Active Learning SVM Classification Algorithm for Complaints Management Process Automatization

2019 60th International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS 2019): Proceedings 2019
Pāvels Gončarovs

Data analytics capabilities integrated with Customer Relationship Management Systems enable automatization activities at financial institutions. This paper reports an experimental study on developing a text-mining model to automate complaints management process. Automatization will reduce the work load and waiting time for the customer to process the complaint and omit manual work. The support-vector machines (SVM), a supervised learning model with associated learning algorithms, is used to analyse input complaints data for classification. The experimental study emphasizes importance of active learning cycle. It has been shown that for Active Learning SVM just a small part of the training set (about 20%) is needed to get a stable and good enough classifier.

Keywords
Active Learning, support-vector machine (SVM), classification, text-mining
DOI
10.1109/ITMS47855.2019.8940658
Hyperlink
https://ieeexplore.ieee.org/document/8940658

Gončarovs, P. Active Learning SVM Classification Algorithm for Complaints Management Process Automatization. In: 2019 60th International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS 2019): Proceedings, Latvia, Riga, 11-11 October, 2019. Piscataway: IEEE, 2019, pp.1-3. ISBN 978-1-7281-5710-8. e-ISBN 978-1-7281-5709-2. Available from: doi:10.1109/ITMS47855.2019.8940658

Publication language
English (en)