Recommendation Systems: Movie Predictor using NLTK and Vectorization
- Created a database of 5000 movies using TMDB Dataset (kaggle) and written a python script that recommends (ranked) 5 Similar Movies based on user input.
- Text Similarity using Vectorization that finds out Similarity Score of Tags using Cosine Similarity (Distance) which in turn predicts closest 5 Substitutes
- Used Streamlit Library in PyCharm to deploy the Movie Recommender System on Local Host where user can interact and get recommendations along with other details(Posters etc.) about movies
Bangalore House Price Prediction Model Using Data Science and Machine Learning
- Performed Data Cleaning operations using Pandas and Numpy Libraries and applied Linear Regression, Lasso and Ridge Models to check the differences in their Prediction Levels
- Achieved an accuracy level of 82% using Linear Regression and displayed the Total Price of Property based on user choice of No .of BHKs, Locality, No of Bathrooms and Carpet Area