Enhanced Root Cause Analysis Using Rule based Aspect Extraction and RNN (RAER)

Blessy Selvam, Dr.S. Ravimaran and Dr. Sheba Selvam

Root cause analysis plays a vital role in the process of decision making in many organizations. Effective decision making requires effective root cause analysis model. However, the enormity of data being processed and the complexity involved in human cognition poses huge challenges to the process. This work presents an effective root cause identification architecture that enables effective decision making. The proposed architecture is composed of three major components; training data preparation, deep learning based sentiment identification and the final ranking mechanism to shortlist the most significant root causes for the user. Experiments were performed with the standard benchmark data and comparisons indicate improvements of ~10% in terms of precision, recall and F-Measure, exhibiting the efficiency of the proposed model.

Volume 11 | 04-Special Issue

Pages: 332-338