Automated IT Service Desk Systems Using Machine Learning Techniques

Managing problem tickets is a key issue in any IT service industry. The routing of a problem ticket to the proper maintenance team is very critical step in any service desk (Helpdesk) system environment. Incorrect routing of tickets results in reassignment of tickets, unnecessary resource utilization, user satisfaction deterioration, and have adverse financial implications for both customers and the service provider. To overcome this problem, this paper proposes a service desk ticket classifier system which automatically classifies the ticket using ticket description provided by user. By mining historical ticket descriptions and label, we have built a classifier model to classify the new tickets. A benefit of building such an automated service desk system includes improved productivity, end user experience and reduced resolution time. In this paper, different classification algorithms like Multinomial Naive Bayes, Logistic regression, K-Nearest neighbor and Support vector machines are used to build such a ticket classifier system and performances of classification models are evaluated using various performance metrics. A real-world IT infrastructure service desk ticket data is used for this research purpose. Key task in developing such a ticket classifier system is that the classification has to happen on the unstructured noisy data set. Out of the different models developed, classifier based on Support Vector Machines (SVM) performed well on all data samples.