Hi, I am a passionate coder with high interest in Data Science, Machine Learning, Deep Learning and AI. I constantly update my knowledge and skills by undergoing courses, blogs and solve problems in Hackathons.
Worked on the E-commerce data for automat the answer for the users query on the product, based on the product specification provided. The solution to this problem involved the usage of classical NLP, ML and Deep learning techniques
The concepts of Collaborative filtering, Matrix factorization and XGBoost are used to predict the rating that a user would give to a movie that he has not yet rated. As it is a recommendation problem, the major task is to handle the Cold Start problem and Minimize the RMSE between predicted and actual rating.
Classical NLP and ML techniques like Tf-Idf, Glove, XGBoost, etc. are used to detect whether the given pair of Quora questions are Semantically Similar or not. The KPI was Multiclass Log Loss.
The concepts of Content Based Recommendation, CNN, Word2Vec, Tf-Idf, etc. with Keras and TensorFlow are used to recommend the apparels to the users.
A multiclass classification problem to detect cancer given its Gene, Mutation, Variation, Medical literature and 9 classes. Techniques like Stacking and Majority Vote Classifiers with NB, Linear SVM, KNN and Random forest are applied. For interpretability, we predict the probability of each data point belonging to each of the 9 classes.
CGPA: 7.4