How to Deal with Liars? Designing Intelligent Rule-Based Expert Systems to Increase Accuracy or Reduce Cost Cai, Yuanfeng Jiang, Zhengrui Jiang, Zhengrui Mookerjee, Vijay
dc.contributor.department Supply Chain Management 2018-02-19T01:17:57.000 2020-07-02T06:25:03Z 2020-07-02T06:25:03Z Fri Jan 01 00:00:00 UTC 2016 2016-01-01
dc.description.abstract <p>Input distortion is a common problem faced by expert systems, particularly those deployed with a Web interface. In this study, we develop novel methods to distinguish liars from truth-tellers, and redesign rule-based expert systems to address such a problem. The four proposed methods are termed <em>split tree</em> (ST), <em>consolidated tree</em> (CT), <em>value-based split tree</em> (VST), and <em>value-based consolidated tree</em> (VCT), respectively. Among them, ST and CT aim to increase an expert system’s accuracy of recommendations, and VST and VCT attempt to reduce the misclassification cost resulting from incorrect recommendations. We observe that ST and VST are less efficient than CT and VCT in that ST and VST always require selected attribute values to be verified, whereas CT and VCT do not require value verification under certain input scenarios. We conduct experiments to compare the performances of the four proposed methods and two existing methods, i.e., the traditional <em>true tree</em> (TT) method that ignores input distortion and the <em>knowledge modification</em> (KM) method proposed in prior research. The results show that CT and ST consistently rank first and second, respectively, in maximizing the recommendation accuracy, and VCT and VST always lead to the lowest and second lowest misclassification cost. Therefore, CT and VCT should be the methods of choice in dealing with users’ lying behaviors. Furthermore, we find that KM is outperformed by not only the four proposed methods, but sometimes even by the TT method. This result further confirms the advantage necessity of differentiating liars from truth-tellers when both types of users exist in the population.</p>
dc.description.comments <p>This is an accepted manuscript published as 3. Cai , Y., Z. Jiang, V. Mookerjee. "How to Deal with Liars? Designing Intelligent Rule-Based Expert Systems to Increase Accuracy or Reduce Cost." INFORMS Journal on Computing, 2017; 29(2); 268-286. <a href="" target="_blank">10.1287/ijoc.2016.0728</a>. Posted with permission.</p>
dc.format.mimetype application/pdf
dc.identifier archive/
dc.identifier.articleid 1029
dc.identifier.contextkey 10984882
dc.identifier.s3bucket isulib-bepress-aws-west
dc.identifier.submissionpath scm_pubs/30
dc.language.iso en
dc.source.bitstream archive/|||Fri Jan 14 23:27:33 UTC 2022
dc.source.uri 10.1287/ijoc.2016.0728
dc.subject.disciplines Business Intelligence
dc.subject.disciplines Management Information Systems
dc.subject.disciplines Marketing
dc.subject.disciplines Operations and Supply Chain Management
dc.subject.disciplines Technology and Innovation
dc.subject.keywords rule-base expert systems
dc.subject.keywords input distortion
dc.subject.keywords classification
dc.subject.keywords value-based methods
dc.title How to Deal with Liars? Designing Intelligent Rule-Based Expert Systems to Increase Accuracy or Reduce Cost
dc.type article
dc.type.genre article
dspace.entity.type Publication
relation.isAuthorOfPublication 1bc6119f-70fc-4d29-a55e-67199a177f4e
relation.isOrgUnitOfPublication ef3ab1b0-d571-4148-84dd-470ef1cdb17a
Original bundle
Now showing 1 - 1 of 1
737.02 KB
Adobe Portable Document Format