Content uploaded by Henry Martin
Author content
All content in this area was uploaded by Henry Martin on Jun 18, 2018
Content may be subject to copyright.
A preview of the PDF is not available
Identifying and Interpreting Clusters of Persons with Similar Mobility Behaviour Change Processes
Abstract and Figures
With the emergence of new mobility options and various initiatives to increase the sustainability of our travel behaviour, it is desirable to gain a deeper understanding of our behavioural reactions to such stimuli. Although it is now possible to use GPS-tracking to record people’s movement behaviour over a longer period, there is still a lack of computational methods which allow to detect and evaluate such behaviour change processes in the resulting datasets. In this study, we propose a data mining method for describing individual persons’ mobility behaviour change processes based on their movement trajectories and clustering participants based on the similarity of these behavioural adaptations. We further propose to use a decision tree classifier to semantically explain the derived clusters in a human-interpretable form. We apply our method to a real, longitudinal movement dataset.
Figures - uploaded by Henry Martin
Author content
All figure content in this area was uploaded by Henry Martin
Content may be subject to copyright.
... This challenge will require knowledge about the mobility behavior of people and the ability to predict it in the near future in order to optimally allocate shared mobility resources. With the recent success of machine learning algorithms [LeCun et al., 2015], research in computational movement analysis [Long et al., 2018] shifted towards using machine learning methods to support data interpretation (e.g., labeling [Toch et al., 2019], clustering tasks , Jonietz et al., 2018) or prediction tasks [Luca, Barlacchi, Lepri, & Pappalardo, 2023;Kumar and Raubal, 2021;Kreil et al., 2020]. ...
... A particularly difficult problem in this situation is the identification of stable groups of users with similar mobility behavior, which are comparable across datasets. Finding such mobility types can help us to enhance our understanding of mobility behavior [Pappalardo et al., 2015], to measure regional similarities between cities McKenzie and Romm [2021] or neighborhoods Calafiore et al. [2021], and to detect changes in mobility behavior over time [Jonietz et al., 2018;Hong et al., 2021]. However, existing solutions are often based on dataset-specific features and can either not be applied to different datasets, or results from different datasets are not comparable. ...
The decarbonization of the transport system requires a better understanding of human mobility behavior to optimally plan and evaluate sustainable transport options (such as Mobility as a Service). Current analysis frameworks often rely on specific datasets or data-specific assumptions and hence are difficult to generalize to other datasets or studies. In this work, we present a workflow to identify groups of users with similar mobility behavior that appear across several datasets. Our method does not depend on a specific clustering algorithm, is robust against the choice of hyperparameters, does not require specific labels in the dataset, and is not limited to specific types of tracking data. This allows the extraction of stable mobility profiles based on several small and inhomogeneous tracking data sets. Our method consists of the following main steps: Representing individual mobility using location-based graphs; extraction of graph-based mobility features; clustering using different hyperparameter configurations; group identification using statistical testing. The method is applied to six tracking datasets (Geolife, Green Class 1 + 2, yumuv and two Foursquare datasets) with a total of 1070 users that visit about 3′000’000 different locations with a total tracking duration of over 200′000 days. We can identify and interpret five mobility profiles that appear in all datasets and show how these profiles can be used to analyze longitudinal and cross-sectional tracking studies.
... The proportion of each cluster size, as well as the number of clusters, may well be available as a priori knowledge for clustering. Clustering is based on a similarity measure to group semblable data objects together, so it is commonly utilized in areas such as market segmentation, vehicle routing selection, and healthcare problems [23,24]. On these occasions, users employ certain conditions for various classification purposes. ...
... The idea behind this heuristic was to induce decision rules with which to identify the variables and categorical values that would allow us to assign individuals to clusters. Inducing decision rules has the additional advantage of generating human-readable information on clusters, as has recently been done in a variety of applications (see [51,52]). ...
