Challenge: Bolt-loosening issues of Nexon suspension joints Performed EDA on the data of the bolted joints. Used Logistic regression model to classify the joint as correct or incorrect. EDA and regression model showed the dependency of loosening on the coating thickness of the bracket.

Developed a movie recommender system using TMDB dataset. After cleaning the data, tags were created using overview, cast, genre, director After tags creation ,we used nltk library to stem the text. Using counter vectorization vectors were created of each row. Using cosine-similarity, simailarity was checked between the movies Project was deployed on heroku using pycharm

Developed a laptop price predictor using various regression algorithms like Linear Regression, Random Forest , Decision Tree, SVR, Ridge, Lasso Main feature was data analysis and feature engineering on this problem Model was trained using column transformers with various algorithms Achieved 88?curacy with random forest Model was deployed on heroku

Developed a message spam classifier using naïve bayes algorithmPerformed EDA and feature engineering on the dataset andmulticollinearity was removed.In pre-processing, tokenization, stemming was done on the text.Used TFIDF Vectorizer for vectorization having max_features =3000After checking various models, multinomial naïve bayes gaveprecision of 100 % and accuracy of 97%