B. Chandra

Blind image Steganalysis is gaining lot of importance these days owing to its application in cyberwarfare, computer forensics, tracking anti-social elements over internet. In this paper a modified discrete bird swarm algorithm (DBSA) referred as Dy_DBSA has been proposed for feature selection. The proposed algorithm not only reduces computational c...

This paper proposes a novel crime prediction algorithm DeepPoint- a hybrid of deep learning and point process. It aims to improve the prediction of high-volume street crimes. In the design of this learning pipeline, census attributes have also been factored in as features in addition to space-time features in the crime dataset. Furthermore, the con...

In this paper, we propose a novel algorithm to improve the state-of-the-art results of homicide prediction in Chicago crime dataset. A marked self-exciting point process (M-SEPP) based epidemic type aftershock sequence (ETAS) model was applied to the Chicago crime dataset by Mohler [1] for improving the prediction rate of homicides over and above t...

In this paper, a novel approach has been proposed for fault diagnosis of internal combustion (IC) engine using Empirical Mode Decomposition (EMD) and Neural Network. Live signals from the engines were collected with and without faults by using four sensors. The vibration signals measured from the large number of faulty engines were decomposed into...

DNA microarray allows to monitor the expression levels of thousands of genes under different experimental conditions. DNA microarray technology helps researchers learn about different kinds of diseases, especially cancer. Using this technology, researchers will be able to classify the types of cancer on the basis of gene expression levels in tumor...

The paper proposes a novel feature selection algorithm based on the density function of features. Density difference between features for paired classes is used for assigning weights to the features. Existing methods select only those features that can distinguish between all classes at the same time. A new filter based feature selection method ter...

An attempt has been made to improve the performance of Deep Learning with Multilayer Perceptron (MLP). Tuning the learning rate or finding an optimum learning rate in MLP is a major challenge. Depending on the value of the learning rate, classification accuracy can vary drastically. This issue has been taken as a challenge in this paper. In this pa...

The paper aims at speeding up Deep Neural Networks (DNN) since this is one of the major bottlenecks in deep learning. This has been achieved by parameterizing the weight matrix using low rank factorization and periodic functions. By parameterization, the weight matrix is split into two matrices of smaller size of rank K with periodic functions. A s...

In recent past, there is an increased interest in multivariate time series (MTS) clustering research due to its wide applications in various areas such as finance, environmental research, multimedia and crime. The traditional similarity measures like correlation, Euclidean distance etc. cannot be applied to measure the similarity among data objects...

An attempt has been made to use efficient Neuron model for blind source separation. Generalized Harmonic Mean Neuron (GHMN) has been used as the neuron model. GHMN model is based on generalized harmonic mean of the inputs applied on it. Information-maximization approach has been used for training the neuron model. In this paper, it has been demonst...

An innovative approach has been proposed for using MLP for handling Big data. There is high computational cost and time involved in using MLP for classification of Big data having large number of features. A parameterized multilayer perceptron (PMLP) has been proposed where the weight matrix has been parameterized using periodic functions. This ens...

The paper proposes an Adaptive Stacked Denoising Autoencoder (ASDA) to overcome the limitations of Stacked Denoising Autoencoder (SDA) [6] in which noise level is kept fixed during the training phase of the autoencoder. In ASDA, annealing schedule is applied on noise where the average noise level of input neurons is kept high during initial trainin...

Purpose – The purpose of this paper is to introduce architecture of an Intelligent Decision Support System to fulfill the emerging responsibilities of law enforcement agencies. Design/methodology/approach – The proposed Intelligent Police System (IPS) is designed to meet the emerging requirements and provide information at all levels of decision m...

The paper discusses how Spiking Wavelet Radial Basis Neural Network can be effectively used for the classification of gene expression data. A new spiking function has been proposed in the non-linear integrate and fire model and its inter spike interval is derived and used in the Wavelet Radial Basis Neural Network for the classification of gene exp...

A biologically realistic spiking neuron model has been proposed which contains a novel non linear spiking function. Proposed neuron model contains a lower order spike generating function in contrast to the spike generating function of Quadratic integrate fire neuron model. It is found that lower order terms of spike generating function is sufficien...

A new octupolar molecule (E, E, E)-2,4,6-Tris [2-(4-N, N-diphenylaminophenyl) vinyl] pyridine (DPATSP) has been synthesized and studied by steady state and time-resolved fluorescence in condensed phase. The large π-conjugation along the chain has been found to be responsible for large solvent-induced shift of fluorescence. Pumping with fs pulses at...