A methodological approach is developed for exploring the relationship between the use of social networking sites and participation in protest activities. Although a recent meta-analysis study demonstrated that there is a positive association between the two, little work examining this association further appears to have been published. The methodology proposed here studies the patterns of the relationship between nine social media and five types of protest activity using the techniques of multiple correspondence analysis, hierarchical cluster analysis and induction of decision rules. The results give insights into the relationship in different segments of individuals' profiles defined as non-activist, offline activist, social media user (two types) and online activist. Significantly, this last segment proves to be a small and heterogeneous group. The results also show that the proposed approach is useful for exploring the patterns of the relationship in a low-dimensional space. Limitations of the methodology and possible extensions are discussed.
Travel mode choice analysis is a central aspect of understanding human mobility and plays an important role in urban transportation and planning. The emergence of passively recorded movement data with spatio-temporal and semantic information offers opportunities for uncovering individuals’ travel mode choice behavior. Considering that many of these choices are highly regular and are performed in similar manners by different groups of people, it is desirable to identify these groups and their characteristic behavior (e.g. for educational or political incentives or to find environmentally-friendly people). Previous research mainly grouped people according to “mobility snapshots”, i.e. mobility patterns exhibited at a single point in time. We argue that especially when considering the change of behavior over time, we need to investigate the behavioral dynamic processes resp. the change of travel mode choices over time. We present a framework that can be used to cluster people according to the dynamics of their travel mode choice behavior, based on automatically tracked GPS data. We test the framework on a large user sample of 107 persons in Switzerland and interpret their travel mode choice behavior patterns based on the clustering results. This facilitates understanding people’s travel mode choice behavior in multimodal transportation and how to design reasonable alternatives to private cars for more sustainable cities.
With the emergence of ubiquitous movement tracking technologies, developing systems which continuously monitor or even influence the mobility behaviour of individuals in order to increase its sustainability is now possible. Currently, however, most approaches do not move beyond merely describing the status quo of the observed mobility behaviour, and require an expert to assess possible behaviour changes of individual persons. Especially today, automated methods for this assessment are needed, which is why we propose a framework for detecting behavioural anomalies of individual users by continuously mining their movement trajectory data streams. For this, a workflow is presented which integrates data preprocessing, completeness assessment, feature extraction and pattern mining, and anomaly detection. In order to demonstrate its functionality and practical value, we apply our system to a real-world, large-scale trajectory dataset collected from 139 users over 3 months.
Information and communication technologies (ICT) have considerable importance for transport systems, as they provide access to travel information, planning tools, opportunities to share transport modes, to work at-a-distance, compare transport mode cost, make payment, improve safety and health, and to communicate travel patterns. Over the past decade, there has been massive growth in the availability of transportation ICT, in particular smartphone applications. There is considerable evidence that ICT has profoundly changed the ways in which transport systems are perceived and used, and mobilities performed; with far-reaching implications for transport mode choices and transport demand. Against this background, the paper seeks to conceptualize ICT with relevance for transport systems, and to discuss the implications for the environmental sustainability of the transport sector. Findings suggest that while some ICT innovations foster and support sustainable transport choices, others raise new and significant barriers to more sustainable transport futures.
The large interest in analyzing one's own fitness led to the development of more and more powerful smartphone applications. Most are capable of tracking a user's position and mode of locomotion, data that do not only reflect personal health, but also mobility choices. A large field of research is concerned with mobility analysis and planning for a variety of reasons, including sustainable transport. Collecting data on mobility behavior using fitness tracker apps is a tempting choice, because they include many of the desired functions, most people own a smartphone and installing a fitness tracker is quick and convenient. However, as their original focus is on measuring fitness behavior, there are a number of difficulties in their usage for mobility tracking. In this paper we denote the various challenges we faced when deploying GoEco! Tracker (an app using the Moves R fitness tracker to collect mobility measurements), and provide an analysis on how to best overcome them. Finally, we summarize findings after one month of large scale testing with a few hundred users within the GoEco! living lab performed in Switzerland.
