Exploration for Stakeholders
Fast and scalable analysis of reviews of mobile application company, automation of marking these reviews by category and harmonization of the taxonomy of these categories involve the creation of solutions in the form of exploratory instruments with automatic categorization and subsequent visualization of the analysis results. The solution allows evaluating feedback sentiment and establishing data trends to help improve the company’s mobile application.
Input Data
- Reviews about the mobile application from the AppStore, Google Play, from other sources
- Assessment reviews, user affixed
- Original taxonomy of review categories
Output Data
- Model categorization taxonomy reviews and predictions sentiment
- Recommendations for taxonomy harmonization
- Web application with aggregation analytics reviews
Main Results
Integration into the BI department of work with clients
Automation Review categorization:
- Increasing accuracy of 15% (reduction of human error)
- Reduction of the analysis time by 85%
Taxonomy harmonization:
- Reducing time to identify a new category by 60%
- Adding and merging 15% of categories
Sentiment assessment and trend analysis:
- Identify growth points when analyzing reviews by 40% more
- Reducing the time to identify critical and reputation-affecting reviews by 70%
Final quality of categorization:
| Quality criterion | Upper | Lower |
|---|---|---|
| Accuracy | 81 % | 70 % |
| Precision | 78 % | 66 % |
| Recall | 68 % | 62 % |
| F1-score | 72 % | 69 % |
Final quality of the prediction of sentiment:
| Quality criterion | Quality |
|---|---|
| MSE | 81 % |
