Please use this identifier to cite or link to this item:
|Title:||Improving the performance of classification models with fuzzy cognitive maps||Authors:||Christodoulou, Panayiotis
Andreou, Andreas S.
|Keywords:||Classification models;Fuzzy Cognitive Maps;Prediction accuracy||Category:||Computer and Information Sciences;Electrical Engineering - Electronic Engineering - Information Engineering||Field:||Natural Sciences;Engineering and Technology||Issue Date:||Jul-2017||Publisher:||Institute of Electrical and Electronics Engineers Inc.||Source:||IEEE International Conference on Fuzzy Systems, 2017, Naples, Italy, 9-12 July||DOI:||https://doi.org/10.1109/FUZZ-IEEE.2017.8015422||Conference:||IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)||Abstract:||This paper presents a novel approach to improve the accuracy of classification models used for prediction purposes by integrating a Fuzzy Cognitive Map (FCM) to produce a hybrid model. The proposed methodology first uses the FCM to discover latent correlations that exist between the data in order to form a single variable. This variable is then fed in the classification model as part of the training and testing phases to enhance its accuracy. Experimental results using datasets describing two different problems suggested noteworthy improvements in the accuracy of various classification models.||ISBN:||978-150906034-4||ISSN:||1558-4739||DOI:||10.1109/FUZZ-IEEE.2017.8015422||Rights:||© 2017 IEEE.||Type:||Conference Papers|
|Appears in Collections:||Δημοσιεύσεις σε συνέδρια/Conference papers|
Show full item record
Page view(s) 5060
checked on Aug 18, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.