AI-Powered Support Ticket Classification and Auto-Routing System using Machine Learning

Support ticket help desks are an essential part of modern companies. As developments in the field of ML advance, it becomes interesting to automate support ticket help desks using ML solutions to lower error rate and cost within the support department. In order to investigate the present state of the art in the field of ML-driven support ticket automating we conducted this Literature Review. We found that every paper analyzed is heavily influenced by the practical use case it was written about and that this is due to the individual nature of support ticket systems, the companies they are used in and the special, technical requirements raised by ML algorithms. We also found that the majority of the papers in the field is published on conferences. We found the topic of creating an automated incident management tool is the dominating topic in this field of research. ML algorithms like RF and SVM are currently best performing in ticket classification, while there is evidence that Deep Learning algorithms will take the lead in the near future. Two other important topics in the field found in this Literature Review were Request Escalation prediction, in which it is the goal to optimize processes within the support system handling customers escalating their requests, and customer sentiment prediction, in which it is the goal to predict the sentiment of a customer based on the tickets he/she created. Additionally, we found a great variety of topics individually dealt with in single papers. We provide an overview over the current technological state of the art, and give suggestions in which directions the research scope can be expanded, and identify research gaps.

[1]   Gupta, M., A. Asadullah, S. Padmanabhuni and A. Serebrenik, Reducing user input requests to improve IT support ticket resolution process. Empirical Software Engineering, 2018. 23(3): p. 1664-1703. [2]   İşcen, E.S. and M.Z. Gürbüz. A Comparison of Text Classifiers on IT Incidents Using WEKA. in 2019 4th International Conference on Computer Science and Engineering (UBMK). 2019. [3]  Qamili, R., S. Shabani and J. Schneider, An Intelligent Framework for Issue Ticketing System Based on Machine Learning. 2018. 79-86. [4]  Chagnon, C.T., A.C.; Djamasbi, S., Creating a decision support system for service classification and assignment through optimization, in AMCIS 2017 - America's Conference on Information Systems: A Tradition of Innovation. 2017, Worcester Polytechnic Institute, Germany. [5]  Montgomery, L., D. Damian, T. Bulmer and S. Quader, Customer support ticket escalation prediction using feature engineering. Requirements Engineering, 2018. 23(3): p. 333-355. [6] Nayebi, M., L. Dicke, R. Ittyipe, C. Carlson and G. Ruhe, ESSMArT way to manage customer requests. Empirical Software Engineering, 2019. 24(6): p. 3755-3789. [7]  Stein, N., C. Flath and C. Böhm, Predictive Analytics for Application Management Services. 2018. [8]  Al-Hawari, F. and H. Barham, A machine learning based help desk system for IT service management. Journal of King Saud University - Computer and Information Sciences, 2019. [9]  Silva, S., R. Pereira and R. Ribeiro. Machine learning in incident categorization automation. in 2018 13th Iberian Conference on Information Systems and Technologies (CISTI). 2018. [10]    Zuev, D., A. Kalistratov and A. Zuev, Machine Learning in IT Service Management. Procedia Computer Science, 2018. 145: p. 675-679.