Communications networks now form the backbone of our digital world, with fast and reliable connectivity. However, even with appropriate redundancy and failover mechanisms, it is difficult to guarantee “five 9s” (99.999%) reliability, requiring rapid and accurate root cause analysis (RCA) during outages. In the event of an outage, rapid and accurate RCA becomes essential to restore service and prevent future disruptions. This study examines the capability and performance of three large language model (LLM) methods for building a knowledge base using support tickets to facilitate RCA and service assurance (SA) in communication networks. We comprehensively evaluate these models encompassing metrics that include lexical and semantic similarity between various LLMs and compare their performance and capabilities. Our experiments on a real dataset demonstrate that the generated knowledge base provides an excellent starting point for accelerating RCA tasks and improving network resilience.
Tran et al. (Thu,) studied this question.