Projects

Investigating Techniques for Aspect-Based Sentiment Analysis in Employee Feedback

Built an aspect‑level sentiment system for employee feedback using BERT/RoBERTa/T5 and GPT variants, with domain adaptation and augmentation. Evaluated using F1 and human judgment.

PythonTransformersscikit‑learnPandasNotebook Experiments
  • Extracted aspect–sentiment pairs (e.g., work‑life balance: positive; salary: negative)
  • Improved interpretability with aspect taxonomy + example rationales
  • Comparing different large language model (LLM) architectures to better understand their progress in fine-grained aspect-based sentiment analysis.

Integrated Mobile Application Portal for the Convinient Service of Farmer, Logistics, Street Vendors AND Consumer

Developed a Mobile Application for Farmers, Consumer, Street Vendors and also the logistics. And the Admin will update the day to day process form the web server.

Mobile Application DevelopmentFlutterDatabase managementSQL
  • They can sell directly to consumers without middlemen, ensuring fair prices and higher revenue. This reduces exploitation and avoids commission/rent deductions.
  • Fresh produce is delivered at a lower cost and with convenience, reducing the need to visit crowded markets. Street vendors also benefit from discounted rates, ensuring sustainability for small businesses.

Databricks DLT Data Warehouse Project

The goal of this project is to build a scalable and incremental Data Lakehouse pipeline using Databricks Delta Live Tables (DLT), which processes raw data into structured dimension and fact tables following the star schema model. The pipeline supports incremental loads using Change Data Capture (CDC) logic and uses surrogate key joins for creating dimensional relationships.

DatabricksPysparkDelta Live TablesDelta Lake
  • Ingest raw data to Bronze via Auto Loader. Clean and transform to Silver using DLT
  • Create SCD-compliant Dimensions (Flights, Passengers, Airports)
  • Incrementally merge into Gold fact using Delta Lake