ICTACT Journal on Soft Computing

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

A Deep Learning Based it Service Desk Ticket Classifier Using CNN


Affiliations
1 Department of Computer Science and Engineering, School of Engineering, Central University of Karnataka, India
2 Department of Studies in Computer Science and Engineering, University B.D.T College of Engineering, India
     

   Subscribe/Renew Journal


Assignment of problem tickets to a proper resolver group is an important aspect and crucial step in any IT Service management tools like IT Service desk systems. Manual categorization of tickets may lead to dispatching of problem tickets to an inappropriate expert group, reassignment of tickets, delays the response time and interrupts the normal functioning of the business. Traditional supervised machine learning approaches can be leveraged to train an automated service desk ticket classifier by using the historical ticket data. Sparsity, non-linearity, overfitting and handcrafting of features are some of the issues concerning the traditional ticket classifiers. In this research work, a deep neural network based on Convolution Neural Network (CNN) is proposed for the automated classification of service desk tickets. CNN automatically extracts the most salient features of the ticket descriptions represented using word embeddings. The extracted features are further used by the output classification layer for efficient ticket category prediction. To corroborate the efficacy of the proposed ticket classifier model, we empirically validated it using a real IT infrastructure service desk data and compared the results with the traditional classifier models like Support Vector machines, Naive Bayes, Logistic Regression and K-nearest neighbour. The proposed CNN model with proper hyperparameters tuning outperforms the traditional classifiers in terms of overall model performance. Assignment of tickets to the correct domain groups, speedy resolution, improved productivity, increased customer satisfaction and uninterrupted business are some of the benefits of the proposed automated ticket classifier model.

Keywords

Service desk, Machine learning, Deep neural networks, Convolution Neural Network, Word Embeddings
Subscription Login to verify subscription
User
Notifications
Font Size

  • P. Kubiak and S. Rass, “An Overview of Data-Driven Techniques for IT-Service-Management”, IEEE Access, Vol.6, pp. 63664-63688, 2018.

  • M. Jantti, A. Cater-Steel and A. Shrestha, “Towards an Improved IT Service Desk System and Processes: A Case Study”, International Journal on Advances in Systems and Measurements, Vol. 5, No. 3-4, pp. 203-215, 2012.

  • F. Al-Hawari and H. Barham, “A Machine Learning Based Help Desk System for IT Service Management”, Journal of King Saud University-Computer and Information Sciences, Vol. 33, No. 6, pp. 702-718, 2021.

  • A. Revina, K. Buza and V.G. Meister, “IT Ticket Classification: The Simpler, the Better”, IEEE Access, Vol. 8, pp. 193380-193395, 2020.

  • S.P. Paramesh and K.S. Shreedhara, “IT Helpdesk Incident Classification Using Classifier Ensembles”, ICTACT Journal of Soft Computing, Vol. 9, No. 4, pp. 1980-1987, 2019.

  • N.A. Harun, S.H. Huspi and N.A. Iahad, “Question Classification Framework for Helpdesk Ticketing Support System using Machine Learning”, Proceedings of International Conference on Research and Innovation in Information Systems, pp. 1-7, 2021.

  • S.P. Paramesh and K.S. Shreedhara, “Automated IT Service Desk Systems using Machine Learning Techniques”, Proceedings of International Conference on Data Analytics and Learning, pp. 331-346, 2018.

  • A. Mandal, N. Malhotra, S. Agarwal and G. Sridhara, “Cognitive System to Achieve Human-Level Accuracy in Automated Assignment of Helpdesk Email Tickets”, Proceedings of International Conference on Service-Oriented Computing, pp. 332-341, 2018.

  • N. Pandey and Amitava Sen, “Automated Classification of Software Issue Reports using Machine Learning Techniques: an Empirical Study”, Innovations in Systems and Software Engineering, Vol. 13, No. 4, pp. 279-297, 2017.

  • S. Roy, D.P. Muni, J.Y.T. Yan, N. Budhiraja and F. Ceiler, “Clustering and Labeling IT Maintenance Tickets”, Proceedings of International Conference on Service- Oriented Computing, pp. 829-845, 2016.

