<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>1534</id>
    <completedYear>2022</completedYear>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>831</pageFirst>
    <pageLast>858</pageLast>
    <pageNumber>28 Seiten</pageNumber>
    <edition/>
    <issue>64</issue>
    <volume>2022</volume>
    <type>article</type>
    <publisherName>Springer Nature</publisherName>
    <publisherPlace>Berlin</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-02-06</completedDate>
    <publishedDate>2022-02-06</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">NAG: neural feature aggregation framework for credit card&#13;
fraud detection</title>
    <abstract language="eng">The state-of-the-art feature-engineering method for fraud classification of electronic pay-ments uses manually engineered feature aggregates, i.e., descriptive statistics of thetransaction history. However, this approach has limitations, primarily that of being dependenton expensive human expert knowledge. There have been attempts to replace manual aggre-gation through automatic feature extraction approaches. They, however, do not consider thespecific structure of the manual aggregates. In this paper, we define the novel Neural Aggre-gate Generator (NAG), a neural network-based feature extraction module that learns featureaggregates end-to-end on the fraud classification task. In contrast to other automatic featureextraction approaches, the network architecture of the NAG closely mimics the structureof feature aggregates. Furthermore, the NAG extends learnable aggregates over traditionalones through soft feature value matching and relative weighting of the importance of differ-ent feature constraints. We provide a proof to show the modeling capabilities of the NAG.We compare the performance of the NAG to the state-of-the-art approaches on a real-worlddataset with millions of transactions. More precisely, we show that features generated with theNAG lead to improved results over manual aggregates for fraud classification, thus demon-strating its viability to replace them. Moreover, we compare the NAG to other end-to-endapproaches such as the LSTM or a generic CNN. Here we also observe improved results. Weperform a robust evaluation of the NAG through a parameter budget study, an analysis of theimpact of different sequence lengths and also the predictions across days. Unlike the LSTMor the CNN, our approach also provides further interpretability through the inspection of itsparameters.</abstract>
    <parentTitle language="eng">Knowledge and Information Systems</parentTitle>
    <identifier type="doi">10.1007/s10115-022-01653-0</identifier>
    <identifier type="urn">urn:nbn:de:101:1-2022060219175235764329</identifier>
    <licence>Creative Commons - CC BY - Namensnennung 4.0 International</licence>
    <author>Kanishka Ghosh Dastidar</author>
    <author>Johannes Jurgovsky</author>
    <author>Wissam Siblini</author>
    <author>Michael Granitzer</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>finance</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>credit card fraud</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>representation learning</value>
    </subject>
    <collection role="ddc" number="004">Datenverarbeitung; Informatik</collection>
    <collection role="ddc" number="006">Spezielle Computerverfahren</collection>
    <collection role="open_access" number="">open_access</collection>
    <collection role="institutes" number="">Fakultät für Informatik und Mathematik</collection>
    <collection role="Transformationsvertrag" number="">DEAL Springer Nature</collection>
    <thesisPublisher>Universität Passau</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-uni-passau/files/1534/GhoshDastidar_NAG.pdf</file>
  </doc>
  <doc>
    <id>762</id>
    <completedYear>2019</completedYear>
    <publishedYear>2019</publishedYear>
    <thesisYearAccepted>2019</thesisYearAccepted>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>xvii, 160 Seiten</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <type>doctoralthesis</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2019-12-20</completedDate>
    <publishedDate>2019-12-20</publishedDate>
    <thesisDateAccepted>2019-12-03</thesisDateAccepted>
    <title language="eng">Context-Aware Credit Card Fraud Detection</title>
    <abstract language="eng">Credit card fraud has emerged as major problem in the electronic payment sector. In this thesis, we study data-driven fraud detection and address several of its intricate challenges by means of machine learning methods with the goal to identify fraudulent transactions that have been issued illegitimately on behalf of the rightful card owner. In particular, we explore several means to leverage contextual information beyond a transaction’s basic attributes on the transaction level, sequence level and user level.&#13;
On the transaction level, we aim to identify fraudulent transactions which, in terms of their attribute values, are globally distinguishable from genuine transactions. We provide an empirical study of the influence of class imbalance and forecasting horizons on the classification performance of a random forest classifier. We augment transactions with additional features extracted from external knowledge sources and show that external information about countries and calendar events improves classification performance most noticeably on card-not-present transactions.&#13;
On the sequence level, we aim to detect frauds that are inconspicuous in the background of all transactions but peculiar with respect to the short-term sequence they appear in. We use a Long Short-term Memory network (LSTM) for modeling the sequential succession of transactions. Our results suggest that LSTM-based modeling is a promising strategy for characterizing sequences of card-present transactions but it is not adequate for card-not-present transactions.&#13;
On the user level, we elaborate on feature aggregations and propose a flexible concept allowing us define numerous features by means of a simple syntax. We provide a CUDA-based implementation for the computationally expensive extraction with a speed-up of two orders of magnitude over a single-core implementation. Our feature selection study reveals that aggregates extracted from users’ transaction sequences are more useful than those extracted from merchant sequences. Moreover, we discover multiple sets of candidate features with equivalent performance as manually engineered aggregates while being structurally different.&#13;
Regarding future work, we motivate the usage of simple and transparent machine learning methods for credit card fraud detection and we sketch a simple user-focused modeling approach.</abstract>
    <identifier type="urn">urn:nbn:de:bvb:739-opus4-7622</identifier>
    <licence>Standardbedingung laut Einverständniserklärung</licence>
    <author>Johannes Jurgovsky</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Credit Card Fraud Detection</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Machine Learning</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Data Augmentation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Feature Engineering</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>swd</type>
      <value>Kreditkartenmissbrauch</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>swd</type>
      <value>Computersicherheit</value>
    </subject>
    <collection role="ddc" number="004">Datenverarbeitung; Informatik</collection>
    <collection role="open_access" number="">open_access</collection>
    <collection role="institutes" number="">Fakultät für Informatik und Mathematik</collection>
    <thesisPublisher>Universität Passau</thesisPublisher>
    <thesisGrantor>Universität Passau</thesisGrantor>
    <file>https://opus4.kobv.de/opus4-uni-passau/files/762/johannes_jurgovsky_creditcardfraud.pdf</file>
  </doc>
</export-example>
