PulseBoard — Customer Review Analytics Platform

A 7-member Hackathon Project

PulseBoard is a collaborative hackathon project that combines Full Stack Development, Machine Learning, Data Analytics, and Business Intelligence to analyze customer reviews. The platform predicts customer sentiment using a Machine Learning model, exposes predictions through FastAPI REST APIs, stores processed data, and visualizes business insights through an interactive Power BI dashboard.

---

1. Project Overview

PulseBoard enables businesses to:

• Upload customer reviews in bulk.
• Perform Machine Learning based Sentiment Analysis.
• Visualize KPIs and review analytics.
• Generate business recommendations.
• Manage users with JWT Authentication.
• View interactive Power BI dashboards.

---

2. Technology Stack

Layer	Technologies
Frontend	React (Vite), Tailwind CSS, JavaScript, Axios, React Router
Backend	FastAPI, Python, SQLAlchemy, Pydantic, JWT Authentication
Database	SQLite
Machine Learning	Scikit-learn, TF-IDF Vectorizer, Sentiemental Naive Bayes, Joblib, Pandas, NumPy
Data Processing	Pandas, NumPy
Business Intelligence	Microsoft Power BI
API Communication	REST APIs, JSON
Version Control	Git, GitHub

---

3. Team Contributions

This project was developed collaboratively as a 7-member Hackathon Project. Each team member contributed across Full Stack Development, Machine Learning, Data Science, Data Analytics, and Business Intelligence.

Team Member	Role	Responsibilities
Shahina Sareen K T	Backend Developer • Frontend Developer • ML Integration Engineer	Developed backend APIs using FastAPI, implemented frontend modules, integrated the trained Machine Learning model into the backend, handled REST API requests/responses, authentication, database integration, frontend-backend communication, and overall project integration.
Rithesh A H	Full Stack Developer	Contributed to frontend and backend development, implemented application features, assisted with testing, debugging, project integration, and overall application development.
Nishanth J C	Machine Learning Engineer	Trained the sentiment analysis model, prepared the ML pipeline, evaluated model performance, optimized predictions, and deployed the trained model for backend integration.
Pradeep Hiremath	Data Scientist	Performed exploratory data analysis (EDA), feature engineering, dataset preparation, model evaluation, performance analysis, and provided insights to improve prediction accuracy.
Nandini Ganesh	Data Analyst	Performed data cleaning, preprocessing, dataset validation, exploratory data analysis, and prepared structured datasets for Machine Learning training.
Harish N	Data Analyst	Assisted in data cleaning, preprocessing, review categorization, statistical analysis, and validation of processed datasets before model training.
Sai Lokesh	Business Intelligence Analyst	Designed interactive Power BI dashboards, created business reports and visualizations, presented sentiment analytics, KPIs, and customer review insights for decision-making.

---

4. Architecture

React (Frontend)
        │
 REST API (Axios)
        │
     FastAPI
        │
 ├── Authentication
 ├── Review Upload
 ├── Dashboard APIs
 ├── Recommendation APIs
 └── ML Integration
        │
 Scikit-learn Sentiment Model
        │
      SQLite Database
        │
     Power BI Dashboard

---

5. Features

• Customer Review Upload
• Sentiment Analysis
• Customer Review Classification
• Dashboard Analytics
• Business Recommendations
• JWT Authentication
• REST API Architecture
• Power BI Integration
• Machine Learning Model Integration

---

6. Project Highlights

• Full Stack Web Application
• Machine Learning Powered Sentiment Analysis
• REST API Architecture
• Customer Review Analytics
• Interactive Power BI Dashboard
• SQLite Database Integration
• Secure Authentication
• Hackathon Project developed by a multidisciplinary team

---

License

This repository is intended for educational and hackathon purposes.