An ambitious project where we use AI/ML to provide a 360-degree analytical view on the existing product such as:
[T1] what’s going right [T2] what needs improvement [T3] market competition [T4] inventory management
Review Sentinel is one of the many modules of the project "Product Zen 360". This is an end-to-end machine learning-driven project designed to detect and flag computer-generated fake product reviews, ensuring authentic customer feedback and maintaining the integrity of online reviews.
- Data Profiling(EDA) | Preliminary Analysis
- Data Preprocessing | Baseline Predictions
- Experiment Tracking and Model Registry
- Workflow Orchestration
- Model Deployment
- Model Monitoring
- Add reproducibility instructions | Documentation
- Containerize development and deployment with localstack in docker
- Add unit and integration tests
- Use linters/formatters
- Add makefile
- Use pre-commit hooks
- CI/CD pipeline
- Webscraping for new reviews
- LLMs based Review generation
Dataset Used : https://osf.io/tyue9/