Case Study: Global Leading Fortune 500 Tech Company achieves improved sentiment classification accuracy with Mu Sigma

Global Leading Fortune 500 Tech Company - Customer Case Study

Global Leading Fortune 500 Tech Company partnered with Mu Sigma to improve its existing sentiment classifier. The customer was struggling with weak data cleaning, inaccurate sentiment detection, and difficulty prioritizing issues in a tool used for analyzing customer feedback and reviews.

Mu Sigma applied its AoPS-based muOBI approach and enhanced the classifier with an RNN model, Stanza NLP, a hybrid NLP/deep learning model, and a TASBA BERT model for target-level sentiment extraction. The upgraded solution improved overall accuracy from about 80% to 91%, delivered roughly 50% better text-to-sentence conversion and 30% better aspect identification, and added multi-target detection plus target-level sentiment extraction.