Learnt skills like Feature Engineering, EDA, Data Visualization and other skills like teamwork, time management and problem-solving. Got hands on experience with tools like Tableau and Scikit-learn.Worked on projects like NYC taxi trip time prediction, credit card default prediction and Netflix movies and tv-shows clustering . Gained proficiency in python programming. Learned and gained proficiency in new technologies like Tableau and NLTK.Worked as a Subject Matter Expert and solved data science related questions and queries over the Doubt Resolution forum.

Built an app that detects the sentiment of the online classroom using live video from the webcam and real-time aggregated feedback to the instructors about the class.Employed essential image preprocessing techniques such as image augmentation, Pixel brightness transformations, Histogram equalization, Region processing, and gray scaling to improve the image quality for better prediction.Build an CNN model, also tried with some transfer learning techniques.

Tags: K-Means, PCA, Clustering, Silhouette, Elbow score and Recommendation systemDeveloped Recommendation system for Movies and Tv- Shows.Performed Data Cleaning, data collection, preprocessing and analysis, Building models to address business problems by finding data patterns.Employed NLP techniques such as Text cleaning, Stemming, Tokenizer, removing Stopwords. Performed vectorization of the final text column using TF-IDF followed by dimensionality reduction using PCA.Visualizing most used words in Title and Cast with WordCloud.Evaluated the optimal clusters using silhouette and got 5 clusters, K-means performed well on the dataset and gives Silhouette score of 0.82.

Tags: Logistic Regression, Random Forrest Classifier, XGBoost, Support Vector Classifier, SMOTE, Class ImbalanceBuilt a Classification model using XGBoost to predict whether a customer will default on his/her credit card.Handled Imbalanced data using SMOTE and performed Hyperparameter Tuning. Benchmarked XGBoost algorithm against Logistic Regression, Random Forrest and Support Vector Classifier.Experimented with Performance metrics for various models found XGBoost as the best model with an Recall of 85.6%.

Tags: Linear Regression, Decision Tree, Gradient Boosting, XGB RegressorBuilt a Regression Model using Gradient Boosting, Decision tree regressor , and XGBoost models to predict taxi trip time in NYC.Gradient Boosting was benchmarked against algorithms such as Linear Regression, Decision Tree and XGB Regressor.Employed processing techniques such as Feature Engineering and Outlier Treatment.Experimented with Hyperparameter Tuning techniques such as GridSearchCV and achieved R2 Score of 73.7% using the Gradient Boosting.

The aim of this project is to Analyze the real-world Business Problem of Airbnb's BookingThis dataset has around 49,000 observations in it. To get a fair idea about the factors affecting the establishment of different types of the hosts at different places across NYC, the aggregate rating of each hosts, and many more.Explored and analyze the data to discover the key understandings.