Skip to Content
Loading Video...
Automatic classification of helpdesk tickets on the Odoo platform through Natural Language Processing techniques
Developer Community Talk
Location: Hall 7.C - 11/8/23, 11:30 AM - 11/8/23, 12:00 PM (Europe/Brussels) (30 minutes)
Automatic classification of  helpdesk tickets on the Odoo platform through Natural Language Processing techniques
Serena Palazzo
AI and Data Science Team Leader
Serena Palazzo
AI and Data Science Team Leader

Serena holds the role of Senior Data Scientist and leads the Data Science & Artificial Intelligence division at Hoverture, an Italian IT Company. She has a strong passion for Physics and programming. She has a PhD in experimental particle physics and worked for the ATLAS Collaboration at CERN. She started using Machine Learning and Deep Learning for research purpose and when deciding to move from Academia to industry, she continued to use Artificial intelligence and to apply the latest cutting-edge techniques for the industry.


In this talk, Serena will elucidate the functionalities of a software application named AI-TSC (Artificial Intelligence - Ticket System Classification). This application is based on Artificial Intelligence (AI) techniques and is designed for the automated categorization of helpdesk tickets within the Odoo platform. The development of AI-TSC involved the utilization of state-of-the-art AI methodologies in the realm of Natural Language Processing (NLP), which focuses on the interaction between human language and computers. The workflow of AI-TSC involves the reception of newly generated helpdesk tickets, subsequent processing that entails labeling, and final allocation to appropriate user groups or individuals based on the specific issue or inquiry. The creation of AI-TSC involved training AI models to accomplish two primary tasks: initially, to extract pertinent entities from the tickets, followed by the categorization of these tickets based on their content. The implementation of AI-TSC has yielded two notable advantages. Firstly, it has significantly expedited the identification and classification of the various scenarios encapsulated within the helpdesk tickets, facilitating prompt and precise resolution of client issues. Secondly, it has contributed to enhancing operational efficiency by enabling the handling of a larger volume of tickets within a reduced timeframe. This has been accompanied by a reduction in errors and anomalies attributed to manual ticket categorization. Beyond its immediate benefits, AI-TSC exhibits versatility as a modular tool. Its applicability extends to diverse processes such as task and project allocation, showcasing its potential as a multifunctional resource.