This paper explores how we can use smart card data for bus passengers to reveal individual and aggregate travel behaviour. More specifically, we measure the extent to which both individual and bus routes exhibit habitual behaviour. To achieve this, we introduce a metric called Stickiness Index to quantify the range of preferences of users that always select to travel on the same route (high stickiness) to those with a more varied patterns of route selection (low stickiness). Adopting a visual analytic and modelling approach using a suite of regression models we find evidence to suggest that stickiness varies across the metropolitan area and over a 24-h period wherein higher stickiness is associated with high frequency users where there is substantial variability of route travel times across all alternatives. We argue that our findings are important in their capacity to contribute to a new evidence base with the potential to inform the (re)-design and scheduling of a public transit systems through unveiling the complexities of transit behaviour.
For the improvement of reliability, safety and efficiency advanced methods of supervision, fault detection and fault diagnosis become increasingly important for many technical processes. This holds especially for safety related processes like aircraft, trains, automobiles, power plants and chemical plants. The classical approaches are limit or trend checking of some measurable output variables. Because they do not give a deeper insight and usually do not allow a fault diagnosis, model-based methods of fault detection were developed by using input. and output signals and applying dynamic process models. These methods are based, e.g., on parameter estimation, parity equations or state observers. Also signal model approaches were developed. The goal is to generate several symptoms indicating the difference between nominal and faulty status. Based on different symptoms fault diagnosis procedures follow, determining the fault by applying classification or inference methods. This contribution gives a short introduction into the field and shows some applications for an actuator, a passenger car and a combustion engine.
Around the world, schooling and other forms of education shape the lives of children and their communities. During her own school reunion, the Noble Peace Prize Laureate from Burma, Daw Aung San Suu Kyi, addressed her former school friends from the Methodist English High School in Rangoon in a speech emphasizing the communal and intergenerational aspect of education: ‘Education is about to enable us to meet any challenge that life might throw at us, not just for ourselves but for those with whom we live. And those with whom we live are not just those in the family or town or country, but in this world today’ (Daw Aung San Suu Kyi, Methodist English High School, Burma Reunion 4th& 5th January 2013). As a matter of fact, the United Nations Millennium Development Goals are committed to improving access to primary education as an important means to promote gender equality and empower women. As a matter of fact, in 2015 the primary school net enrolment rate in poor countries has reached 91 % and the literacy rate among youth aged 15–24 years has increased globally to 91 %. The gap between educated women and men has also narrowed (United Nations 2015: 4–5). In spite of this progress, many children are still excluded from formal education systems and those who access school feel they are not learning the skills they need for their adult lives (Department for International Development 2013: 4). According to the Education for All Global Monitoring Report 2015 worldwide there were around 58 million children who were not attending school and around 100 million children who dropped out of primary school. Out-of-school children often come from poor families and frequently live in conflict zones. As a consequence, many young people lack the formal skills they need for the white-collar jobs they aspire to have in the future. Therefore, within a global context of neo-liberal markets and economic restructuring, it is increasingly necessary that the learning needs of children and young people as well as adults are met through equal access to learning programmes that encourage sociocultural skills (Banerjee and Duflo 2011: 74; Pells and Woodhead 2014: 42–44; UNESCO 2015a: 110–111). This is particularly relevant in settings where increased aspirations for schooling impact on intergenerational relations, as parents invest in the education of their children and expect them to become bearers of knowledge and skills (Wagner 2015: 14). This book develops the interplay between learning, migration and intergenerational relations through the experiences of Karen boys and girls at different historical times and different geographical locations. The Karen value education as a gift that cannot be immediately reciprocated. As one research participant put it: ‘I really believe in education. So, to give some education is a gift, priceless’ (Nant Bwa Bwa Phan, interview, 10 April 2015). Anthropological theory questions the gratuitousness of a gift (Hendry and Underdown 2012: 65). In Argonauts of the Western Pacific (1922), Malinowski described how the economic system of the Trobriand Islanders is based on a system of gift exchange. Gifts included armshells, curved tusks and fine long necklaces. The opening gift of the exchange always has to be reciprocated by a counter-gift. Yet, just so these gifts do not end up as barter, the gift exchange requires that gifts are of different nature and also the passage of time between giving and returning a gift is mandatory. If a person is for a longer while not able to return a gift equivalent in