|
Automated Prioritization for Monitoring and Tracking Service Requests using Named Entity Recognition Technique |
|---|---|
| รหัสดีโอไอ | |
| Creator | Chumpol Mokarat |
| Title | Automated Prioritization for Monitoring and Tracking Service Requests using Named Entity Recognition Technique |
| Contributor | Pattarak Sawatdee, Pimpika Intutsing |
| Publisher | Faculty of Informatics, Mahasarakham University |
| Publication Year | 2569 |
| Journal Title | Journal of Applied Informatics and Technology |
| Journal Vol. | 8 |
| Journal No. | 2 |
| Page no. | 262860 |
| Keyword | Automated Prioritization, Monitoring and Tracking, Named Entity Recognition, Service Requests |
| URL Website | https://ph01.tci-thaijo.org/index.php/jait |
| Website title | Journal of Applied Informatics and Technology |
| ISSN | 3088-1803 |
| Abstract | Service request prioritization in IT service management presents signifi-cant challenges in terms of accuracy and efficiency, particularly in Thai-language environments where automated support tools remain limited.This study develops and evaluates a web-based application designed tooptimize the management and tracking of organizational service requests.By incorporating automated prioritization through Named Entity Recog-nition (NER), the system aims to improve operational efficiency and re-duce the time staff spend on maintenance-related tasks. In addition toreal-time tracking, the application serves as a centralized repository forservice request data, enabling in-depth analysis and supporting strate-gic planning. This capability facilitates more informed decision-makingin organizational service management. A key feature of the proposedsystem is its ability to automatically prioritize incoming service requestsbased on extracted information. The core component is an NER modelfine-tuned using 500 service request samples. Four transformer-basedlanguage models, namely BERT-Base-Thai, BERT-Base-Multilingual,WangchanBERTa-Base, and XLM-RoBERTa-Base, were evaluated andcompared. The evaluation employed multiple performance metrics, in-cluding evaluation loss, precision, recall, F1-score, and accuracy, to deter-mine the most suitable model for the task. Experimental results indicatethat BERT-Base-Multilingual achieved the best performance, obtainingan evaluation loss of 0.0272, precision of 0.9888, recall of 0.9946, F1-scoreof 0.9917, and accuracy of 0.9923. System performance testing furtherdemonstrated that automatic priority assignment required less than 3.5seconds per request, while service resolution was completed within 48hours, resulting in an overall user satisfaction score of 4.16 out of 5. Thefindings confirm the effectiveness of the proposed NER model in accu-rately identifying multiple entity categories and demonstrate that the de-veloped system successfully enhances service request management whilereducing request-tracking time through automated log analysis. |
