Hey DEV Community!
I recently participated in a hackathon and built Samiksha AI, a universal review and comment analyzer designed to turn messy customer feedback into structured business intelligence.
The Problem
Generic star ratings on e-commerce platforms don't give business owners the full picture. Rule-based sentiment tools often fail to catch sarcasm, background noise (like shipping complaints), or aspect-specific product feedback.
How I Built It
I developed a python-based architecture that leverages:
- Google AI Studio (Gemini API): For advanced parsing, noise filtering, and Aspect-Based Sentiment Analysis (ABSA).
- Streamlit Framework: To build an interactive, real-time executive report card dashboard.
- Plotly/Matplotlib: For rendering data distribution metrics dynamically.
Repository & Code Architecture
The code is structured with a clean separation between the data backend, core analytical engine, and user interface. You can check out the full open-source implementation here:
GitHub Repository: https://github.com/TechSakhi/Samiksha-AI
Future Roadmaps
Building this pipeline gave me a deep appreciation for computational linguistics and structured datasets. Next, I plan to leverage these foundational concepts to explore building an AI agent focused on language learning systems!
I would love to hear your thoughts, feedback, or suggestions on how to improve the architecture!
Top comments (0)