value to the gift given, he can meanwhile use intermediary gifts of inferior value to fill in the gap. As soon as a gift of equivalent value has been returned the transaction of giving and receiving is concluded by a clinching gift (Malinowski 1922: 352–358). According to Mauss (11925), the act of giving and receiving gifts tells about power relationships between two parties because ‘the bond established between donor and recipient is too strong for both of them (...) the recipient puts himself in a position of dependence vis-à-vis the donor’. Yet the power relations thus established differ from markets: ‘It is all a matter of etiquette, it is not like in the market where, objectively, and for a price, one takes something. Nothing is unimportant. Contracts, alliances, the passing of goods, the bonds created by these goods passing between those giving and receiving—this form of economic morality takes account of all this. The nature and intention of the contacting parties, the nature of the thing given, are all indivisible’ (Mauss 1990: 76–77). In this study I suggest considering education as a non-material gift, because similar to material gift exchanges, education always engages people in permanent commitments, therefore creating and sustaining relationships between educators and pupils of different ages. This intergenerational dimension is important to understanding processes of giving and receiving education. Like other gifts, education cannot in most instances be reciprocated immediately. Time is needed to return the intellectual and practical benefits of learning. While this is often a question of years, sometimes it may only take hours for children to bring new knowledge home and cause change in the economy of the household. Conversely, when young people miss the chance to return what they received, they remain indebted towards their elders. In this case, the value of education as a priceless gift is called into question. This is particularly relevant for humanitarian aid in refugee situations (Harrell-Bond 1999) and in modern economies where young people increasingly migrate for secondary and tertiary education, but are left without meaningful employment (Boyden and Crivello 2014; Kabeer 2000: 473). The book also relates to a conceptual framework that recognizes that education is about different learning processes and learning contexts ranging from formal and highly structured settings (for example schools with professional teachers) to informal highly structured processes (for example government sponsored or non-state forms of education, such as independent school projects for refugees), formal unstructured processes (for example when children learn informally from their peers at school) and informal settings where learning takes place informally (for example transmission of sociocultural learning at home). Recognizing how these areas operate in interdependence allows us to discern how children not only learn at school, but also learn at home from their parents and grandparents. Such a framework is also sensitive to the social dissonance during childhood.
Travel activities are embodied as people’s needs to be physically present at certain locations. The development of Information and Communication Technologies (ICTs, such as mobile phones) has introduced new data sources for modeling human activities. Based on the scattered spatiotemporal points provided in mobile phone datasets, it is feasible to study the patterns (e.g., the scale, shape, and regularity) of human activities. In this paper, we propose methods for analyzing the distribution of human activity space from both individual and urban perspectives based on mobile phone data. The Weibull distribution is utilized to model three predefined measurements of activity space (radius, shape index, and entropy). The correlation between demographic factors (age and gender) and the usage of urban space is also tested to reveal underlying patterns. The results of this research will enhance the understanding of human activities in different urban systems and demographic groups, as well as providing novel methods to expand the important and widely applicable area of geographic knowledge discovery in the age of instant access.
The public transport networks of dense cities such as London serve passengers with widely different travel patterns. In line with the diverse lives of urban dwellers, activities and journeys are combined within days and across days in diverse sequences. From personalized customer information, to improved travel demand models, understanding this type of heterogeneity among transit users is relevant to a number of applications core to public transport agencies’ function. In this study, passenger heterogeneity is investigated based on a longitudinal representation of each user’s multi-week activity sequence derived from smart card data. We propose a methodology leveraging this representation to identify clusters of users with similar activity sequence structure. The methodology is applied to a large sample (n = 33,026) from London’s public transport network, in which each passenger is represented by a continuous 4-week activity sequence. The application reveals 11 clusters, each characterized by a distinct sequence structure. Socio-demographic information available for a small sample of users (n = 1973) is combined to smart card transactions to analyze associations between the identified patterns and demographic attributes including passenger age, occupation, household composition and income, and vehicle ownership. The analysis reveals that significant connections exist between the demographic attributes of users and activity patterns identified exclusively from fare transactions.