A Stacked Generalization Ensemble Model for Help Desk Ticket Assignment

The assignment of Help Desk Support tickets (HDTAP) to programmers or developers is an important problem for Information Technology and software development companies. When the number of tickets and programmers are significant, the assignment becomes a time-consuming task. To handle this issue, companies develop software to manage the generation, the tracking, and the assignment of tickets to employees. Current advances in machine learning can automate the HDTAP, and make it more efficient. In this paper, we propose the use of an ensemble model based on Stacked Generalization that replicates the expert behavior by reducing the classification errors. The model is tested using data of a Chilean company that develops management software for hospitals. The results show that our proposal obtains the best accuracy and $F_{1}$ score outperforming, consistently, most state-of-the-art models.