The paper proposes a novel methodology of finding frequent itemsets in data stream. Fuzzification of support of the closed frequent itemsets in conjunction with a jumping window has been used for finding frequent itemsets. Closed frequent itemsets help in retaining all frequent itemsets in a reduced memory space. Fuzzifying the support of the close...

A biologically realistic non linear integrate and fire model is proposed in this paper. Its complete solution is derived and used for the construction of aggregation function in Multi layer perceptron model for classification of UCI Machine learning datasets. It is found that a single neuron in the conventional neural network is sufficient for the...

Semi-supervised clustering is gaining importance these days since neither supervised nor unsupervised learning methods in a stand-alone manner provide satisfactory results. Existing semi-supervised clustering techniques are mostly based on pair-wise constraints, which could be misleading. These semi-supervised clustering algorithms also fail to add...

A new octupolar molecule (E,E,E)-2,4,6-Tris [2-(4-N, N-diphenylaminophenyl) vinyl] pyridine (DPATSP) has been synthesized and studied by steady state and time-resolved fluorescence in condensed phase. The large p-conjugation along the chain has been found to be responsible for large solvent-induced shift of fluorescence. Pumping with fs pulses at 7...

This paper proposes a novel approach for classification of patterns having categorical attributes using decision trees. A new split measure has been proposed for construction of decision trees. Main focus of the proposed split measure is to improve the classification accuracy. Performance of the proposed split measure has been compared with the wel...

A new heterogeneous node split measure (HSM) has been proposed in this paper for decision tree construction. The split measure HSM is derived from quasilinear mean of information gain. This helps in including proportionalities of class values from the sub-partitions and the entire dataset at the same time. This results in acquiring more information...

The concept of finding frequent itemsets without pre-assigned weights is of great importance in Association Rule Mining (ARM). The prime advantage of this approach is that weights can be derived from the dataset itself rather than being given by domain expert. The modification of Apriori algorithm for Weighted Association Rule Mining (WARM) without...

Classification of gene expression data plays a significant role in prediction and diagnosis of diseases. Gene expression data has a special characteristic that there is a mismatch in gene dimension as opposed to sample dimension. All genes do not contribute for efficient classification of samples. A robust feature selection algorithm is required to...

The paper proposes an improved architecture for Probabilistic Neural Networks (IAPNN) with an aggregation function based on f-mean of training patterns. The improved architecture has reduced number of layers and that reduces the computational complexity. Performance of the proposed model was compared with the traditional Probabilistic Neural Networ...

The paper proposes a novel methodology of finding frequent itemsets in data stream. Fuzzification of support of closed frequent itemsets in conjunction with jumping window has been used for finding frequent itemsets. Fuzzification of support of closed frequent itemsets helps in preserving information regarding the frequent itemsets at different poi...

Prediction of option prices has always been a challenging task. Various models have been used in the past but there has been no effort to point out which model is suited best for predicting option prices. Computational time plays an important role in prediction of option prices since these time series are usually large. It is computationally expens...

Mining frequent substructures has gained importance in the recent past. Number of algorithms has been presented for mining undirected graphs. Focus of this paper is on mining frequent substructures in directed labeled graphs since it has variety of applications in the area of biology, web mining etc. A novel approach of using equivalence class prin...

Classical data mining algorithms require expensive passes over the entire database to generate frequent items and hence to generate association rules. With the increase in the size of database, it is becoming very difficult to handle large amount of data for computation. One of the solutions to this problem is to generate sample from the database t...

Naïve–Bayes Classifier (NBC) is widely used for classification in machine learning. It is considered as the first choice for many classification problems because of its simplicity and classification accuracy as compared to other supervised learning methods. However, for high dimensional data like gene expression data, it does not perform well due t...

The key element of neurocomputing research in complex domain is the development of artificial neuron model with improved computational power and generalization ability. The non-linear activities in neuronal interactions are observed in biological neurons. This paper presents architecture of a neuron with a non-linear aggregation function for comple...

The paper proposes a new neuron model (geometric mean neuron model) with an aggregation function based on geometric mean of all inputs. Performance of the geometric mean neuron model was evaluated using various learning algorithms like the back-propagation and resilient propagation on various real life data sets. Comparison of the performance of th...

The complex interaction of snow and meteorological features leads to formation of avalanches. Thus, the identification of significant governing features in the avalanche formation process is very critical aspect for developing any prediction scheme. This has direct implication on the design complexity of prediction scheme, computational effort invo...

A new node splitting measure termed as distinct class based splitting measure (DCSM) for decision tree induction giving importance to the number of distinct classes in a partition has been proposed in this paper. The measure is composed of the product of two terms. The first term deals with the number of distinct classes in each child partition. As...

The paper proposes a novel neuron model termed as Generalized Power Mean Neuron model (GPMN). The paper focuses on illustrating the computational power and the generalization capability of this model. In this model, the aggregation function is based on generalized power mean of the inputs. The performance of the neural network using GPMN model is c...

The paper presents a new graph based clustering algorithm. Traditional clustering algorithms have the drawback that it takes large number of iterations in order to come up with the desired number of clusters. The advantage of this approach is that the size of the dataset is reduced using graph based clustering approach and the required number of cl...

This paper proposes a complex valued generalized product neuron (GPN) which tries to incorporate polynomial structure in the aggregation of inputs. The advantage of using this model is to bring in the non-linearity in aggregation function by taking a product of linear terms in each dimension of the input space. This aggregation function has the abi...

Crisp decision tree algorithms face the problem of having sharp decision boundaries which may not be found in all real life classification problems. A fuzzy decision tree algorithm Gini Index based (G-FDT) is proposed in this paper to fuzzify the decision boundary without converting the numeric attributes into fuzzy linguistic terms. Gini Index is...

Motivated by the desire to construct compact (in terms of expected length to be traversed to reach a decision) decision trees, we propose a new node splitting measure for decision tree construction. We show that the proposed measure is convex and cumulative and utilize this in the construction of decision trees for classification. Results obtained...

The paper aims at training multilayer perceptron with different new error measures. Traditionally in MLP, Least Mean Square error (LMSE) based on Euclidean distance measure is used. However Euclidean distance measure is optimal distance metric for Gaussian distribution. Often in real life situations, data does not follow the Gaussian distribution....

In recent past, there is an increased interest in time series clustering research, particularly for finding useful similar trends in multivariate time series in various applied areas such as environmental research, finance, and crime. Clustering multivariate time series has potential for analyzing large volume of crime data at different time points...

The paper proposes new neuron architecture for Neural Network models with an aggregation function based on geometric mean of all inputs. This new neuron model gives better accuracy compared to Multilayer Perceptron model (MLP) without increasing the number of parameters. Various error measures have been used with this model. The effectiveness of th...

Traditional decision tree algorithms face the problem of having sharp decision boundaries which are hardly found in any real-life classification problems. A fuzzy supervised learning in Quest (SLIQ) decision tree (FS-DT) algorithm is proposed in this paper. It is aimed at constructing a fuzzy decision boundary instead of a crisp decision boundary....

Machine learning techniques like self organizing maps has been explored for the first time for efficient key clustering in cryptanalysis. In the first phase, transformations have been employed on the encrypted messages to find the dominance in the features and n, in the second phase, independent component analysis -ICA to find important features am...

The paper proposes new neuron model with an aggregation function based on Generalized harmonic mean of the inputs. Information-maximization approach has been used for training the new neuron model. The paper focuss on illustrating the efficiency of the proposed neuron model for blind source separation. It has been shown on various generated mixture...

Naive-Bayes classifier is a popular technique of classification in machine learning. Improving the accuracy of naive-Bayes classifier will be significant as it has great importance in classification using numerical attributes. For numeric attributes, the conditional probabilities are either modeled by some continuous probability distribution over t...

A new algorithm called STAG (Stacked Graph) for association rule mining has been proposed in this paper using graph theoretic approach. A structure is built by scanning the database only once or at most twice that can be queried for varying levels of minimum support to find frequent item sets. Incremental growth is possible as and when new transact...

In present day scenario, law enforcement agencies are looked upon not only to control crime but also to analyze the crime so that future occurrences of similar incidents can be overcome. There is need for user interactive interfaces based on current technologies to meet and fulfill the new emerging responsibilities and tasks of the Police. The pape...

