Improving the Prediction Resolution Time for Customer Support Ticket System


Haw, Su Cheng and Ong, Kyle and Chew, Lit Jie and Ng, Kok Why and Naveen, Palanichamy and Anaam, Elham Abdulwahab (2022) Improving the Prediction Resolution Time for Customer Support Ticket System. Journal of System and Management Sciences, 12 (6). pp. 1-16. ISSN 1816-6075, 1818-0523

[img] Text
Vol.12.No.06.01.pdf - Published Version
Restricted to Repository staff only

Download (1MB)


Processing customer queries on time will able to engage customer satisfaction, and thus improve the customer retention of a company. Increasing the labour to process these queries is certainly not an ideal solution. Advancing technology such as artificial intelligence and machine learning has led to the goal of automating this process, by predicting the time needed to resolve certain issues based on past similar cases. In this paper, we present the architecture for the Customer Support Ticket System to improve the accuracy of the predicted resolution time. In this research, we first perform the one hot encoding on the categorical variables, followed by feature selection. Next, a combination of classification and regression models is being utilised in our prediction pipeline. Experimental evaluations demonstrated that the Random Forest (RF) regression model has the best performance as compared to Neural Network and ADA boost. In addition, by adding the extremity feature as the attention, a significant performance boost for RF is observed.

Item Type: Article
Uncontrolled Keywords: Prediction, predictive analytics, resolution time, customer support, ticket system
Subjects: H Social Sciences > HD Industries. Land use. Labor > HD9000-9999 Special industries and trades
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 22 Mar 2023 02:15
Last Modified: 22 Mar 2023 02:15


Downloads per month over past year

View ItemEdit (login required)