  • Cristina Kadar, Dorothea Wiesmann, Jose Iria, Dirk Husemann and Mario Lucic, “Automatic Classification of Change Requests for Improved it Service Quality”, Proceedings of Annual SRII Global Conference, pp. 430-439, 2011.

  • M. Ikonomakis, S. Kotsiantis and V. Tampakas, “Text Classification using Machine Learning Techniques”, WSEAS Transactions on Computers, Vol. 4, No. 8, pp. 966-974, 2005.

  • M.M. Mironczuk and J. Protasiewicz, “A Recent Overview of the State-of-the-Art Elements of Text Classification”, Expert Systems with Applications, Vol. 106, pp. 36-54, 2018.

  • K. Kowsari, K.J. Meimandi, M. Heidarysafa, S. Mendu, L.E. Barnes and D.E. Brown, “Text Classification Algorithms: A Survey”, Proceedings of International Conference on Computation and Language, pp. 1-7, 2019.

  • R. Collobert, J. Weston, L. Bottou, M. Karlen, K. Kavukcuoglu and P. Kuksa, “Natural Language Processing (Almost) from Scratch”, Journal of Machine Learning Research, Vol. 12, No. 3, pp. 2493-2537, 2011.

  • Yoon Kim, “Convolutional Neural Networks for Sentence Classification”, Proceedings of International Conference on Empirical Methods in Natural Language Processing, pp. 1746-1751, 2014.

  • N. Kalchbrenner, E. Grefenstette and P. Blunsom, “A Convolutional Neural Network for Modelling Sentences”, Proceedings of Annual Meeting of the Association for Computational Linguistics, 2014.

  • Y. Zhang and B. Wallace, “A Sensitivity Analysis of (and Practitioners’ Guide to) Convolutional Neural Networks for

  • Sentence Classification”, Proceedings of International Joint Conference on Natural Language Processing, pp. 253-263, 2017.

  • T. Young, D. Hazarika, S. Poria and E. Cambria, “Recent Trends in Deep Learning Based Natural Language Processing”, IEEE Computational Intelligence Magazine, Vol. 13, No. 3, pp. 55-75, 2018.

  • N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever and R. Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting”, The Journal of Machine Learning Research, Vol.15, No. 1, pp. 1929-1958, 2014.

  • A. Krizhevsky, I. Sutskever, and G.E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks”, Proceedings of International Conference on Neural Information Processing Systems, pp.1097-1105, 2012.

  • David E. Rumelhart and James L. McClelland, “Learning Internal Representations by Error Propagation”, MIT Press, 1986.


Abstract Views: 180

PDF Views: 2




  • A Deep Learning Based it Service Desk Ticket Classifier Using CNN

Abstract Views: 180  |  PDF Views: 2

Authors

S. P. Paramesh
Department of Computer Science and Engineering, School of Engineering, Central University of Karnataka, India
K. S. Shreedhara
Department of Studies in Computer Science and Engineering, University B.D.T College of Engineering, India

Abstract


Assignment of problem tickets to a proper resolver group is an important aspect and crucial step in any IT Service management tools like IT Service desk systems. Manual categorization of tickets may lead to dispatching of problem tickets to an inappropriate expert group, reassignment of tickets, delays the response time and interrupts the normal functioning of the business. Traditional supervised machine learning approaches can be leveraged to train an automated service desk ticket classifier by using the historical ticket data. Sparsity, non-linearity, overfitting and handcrafting of features are some of the issues concerning the traditional ticket classifiers. In this research work, a deep neural network based on Convolution Neural Network (CNN) is proposed for the automated classification of service desk tickets. CNN automatically extracts the most salient features of the ticket descriptions represented using word embeddings. The extracted features are further used by the output classification layer for efficient ticket category prediction. To corroborate the efficacy of the proposed ticket classifier model, we empirically validated it using a real IT infrastructure service desk data and compared the results with the traditional classifier models like Support Vector machines, Naive Bayes, Logistic Regression and K-nearest neighbour. The proposed CNN model with proper hyperparameters tuning outperforms the traditional classifiers in terms of overall model performance. Assignment of tickets to the correct domain groups, speedy resolution, improved productivity, increased customer satisfaction and uninterrupted business are some of the benefits of the proposed automated ticket classifier model.

Keywords


Service desk, Machine learning, Deep neural networks, Convolution Neural Network, Word Embeddings

References