Decision trees have been widely used for classification in Data mining. Number of decision tree algorithms has been developed in the past. The SLIQ algorithm [ 2 ] was developed with an aim to reduce diversity of the decision tree at each split. However the number of split points which needs to be examined while building the decision tree becomes e...

Cryptography is important in the field of information security. Cryptography, the science of code making and code breaking, keeps the defense, business and financial transactions, safe from unwanted elements and terrorists. Covert communications are kept secure by using a secret code to encrypt sensitive information. In the new digital age, secrecy...

Decision trees algorithms have been suggested in the past for classification of numeric as well as categorical attributes. SLIQ algorithm was proposed (Mehta et al., 1996) as an improvement over ID3 and C4.5 algorithms (Quinlan, 1993). Elegant Decision Tree Algorithm was proposed (Chandra et al. 2002) to improve the performance of SLIQ. In this pap...

In this paper, we present an open and safe nested transaction model and discuss the crash recovery issues. We introduce the notion of a recovery point subtransaction in a nested transaction tree. We introduce prewrite operations to increase concurrency. Our model is open and safe as prewrites allow early reads (before database writes on disk) witho...

The paper presents a new approach to interrelated two-way clustering of gene expression data. Clustering of genes has been effected using entropy and a correlation measure, whereas the samples have been clustered using the fuzzy C-means. The efficiency of this approach has been tested on two well known data sets: the colon cancer data set and the l...

Medical image analysis is an important part of image analysis applications. Cluster analysis can be effectively used in pattern recognition and image processing. This work presents a novel way of classification and clustering medical image data using self organizing neural networks with quadratic neural type junctions. The medical image data compri...

There has been an enormous increase in the crime in the recent past. Crime deterrence has become an upheaval task. The cops in their role to catch criminals are required to remain convincingly ahead in the eternal race between law breakers and law enforcers. One of the key concerns of the law enforcers is how to enhance investigative effectiveness...

The paper proposes a new selection measure for classification using decision trees for Data mining. Various algorithms have been proposed in the past for classification using decision trees viz. ID3, CART, SLIQ, etc. Selection measures like the Gain, Gain ratio, and Gini index have been proposed in these algorithms. However, none of the selection m...

There are some basic real life problems that cannot be solved using classical mathematical techniques. In this paper functional networks have been effectively used to solve practical CAD problems related to plant engineering industry. Modular construction of plants is becoming popular due to severe weather conditions at plant sites. The modules are...

Decision trees have been found very effective for classification especially in Data Mining. This paper aims at improving the performance of the SLIQ decision tree algorithm (Mehta et. al,1996) for classification in data mining The drawback of this algorithm is that large number of gini indices have to be computed at each node of the decision tree....

In this paper, we formalize and prove the correctness of a nested transaction version of the concurrency control algorithm using a linear hash structure. Nested transactions allow increased parallel execution of transactions, and handle transaction aborts in our system. We present our nested transaction model in a linear hash structure environment...

In this paper, we present, formalize and prove the correctness of recovery algorithm for our open and safe nested transaction model using I/O automaton model. Our nested transaction model uses the notion of a recovery point subtransaction in the nested transaction tree. It introduces a prewrite operation before each write operation to increase the...

In this paper, we present a formal description of the virtual partition algorithm in a nested transaction environment and prove its correctness. We model the virtual partition algorithm in a nested transaction environment using the I/O automaton model. The formal description is used to construct a complete correctness proof that is based on standar...

In this paper, we present an open and safe nested transaction model. We discuss the concurrency control and recovery algorithms for our model. Our nested transaction model uses the notion of a recovery point subtransaction in the nested transaction tree. It incorporates a prewrite operation before each write operation to increase the potential conc...

In this paper, we model the virtual partition algorithm in a nested transaction environment using I/O automaton model. The formal description is used to construct a complete correctness proof that is based on standard assertional techniques and on a natural correctness condition, and takes advantage of modularity that arises from describing the alg...

This paper discusses the estimation of parameters in second order exponential autoregressive models.

The paper discusses the implementation of the Newton-Raphson iterative method of estimation of parameters in the autoregressive integrated moving average (ARIMA) models. The efficiency of this method has been compared with other well known methods of estimation.

In this paper, a novel approach has been proposed to rank police administration units on the basis of their effective enforcement of crime prevention measures using Data Envelopment Analysis (DEA) and Clustering. The proposed approach will offer an effective mechanism not only to rank police administration units but also provide an evaluation tool...