2,260 Hits in 8.2 sec

A Novel Gaussian Based Similarity Measure for Clustering Customer Transactions Using Transaction Sequence Vector

M.S.B. Phridviraj, Vangipuram RadhaKrishna, Chintakindi Srinivas, C.V. GuruRao
2015 Procedia Technology - Elsevier  
use them for defining the proposed measure.  ...  We then carry out the analysis for worst case, average case and best case situations. The Similarity measure designed is Gaussian based and preserves the properties of Gaussian function.  ...  In this paper, we design and define a similarity measure by defining two terms transaction sequence vector and transaction vector.  ... 
doi:10.1016/j.protcy.2015.02.126 fatcat:r5vzr23fivf55kiv7na3i2kwku

Identifying Hidden Buyers in Darknet Markets via Dirichlet Hawkes Process [article]

Panpan Zheng, Shuhan Yuan, Xintao Wu, Yubao Wu
2019 arXiv   pre-print
Case studies on real transaction sequences explicitly show that our approach can group transactions with similar patterns into the same clusters.  ...  To tackle this challenge, we propose a hidden buyer identification model, called UNMIX, which can group the transactions from one hidden buyer into one cluster given a transaction sequence from an anonymized  ...  (restaurant customer) are similar to it.  ... 
arXiv:1911.04620v1 fatcat:z5ahy75cdfhihdwtzcohrl6fsy

A Novel Statistical Approach to Detect Card Frauds Using Transaction Patterns

Chae Chang LEE, Ji Won YOON
2015 IEICE transactions on information and systems  
In this paper, we present new methods for learning the individual patterns of a card user's transaction amount and the region in which he or she uses the card, for a given period, and for determining whether  ...  Then, we classify legitimate transactions and fraudulent transactions by setting thresholds based on the learned individual patterns.  ...  Acknowledgments This research was supported by the MSIP (Ministry of Science, ICT & Future Planning), Korea, under the "Employment Contract based Master's Degree Program for Information Security" supervised  ... 
doi:10.1587/transinf.2014edp7071 fatcat:qbhtinyoirbvnjhv7wvnnemaby

Variational Autoencoders and Wasserstein Generative Adversarial Networks for Improving the Anti-Money Laundering Process

ZhiYuan Chen, Waleed Soliman, Amril Nazir, Mohammad Shorfuzzaman
2021 IEEE Access  
Also, Wasserstein Generative Adversarial Network (WGAN) is used to generate fraud transactions, which are then mixed with the base dataset to form a more balanced mixed dataset.  ...  We preprocess the given dataset to separate the Transaction Date attribute into its base components to capture time-related fraud patterns.  ...  However, the data were clustered by customer attributes to build a customer profiles table, and then the PART algorithm was used for rule generation.  ... 
doi:10.1109/access.2021.3086359 fatcat:24qmwl63ebaffdjwg6fge3eyei

Approaches to Fraud Detection on Credit Card Transactions Using Artificial Intelligence Methods [article]

Yusuf Yazici
2020 arXiv   pre-print
These researches generally use rule-based or novel artificial intelligence approaches to find eligible solutions.  ...  Also, adapting the detection system to real time scenarios is a challenge since the number of credit card transactions in a limited time period is very high.  ...  The features created by HMM quantify how similar a sequence is to a past sequence of a cardholder or terminal [10] .  ... 
arXiv:2007.14622v1 fatcat:jt4pzmk6qjcr3jhng2orv5opvm

Relationship-Based Clustering and Visualization for High-Dimensional Data Mining

Alexander Strehl, Joydeep Ghosh
2003 INFORMS journal on computing  
While two-dimensional visualization of a similarity matrix is by itself not novel, its combination with the order-sensitive partitioning of a graph that captures the relevant similarity measure between  ...  This paper proposes a relationship-based approach that alleviates both problems, side-stepping the "curseof-dimensionality" issue by working in a suitable similarity space instead of the original high-dimensional  ...  We also thank Arindam Banerjee for processing the web-log data into web sessions.  ... 
doi:10.1287/ijoc. fatcat:ftbozoo27rbrfob2jmmmvkglay

A data mining approach using transaction patterns for card fraud detection [article]

Chae Chang Lee, Ji Won yoon
2013 arXiv   pre-print
However, card transaction features are targeted by criminals seeking to use a lost or stolen card and looking for a chance to replicate it.  ...  In this paper, we present methods for learning the individual patterns of a card user's transaction amount and the region in which he or she uses the card, for a given period, and for determining whether  ...  used a binary support vector system based on the support vectors in support vector machines and a genetic algorithm to solve credit card fraud problems that had not been well identified (?).  ... 
arXiv:1306.5547v1 fatcat:2cekkdawpba7vauwivikjolsb4

Role-based lateral movement detection with unsupervised learning [article]

Brian A. Powell
2021 arXiv   pre-print
We use unsupervised learning to cluster systems according to role and identify connections to systems with novel roles as potentially malicious.  ...  We apply frequent-itemset mining to process sequences to establish regular patterns of communication between systems based on role, and identify rare process sequences as signalling potentially malicious  ...  A combination of Naive Bayes and similarity measures was proposed in [54] , where similarity based on Gaussian kernels was shown to yield general improvement over Naive Bayes.  ... 
arXiv:2108.02713v1 fatcat:yjfztwyfojbxpbq2qsv4ymfyqa

