Experience

GenZ Technologies

• United States
• Information Technology & Services
• 1 - 100 Employee

• Junior Data Scientist

  • Jul 2023 - Present

  Hyderabad, Telangana, India



Edureka

• India
• E-Learning Providers
• 700 & Above Employee

• Data science Trainer

  • Oct 2022 - Jul 2023

  Hyderabad, Telangana, India



AlmaBetter

• India
• Higher Education
• 700 & Above Employee

• Data Scientist

  • Jan 2022 - Dec 2022

  Bengaluru, Karnataka, India

• Online Retail Customer Segmentation

  • Apr 2022 - May 2022

  Tags: K-Means, RFM, Elbow method, Silhouette score, Hierarchical Clustering, Dendrogram, t-SNE, DBSCAN. The UK- based Online Retail company wants to Build customer segmentation Based on different factors to customize to Target them. Preparing the data to build models and then Evaluation of the model like K-means clustering, DBSCAN, Hierarchical _clustering with and without silhouette_score, Elbow method. when we cluster them by Recency , Frequency and Monetary . K-Means… Show more Tags: K-Means, RFM, Elbow method, Silhouette score, Hierarchical Clustering, Dendrogram, t-SNE, DBSCAN. The UK- based Online Retail company wants to Build customer segmentation Based on different factors to customize to Target them. Preparing the data to build models and then Evaluation of the model like K-means clustering, DBSCAN, Hierarchical _clustering with and without silhouette_score, Elbow method. when we cluster them by Recency , Frequency and Monetary . K-Means with silhouette_score, Hierarchical clustering with Elbow method we get the optimal segments Two . This segment two Allows us to understand our customers better way help us target these customers more efficiently and improve the customer experience.

• Cardiovascular Risk prediction

  • Mar 2022 - Apr 2022

  Tags: Classification, SMOTE, XG Boost , Grid search CV, Feature Engineering, Pearson Correlation, Logistic Regression, SVC, K-NN, Naive Bayes Classifier,Accuracy,Precision,Recall,F-1 score. Predict the Risk a patient of coronary heart disease (CHD).To Build Preventive methods against CHD for future prediction of CHD. Model building and then Evaluation of the model Logistic Regression, Random Forest, Naive Bayes, Support vector machine, Adaboost, KNN, Decision Tree with different… Show more Tags: Classification, SMOTE, XG Boost , Grid search CV, Feature Engineering, Pearson Correlation, Logistic Regression, SVC, K-NN, Naive Bayes Classifier,Accuracy,Precision,Recall,F-1 score. Predict the Risk a patient of coronary heart disease (CHD).To Build Preventive methods against CHD for future prediction of CHD. Model building and then Evaluation of the model Logistic Regression, Random Forest, Naive Bayes, Support vector machine, Adaboost, KNN, Decision Tree with different evaluation metrics. We've got the model accuracy in the range of 0.82 to 0.88. So, the accuracy of our best model is 0.845 is 84.5% which is by Naive Bayes so, Naive Bayes is our optimal model. This model was Implemented finally to prevent People and patients from this 10-year CHD using the ML model. Doctors and scientists easily conclude which features are more affecting CHD.

• Bike Sharing Demand Prediction

  • Feb 2022 - Mar 2022

  Tags: Linear ,Lasso, Ridge Regressions, Random Forest, Gradient Boost, XG Boost, EDA,, Multicollinearity, Grid Search Optimization, R2-Score, Adj-R2- Score, MSE, RMSE,MAE A stable supply of rental bikes becomes a major concern and a count is required at each hour for the stable supply of rental bikes with optimal prices. Arrange the Bike for the public at the right time as it less the waiting time. