al. (A working group of the American Statistical Association).
2014. Discovery with data: leveraging statistics with computer science
to transform science and society.
8. Porter MD. A statistical approach to crime linkage. ArXiv e-prints, 2014.
9. Cocx TK, Kosters WA. A distance measure for determining similarity be-
tween criminal investigations. In: Adderley R, Musgrove P, (eds.), Advan-
ces in Data Mining. Applications in Medicine, Web Mining, Marketing,
Image and Signal Mining, Volume 4065 of Lecture Notes in Computer
Science. Berlin: Springer, 2006, pp. 511–525.
10. Lin S, Brown DE. An outlier-based data association method for linking
criminal incidents. Decis Supp Syst 2006;41:604–615.
11. Nath SV. Crime pattern detection using data mining. In: Proceedings of
the Web Intelligence and Intelligent Agent Technology Workshops; 2006,
pp. 41–44.
12. Singla CR, Dembla D, Chaba Y, Singla K. 2008. An optimal kd model for
crime pattern detection based on semantic link analysis-a data mining
Table 6. Cores and Their Defining Features for Example 4
Cores Crimes
of entry
of entry Premises Ransacked Residents
of day Day Suspect Victim
_ Model_for_Crime_Pattern_Detection_Based_on_Semantic_Link
_Analysis-A_ Data_Mining_Tool.
13. Bennell C, Bloomfield S, Snook B, et al. Linkage analysis in cases of serial
burglary: comparing the performance of university students, police pro-
fessionals, and a logistic regression model. Psychol Crime Law.
14. Bennell C, Canter DV. Linking commercial burglaries by modus
operandi: tests using regression and ROC analysis. Sci Justice. 2002;42:
15. Bennell C, Jones N. Between a ROC and a hard place: a method for linking
serial burglaries by modus operandi. J Invest Psychol Offender Profiling.
16. Bennell C, Jones N, Melnyk T. Addressing problems with traditional crime
linking methods using receiver operating characteristic analysis. Legal
Criminol Psychol. 2009;14:293–310.
17. Bennell C, Snook B, Macdonald S, et al. Computerized crime linkage
systems a critical review and research agenda. Criminal Justice Behav
18. Brown DE, Hagen S. Data association methods with applications to law
enforcement. Decis Support Syst. 2003;34:369–378.
19. Ellingwood H, Mugford R, Bennell C, et al. Examining the role of similarity
coefficients and the value of behavioural themes in attempts to link serial
arson offences. J Invest Psychol Offender Profiling. 2012;10:1–27.
20. Markson L, Woodhams J, Bond JW. Linking serial residential burglary:
comparing the utility of modus operandi behaviours, geographical
proximity, and temporal proximity. J Invest Psychol Offender Profiling
21. Tonkin M, Santtila P, Bull R. The linking of burglary crimes using offender
behaviour: testing research cross-nationally and exploring methodology.
Legal Criminol Psychol. 2012;17:276–293.
22. Tonkin M, Grant T, Bond JW. To link or not to link: a test of the case linkage
principles using serial car theft data. J Invest Psychol Offender Profiling.
23. Tonkin M, Woodhams J, Bull R, et al. Linking different types of crime using
geographical and temporal proximity. Criminal Justice Behav.
24. Wang T, Rudin C, Wagner D, Sevieri R. Learning to detect patterns in
crime. In: Machine Learning and Knowledge Discovery in Databases.
Springer, 2013, pp. 515–530.
25. Wang T, Rudin C, Wagner D, Sevieri R. Detecting patterns of crime with
series finder. In: Proceedings of AAAI Late Breaking Track; 2013.
26. Huggins JH, Rudin C. A statistical learning theory framework for super-
vised pattern discovery. In: Proceedings of SIAM Conference on Data
Mining (SDM); 2014.
27. Adderley R. The use of data mining techniques in operational crime
fighting. In: Intelligence and Security Informatics, Volume 3073 of Lecture
Notes in Computer Science. Chen H, Moore R, Zeng DD, Leavitt J (eds.).
Berlin: Springer, 2004, pp. 418–425.
28. Adderley A, Musgrove P. Modus operandi modelling of group offending:
a data-mining case study. Int J Police Sci Manage. 2003;5:265–276.
29. Dahbur K, Muscarello T. Classification system for serial criminal patterns.
Artif Intell Law. 2003;11:251–269.
30. Hering AS, Kazor K. A permutation test to identify important attributes for
linking crimes of serial offenders. Stat. 2013;2:211–226.
31. Ma L, Chen Y, Huang H. Ak-modes: a weighted clustering algorithm for
finding similar case subsets. In: Proceedings of the International Confer-
ence on ISKE, 2010; 2010, pp. 218–223.
32. Reich BJ, Porter MD. Partially supervised spatiotemporal clustering for
burglary crime series identification. J R Stat Soc Ser A. 2014;178:465–480.
33. Santtila P, Fritzon K, Tamelander AL. Linking arson incidents on the basis
of crime scene behavior. J Police Criminal Psychol. 2004;19:1–16.
34. Santtila P, Junkkila J, Kenneth Sandnabba N. Behavioural linking of
stranger rapes. J Invest Psychol Offender Profiling 2005;2:87–103.
35. Sorochinski M, Gabrielle Salfati C. The consistency of inconsistency in
serial homicide: patterns of behavioural change across series. J Invest
Psychol Offender Profiling. 2010;7:109–136.
36. Kriegel H-P, Kro
¨ger P, Zimek A. Clustering high-dimensional data: a survey
on subspace clustering, pattern-based clustering, and correlation clus-
tering. ACM Trans Knowledge Discov Data. 2009;3:1–58.
37. Pei J, Zhang X, Cho M, et al. Maple: a fast algorithm for maximal pattern-
based clustering. In: Proceedings of the International Conference on Data
Mining (ICDM). New York: IEEE, 2003, pp. 259–266.
38. Wang H, Wang W, Yang J, Yu PS. Clustering by pattern similarity in large
data sets. In: Proceedings of ACM SIGMOD; 2002, pp. 394–405.
39. Domeniconi C, Papadopoulos D, Gunopulos D, Ma S. Subspace clustering
of high dimensional data. In: Proceedings of the SIAM Conference on
Data Mining (SDM); 2004, pp. 517–521.
40. Vidal R. Subspace clustering. IEEE Signal Process Mag. 2011;28:52–68.
41. Gu¨ nnemann S, Boden B, Seidl T. DB-CSC: a density-based approach for
subspace clustering in graphs with feature vectors. In: Proceedings of the
European Conference on Machine Learning and Principles and Practice of
Knowledge Discovery in Databases (ECML-PKDD). New York: Springer,
2011, pp. 565–580.
42. Moser F, Colak R, Rafiey A, Ester M. Mining cohesive patterns from graphs
with feature vectors. In: Proceedings of the SIAM Conference on Data
Mining (SDM); 2009, volume 9, pp. 593–604.
43. McFowland E, Speakman S, Neill DB. Fast generalized subset scan
for anomalous pattern detection. J Mach Learn Res. 2013;14:
44. Neill DB, Cooper GF. A multivariate bayesian scan statistic for early event
detection and characterization. Mach Learn. 2010;79:261–282.
45. Guan Y, Dy JG, Jordan MI. A unified probabilistic model for global
and local unsupervised feature selection. In: Proceedings of the 28th
International Conference on Machine Learning (ICML); 2011, pp. 1073–
46. Chen H, Chung W, Xu JJ, et al. Crime data mining: a general framework
and some examples. IEEE Comput. 2004;37:50–56.
47. Thongtae P, Srisuk S. An analysis of data mining applications in crime
domain. In: Proceedings of the IEEE 8th International Conference on
Computer and Information Technology Workshops; 2008, pp. 122–126.
48. Berk RA. Algorithmic criminology. Secur Inform. 2013;2:5.
49. Berk R. Criminal Justice Forecasts of Risk: A Machine Learning Approach.
New York: Springer, 2012.
50. Berk R, Sherman L, Barnes G, et al. Forecasting murder within a popula-
tion of probationers and parolees: a high stakes application of statistical
learning. J R Stat Soc Ser A 2009;172:191–211.
51. Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning:
Data Mining, Inference and Prediction. New York: Springer, 2009.
52. Ghahramani Z, Heller K. Bayesian sets. In: Proceedings of Neural Infor-
mation Processing Systems (NIPS); 2005.
53. Letham B, Rudin C, Heller KA. Growing a list. Data Mining Knowledge
Disco. 2013;27:372–395.
Cite this article as: Wang T, Rudin C, Wagner D, Sevieri R (2015)
Finding patterns with a rotten core: data mining for crime series with
cores. Big Data 3:1, 3–21, DOI: 10.1089/big.2014.0021.
Abbreviations Used
M.O. ¼modus operandi
HAC ¼hierarchical agglomerative clustering
NN ¼nearest neighbor
SL ¼single linkage
CL ¼complete linkage
GA ¼group average
ILP ¼integer linear programming
... When comparing the similarity of criminal behavior, every two cases form a case pair, and then crime linkage becomes the task of judging whether a case pair is serial or nonserial [30] . Many methods have been applied to detect crime series. ...
Crime linkage is a difficult task and is of great significance to maintaining social security. It can be treated as a binary classification problem. Some crimes are difficult to determine whether they are serial crimes under the existing evidence, so the two-way decisions are easy to make mistakes for some case pairs. Here, the three-way decisions based on the decision-theoretic rough set are applied and its key issue is to determine thresholds by setting appropriate loss functions. However, sometimes the loss functions are difficult to obtain. In this paper, a method to automatically learn thresholds of the three-way decisions without the need to preset explicit loss functions is proposed. We simplify the loss function matrix according to the characteristic of crime linkage, re-express thresholds by loss functions, and investigate the relationship between overall decision cost and the size of the boundary region. The trade-off between the uncertainty of the boundary region and the decision cost is taken as the optimization objective. We apply multiple traditional classification algorithms as base classifiers, and employ real-world cases and some public datasets to evaluate the effect of our proposed method. The results show that the proposed method can reduce classification errors.
... Serial crime detection is generally based on the similarity of criminal behaviour [5]. Research on the use of behavioural similarity to detect serial crimes concerns various types of crimes, including serious crimes, such as rape [6,7] and murder [8], and large-scale crimes, such as theft [9,10], robbery [11][12][13] and fraud [14]. The modus operandi (M.O.) concerns the behaviour of criminals when they commit a crime, escape, and protect themselves from arrest [15]. ...
Full-text available
Detecting serial crimes is to find criminals who have committed multiple crimes. A classification technique is often used to process serial crime detection, but the pairwise comparison of crimes is of quadratic complexity, and the number of nonserial case pairs far exceeds the number of serial case pairs. The blocking method can play a role in reducing pairwise calculation and eliminating nonserial case pairs. But the limitation of previous studies is that most of them use a single criterion to select blocks, which is difficult to guarantee an excellent blocking result. Some studies integrate multiple criteria into one comprehensive index. However, the performance is easily affected by the weighting method. In this paper, we propose a combined blocking (CB) approach. Each criminal behaviour is defined as a behaviour key (BHK) and used to form a block. CB learns several weak blocking schemes by different blocking criteria and then combines them to form the final blocking scheme. The final blocking scheme consists of several BHKs. Because rare behaviour can better identify crime series, each BHK is assigned a score according to its rarity. BHKs and their scores are used to determine whether a case pair need to be compared. After comparing with multiple blocking methods, CB can effectively guarantee the number of serial case pairs while greatly reducing unnecessary nonserial case pairs. The CB is embedded in a supervised machine learning framework. Experiments on real-world robbery cases demonstrate that it can effectively reduce pairwise comparison, alleviate the class imbalance problem and improve detection performance.
... Furthermore, predictive policing, which utilizes historic crime data to inform current and future decision making on crime prevention, provides an example of how data analytics and machine learning can be used in crime prevention [28][29][30]. The ability to develop new techniques to create interventions provides insight into tools law enforcement can make decisions on in the future [31][32][33]. These methods and concepts may be the result of researcher-practitioner partnerships, which are a foundation of translational criminology. ...
