The text mining method for failure analysis case based on Apriori algorithm

In complex application scenarios, traditional failure analysis techniques have problems such as high complexity, difficulty, and excessive reliance on expert experience. A framework for text mining of failure analysis cases based on Apriori algorithm was proposed, which realized auxiliary diagnosis of failure causes and explanation of fault action mechanisms. The framework was applied and verified based on historical failure case texts. The results show that the framework consisted of three main parts, a text preprocessing method for failure analysis cases, a text feature representation model for failure analysis cases based on TF-IDF algorithm, and a two-stage association rule mining method for failure analysis cases based on Apriori algorithm. The visualized propagation path of failure factors, failure modes, and failure causes established by this framework had reference significance for improving the efficiency of failure analysis and decoupling the process of failure cause inference.