Performance Evaluation of Machine Learning Algorithms for Credit Card Fraud Detection
In this work, popular supervised and unsupervised machine learning algorithms have been applied to detect credit card frauds in a highly imbalanced dataset and it was found that unsuper supervised machineLearning algorithms can handle the skewness and give best classification results.
A review of some Bayesian Belief Network structure learning algorithms
- S. Mittal, S. Maskara
- Computer ScienceInternational Conference on Information…
- 1 December 2011
Six different algorithms have been reviewed by constructing BBN structures for two different datasets using various algorithms to draw inferences which may help in decision making.
Situation recognition in sensor based environments using concept lattices
- S. Mittal, A. Aggarwal, S. Maskara
- Computer ScienceCUBE International IT Conference & Exhibition
- 3 September 2012
A variety of sensors are available nowadays for fine grain continuous monitoring of the authors' environments in many desired ways and contexts, which may affect their actions and decisions are deduced using probabilistic methods like maximum likelihood estimation and Bayesian probabilities.
A threshold secret sharing technique based on matrix manipulation
- Shardha Porwal, S. Mittal
- Computer Science
- 1 March 2020
Computational Techniques for Real-Time Credit Card Fraud Detection
This chapter focuses on the security issues arising out of online credit card usage and most common attributes and open datasets of credit card transactions have been compiled to provide a starting point for new researchers.
Neutrosophic Concept Lattice based Approach for Computing Human Activities from Contexts
- S. Mittal, K. Gopal, S. Maskara
- Computer Science
- 1 January 2015
The proposed neutrosophic formal concept analysis method has been proposed to quantify non-determinism leading to ambiguity of interpretation and utilize it in activity recognition and successfully identified nondeterminism in activities description without compromising recognition performance of deterministic activities.
Implementation of Ciphertext Policy-Attribute Based Encryption (CP-ABE) for fine grained access control of university data
- Shardha Porwal, S. Mittal
- Computer ScienceInternational Conference on Contemporary…
- 1 August 2017
Experimental simulations demonstrate the capability of proposed solution in providing real time encryption/ decryption services on varying number of attributes, file sizes and types.
A Versatile Lattice Based Model For Situation Recognition From Dynamic Ambient Sensors
- S. Mittal, K. Gopal, S. Maskara
- Computer Science
- 1 January 2013
A versatile method based on formal concept analysis that works on any number and type of sensors data, and is also instance-independent and eliminates need of training, when applied to various instances of similar application.
Sampling Approaches for Imbalanced Data Classification Problem in Machine Learning
It has been observed that adaptive synthetic oversampling approach can best improve the imbalance ratio as well as classification results, however, undersampling approaches gave better overall performance on all datasets.
Application of Bayesian Belief Networks for context extraction from wireless sensors data
- S. Mittal, A. Aggarwal, S. Maskara
- Computer ScienceInternational Conference on Advanced…
- 3 April 2012
Given incomplete and erroneous nature of sensor data, Bayesian Belief Networks (BBN) are used here to obtain features defining context and simple rule based matching is applied to map the features to already defined context.
...
...