Full-text available
Scientists should not play a role in investigations nor should investigators play a role in the scientific analyses. One way to bridge the relationship between the forensic scientist and the police investigator is through an Intelligence Analyst (IA) who is part of the forensic services operation. The IA offers the ability to walk between the role of scientist and law enforcement, receiving data after completion of scientific analyses and translating the information into actionable intelligence. The additional bridging and translating services represent a paradigm shift with increased emphasis on investigative contributions from forensic analysis. Forensic intelligence incorporates forensic data early in an investigation in a holistic case approach that incorporates possible datasets and information that could be relevant to the investigation. We present a brief review of the value added when an IA provides the bridge between the forensic laboratory and police investigators to enhance the use of forensic evidence.
... With the accumulation of Big data in various real-world domains, the usefulness of core periphery structures has increased even more. In a recent paper [19], the authors perform an analysis of cores in a series of big crime data so as to characterize the modus operandi of criminals across the crime series. ...
Core periphery structures exist naturally in many complex networks in the real-world like social, economic, biological and metabolic networks. Most of the existing research efforts focus on the identification of a meso scale structure called community structure. Core periphery structures are another equally important meso scale property in a graph that can help to gain deeper insights about the relationships between different nodes. In this paper, we provide a definition of core periphery structures suitable for weighted graphs. We further score and categorize these relationships into different types based upon the density difference between the core and periphery nodes. Next, we propose an algorithm called CP-MKNN (Core Periphery-Mutual K Nearest Neighbors) to extract core periphery structures from weighted graphs using a heuristic node affinity measure called Mutual K-nearest neighbors (MKNN). Using synthetic and real-world social and biological networks, we illustrate the effectiveness of developed core periphery structures.
... Clasificar los llamados atendiendo a este campo, rico en información, dará acceso a un conjunto de palabras claves, indicadores y datos que complementarán los informes periódicos, esta "nueva información" (que no es nueva porque siempre estuvo presente pero no en un formato accesible) abrirá la posibilidad de manejar informes detallados y específicos, atendiendo no solo a la causa u origen del llamado origen, sino también a las relaciones entre los eventos/sucesos. Este tipo de relaciones tiene que ver con los patrones que se logren identificar (Wang, Rudin, Wagner, & Sevieri, 2015). ...
Conference Paper
Full-text available
La minería de texto o text mining la englobamos dentro de las técnicas y modelos de minería de datos o data mining, que es el análisis matemático para deducir patrones y tendencias que existen en los datos, patrones que no pueden detectarse mediante una exploración tradicional de los datos porque las relaciones son demasiado complejas o por el volumen de datos que se maneja. A diferencia del data mining, el text mining trabaja sobre datos no estructurados en grandes colecciones de datos. La base de datos del Servicio del 911 de la provincia de Jujuy presenta un panorama muy interesante, ya que si bien la información que maneja dicho servicio es almacenada en una base de datos estructurada, la tabla principal que almacena los incidentes que son cargados por los tele-operadores posee un campo llamado descripción donde allí se vuelca la información más importante y los detalles del llamado de emergencia, información que no puede almacenarse en los campos del formulario de carga. Este campo ofrece una alternativa muy interesante de exploración pues el contenido no estructurado de dicho campo no es analizado ni tenido en cuenta por los informes tradicionales que utilizan en la actualidad.
We consider the sequential anomaly detection problem in the one-class setting when only the anomalous sequences are available and propose an adversarial sequential detector by solving a minimax problem to find an optimal detector against the worst-case sequences from a generator. The generator captures the dependence in sequential events using the marked point process model. The detector sequentially evaluates the likelihood of a test sequence and compares it with a time-varying threshold, also learned from data through the minimax problem. We demonstrate our proposed method’s good performance using numerical experiments on simulations and proprietary large-scale credit card fraud data sets. The proposed method can generally apply to detecting anomalous sequences. Funding: This work is partially supported by the National Science Foundation [Grants CAREER CCF-1650913, DMS-1938106, and DMS-1830210] and grant support from Macy’s Technology. Supplemental Material: The online appendix is available at .
