How kishanchauhan11/google-playstore-analytics works
Project type and purpose
• This is a data-analytics and visualization project built primarily in Python.
• It ingests Google Play Store data (two CSVs), cleans and processes it, performs sentiment analysis and machine-learning predictions, and exposes the results through interactive dashboards (web via Dash, desktop via Tkinter).
File structure
• Top-level files:
• Analysis.ipynb, Analysis2.ipynb, Analysis3.ipynb – each notebook contains chunks of the ETL, analytics, ML, and visualization code.
• Play Store Data.csv, User Reviews.csv – raw data sources.
• README.md – project overview.
• No dedicated “src” or “backend/frontend” folders. Functionally, notebooks can be divided into:
• Data Ingestion & Cleaning
• Sentiment Analysis
• Rating Prediction (ML)
• Dashboard (Dash)
• Desktop GUI (Tkinter)
Key insights from the README
• Technologies/libraries used: pandas, numpy, matplotlib/seaborn, Plotly, Dash, Tkinter, scikit-learn, NLTK.
• Main features: data cleaning, category/rating/installation analysis, sentiment analysis, rating prediction, interactive dashboards.
Elements
• Blue for data sources
• Green for processing/analysis modules
• Orange for machine learning
• Purple for UI layers
• Rectangles for compute modules
• Rounded rectangles for UI
Detailed capturing of essential elements
• Both CSV inputs separately feeding into a common ETL.
• Separate branches for sentiment vs. rating prediction.
• Converge results into a “Dashboard Data Store” layer feeding both UIs.