Mining web content usage patterns of electronic commerce transactions for enhanced customer services

Sylvanus A. Ehikioya, Jinbo Zeng
2021 Engineering Reports  
A successful business intelligence solution can help organizations improve the quality and speed of their decision-making processes by analyzing the consolidated information collected from their websites  ...  To calculate the similarity matrix for the derived association rules, they used the Euclidean distance measure, and then used the single linkage agglomerative clustering algorithm to cluster the association  ...  Jiang and Yu 118 use K-means algorithm to cluster transactions data based on customer usage data of various e-commerce Websites.  ... 
doi:10.1002/eng2.12411 fatcat:yp57yt6jkvgv7h6oz6vvzxk4fm

A Comprehensive Survey on Machine Learning Techniques and User Authentication Approaches for Credit Card Fraud Detection [article]

Niloofar Yousefi, Marie Alaghband, Ivan Garibay
2019 arXiv   pre-print
and identifying illicit transactions as quickly as possible to protect themselves and their customers.  ...  In the second part of our survey, we review more advanced techniques of user authentication, which use behavioral biometrics to identify an individual based on his/her unique behavior while he/she is interacting  ...  This may lead to instability problems in clustering-based methods. A K-means clustering algorithm is used in (Vaishali, 2014) to detect fraudulent transactions.  ... 
arXiv:1912.02629v1 fatcat:cnedguwpkveydnlbmdljm65t24

An Artificial Intelligence Approach to Financial Fraud Detection under IoT Environment: A Survey and Implementation

Dahee Choi, Kyungho Lee
2018 Security and Communication Networks  
Therefore, we have surveyed financial fraud methods using machine learning and deep learning methodology, mainly from 2016 to 2018, and proposed a process for accurate fraud detection based on the advantages  ...  Financial fraud under IoT environment refers to the unauthorized use of mobile transaction using mobile platform through identity theft or credit card stealing to obtain money fraudulently.  ...  BLAST and SSAHA algorithm are sequence alignment algorithms and used as the alignment of sequences for an efficient technique to examine the spending behavior of customers [11] .  ... 
doi:10.1155/2018/5483472 fatcat:ne3rugm3xrg6hgo4h3b7pqik2i

Calibration of Shared Equilibria in General Sum Partially Observable Markov Games [article]

Nelson Vadori and Sumitra Ganesh and Prashant Reddy and Manuela Veloso
2020 arXiv   pre-print
Form Game (FFG) and prove convergence to the latter for a certain class of games using self-play.  ...  In addition, it is important that such equilibria satisfy certain constraints so that MAS are calibrated to real world data for practical use: we solve this problem by introducing a novel dual-Reinforcement  ...  We split 500 customers into 10 customer clusters, cluster i ∈ [1, 10] being associated to quantity i. For example, a customer belonging to cluster 5 will generate transactions of quantity 5.  ... 
arXiv:2006.13085v5 fatcat:zuhkmda6lbhsfdwb6mdd5iqdk4

Bayesian-OverDBC: A Bayesian Density-Based Approach for Modeling Overlapping Clusters

Mansooreh Mirzaie, Ahmad Barani, Naser Nematbakkhsh, Majid Mohammad-Beigi
2015 Mathematical Problems in Engineering  
Therefore, a probability model for overlap density-based clustering is a critical need for large data analysis.  ...  In this paper, a new Bayesian density-based method (Bayesian-OverDBC) for modeling the overlapping clusters is presented. Bayesian-OverDBC can predict the formation of a new cluster.  ...  This algorithm is a novel density-based clustering algorithm that has several advantages over traditional algorithms.  ... 
doi:10.1155/2015/187053 fatcat:wdjappv2uvao7m6s3tnzzs7rlu

MEMPower: Data-Aware GPU Memory Power Model [chapter]

Jan Lucas, Ben Juurlink
2019 Lecture Notes in Physics  
We explain how the model was calibrated using special micro benchmarks as well as a high-resolution power measurement testbed.  ...  A novel technique to identify the number of memory channels and the memory channel of a specific address is presented.  ...  A prior version of this work is part of the first author's defended, but currently unpublished, doctoral thesis.  ... 
doi:10.1007/978-3-030-18656-2_15 fatcat:bl32vsmrj5g5vp3jqrumzsojwu

On Approximation Algorithms for Data Mining Applications [chapter]

Foto N. Afrati
2006 Lecture Notes in Computer Science  
Acknowledgements: Thanks to Chen Li and Ioannis Milis for reading and providing comments in an earlier version of this chapter.  ...  The proposed measure of similarity is based on the L p norm for various values for p (non-integral too).  ...  of data exhibit "similar features". • Sequence matching: Given a collection of sequences and a sequence query, find the sequence which is closest to the query-sequence. • Clustering: Partition a given  ... 
doi:10.1007/11671541_1 fatcat:g2nm45u545audjojlxh6a7zpzu
« Previous Showing results 1 — 15 out of 2,260 results