Crime prediction plays a vital role in public security. Existing studies infer crime locations or crime groups without considering individual GPS trajectory data. They ignore joint influence on crime patterns coming from the internal relationship between criminals, locations, and time. In this study, we propose Fusion Information Graph Attention Networks (FIGAT), which classifies individuals into high and low risks with personal movement time series and location trajectories. To solve the independence of individual crime behavior and the fusion information loss problem, FIGAT proposes Multi-dimension Fusion Information Graph to combine semantic correlation features with conventional person basic features, time features, and location features. FIGAT constructs a multi-relation graph attention layer, which utilizes the semantic relationship and node information to accurately classify individuals into high and low risks. We evaluate FIGAT with 14,625,884 GPS trajectories from 1038 individuals collected by a real-world public safety department. The results demonstrate that FIGAT improves F1 score by 41%, 32%, and 23% compared with legacy machine learning, RNN-based deep learning, and graph neural network SOTA methods, respectively. T-SNE results and ablation experiments further prove the effectiveness of FIGAT.
In the current scenario as crime is increasing day by day it is becoming a challenge for the investigating agencies to have control on this. Again criminals are adopting new techniques and formulae to perform crime so that it would be difficult for law enforcement department to catch them. Hence crime analysis, detection and prediction becomes essential now‐a‐days. For this we need to go through previous crime records, analyze them, then classification/regression algorithms can be applied basing on requirement in order to predict the next occurrence of crime. So that next crime can be checked before being committed. As a procedure learned rules from training set are applied on test set in order to obtain predicted output. Several other data mining techniques can be a great help towards solving the problem. By going through this literature one can have a brief idea regarding what has been done in this regard and can acquire knowledge in facing the problem in a different way. This paper also puts light on several criminal analysis methods in use till date.
Full-text available
Many crimes can happen every day in a major city, and figuring out which ones are committed by the same individual or group is an important and difficult data mining challenge. To do this, we propose a pattern detection algorithm called Series Finder, that grows a pattern of discovered crimes from within a database, starting from a "seed" of a few crimes. Series Finder incorporates both the common characteristics of all patterns and the unique aspects of each specific pattern. We compared Series Finder with classic clustering and classification models applied to crime analysis. It has promising results on a decade's worth of crime pattern data from the Cambridge Police Department. Copyright © 2013, Association for the Advancement of Artificial Intelligence. All rights reserved.
Full-text available
This chapter begins by explaining the purposes of linking crimes committed by the same offender and what case linkage can add to a police investigation and prosecution. The various steps involved in the process of case linkage are explained. The assumptions of behavioral consistency and inter-individual behavioral variation, which case linkage rests on, are outlined, and the research that has begun to test these assumptions is reported. The effect of poor-quality data on the case linkage process and on empirical research is examined. Current methods and future developments for overcoming this difficulty are described. The obstacles to identifying linked crimes across police boundaries are discussed. Case linkage research and practice are compared with various criteria for expert evidence with promising results. The chapter closes by considering future avenues for research and practice in case linkage.
Conference Paper
Full-text available
Our goal is to automatically detect patterns of crime. Among a large set of crimes that happen every year in a major city, it is challenging, time-consuming, and labor-intensive for crime analysts to determine which ones may have been committed by the same individual(s). If automated, data-driven tools for crime pattern detection are made available to assist analysts, these tools could help police to better understand patterns of crime, leading to more precise attribution of past crimes, and the apprehension of suspects. To do this, we propose a pattern detection algorithm called Series Finder, that grows a pattern of discovered crimes from within a database, starting from a \seed" of a few crimes. Series Finder incorporates both the common characteristics of all patterns and the unique aspects of each speci c pattern, and has had promising results on a decade's worth of crime pattern data collected by the Crime Analysis Unit of the Cambridge Police Department.
Full-text available
The object of this paper is to develop a statistical approach to criminal linkage analysis that discovers and groups crime events that share a common offender and prioritizes suspects for further investigation. Bayes factors are used to describe the strength of evidence that two crimes are linked. Using concepts from agglomerative hierarchical clustering, the Bayes factors for crime pairs are combined to provide similarity measures for comparing two crime series. This facilitates crime series clustering, crime series identification, and suspect prioritization. The ability of our models to make correct linkages and predictions is demonstrated under a variety of real-world scenarios with a large number of solved and unsolved breaking and entering crimes. For example, a naive Bayes model for pairwise case linkage can identify 82% of actual linkages with a 5% false positive rate. For crime series identification, 74%-89% of the additional crimes in a crime series can be identified from a ranked list of 50 incidents.
Statistical clustering of criminal events can be used by crime analysts to create lists of potential suspects for an unsolved crime, to identify groups of crimes that may have been committed by the same individuals or group of individuals, for offender profiling and for predicting future events. We propose a Bayesian model-based clustering approach for criminal events. Our approach is semisupervised, because the offender is known for a subset of the events, and utilizes spatiotemporal crime locations as well as crime features describing the offender's modus operandi. The hierarchical model naturally handles complex features that are often seen in crime data, including missing data, interval-censored event times and a mix of discrete and continuous variables. In addition, our Bayesian model produces posterior clustering probabilities which allow analysts to act on model output only as warranted. We illustrate the approach by using a large data set of burglaries in 2009–2010 in Baltimore County, Maryland.
We propose Fast Generalized Subset Scan (FGSS), a new method for detecting anomalous patterns in general categorical data sets. We frame the pattern detection problem as a search over subsets of data records and attributes, maximizing a nonparametric scan statistic over all such subsets. We prove that the nonparametric scan statistics possess a novel property that allows for efficient optimization over the exponentially many subsets of the data without an exhaustive search, enabling FGSS to scale to massive and high-dimensional data sets. We evaluate the performance of FGSS in three real-world application domains (customs monitoring, disease surveillance, and network intrusion detection), and demonstrate that FGSS can successfully detect and characterize relevant patterns in each domain. As compared to three other recently proposed detection algorithms, FGSS substantially decreased run time and improved detection power for massive multivariate data sets.
Using 31 unique crime series from two US cities, the spatial pattern that individual criminal decision makers follow is investigated. Locations within a city vary in the likelihood for crimes of a given type to occur, and this is accounted for by using kernel density estimation based on all crimes of each type. Kernel density estimation is highly influenced by the bandwidth, so an objective approach is used to select the estimate. Then, the kernel density estimate for each type of crime is incorporated in an inhomogeneous K-function that can identify significant spatial clustering and/or uniformity at various spatial scales for each crime series. We find that robbery series are more likely to exhibit uniformity than burglary series, which tend to show strong clustering. In addition, the order in which new crimes are added to a series is found to follow an interesting pattern that does not always support the theory of offender as forager in which criminals first cluster crimes and then gradually disperse them. Half of the robbery series exhibit a spatial distribution that begins dispersed and develops clusters as increasingly more events are added to the series, indicating that they often return to locations of earlier crimes after having committed crimes elsewhere.
The modus operandi (MO) of a crime describes the unique characteristics that an offender imparts to it. Although in some instances, a serial offender's behavior is circumstantial, some MO behaviors may be consistent from one crime to the next. By investigating these behaviors, similar crimes can be linked to the same individual, but some attributes describing a crime may be more important for linking than others. Two strategies have historically been used to link crimes. One relies on expert criminal judgment, which may require manually sifting through thousands of crime records. In the second, similar attributes are grouped, and logistic regression is applied to coarse frequency summaries of those groupings. In this work, we introduce an intuitive statistical permutation test for assessing the importance of individual attributes in linking crimes, and we show how the results can be used to weight each one's importance to link crimes. By using the serial offenses of four sets of residential burglaries and six sets of robberies identified in Tempe, Arizona, the test is illustrated, and differences among attribute importance of these two types of crimes are highlighted. We demonstrate greater success in linking crimes when the test results are incorporated into a linking analysis. Copyright © 2013 John Wiley & Sons Ltd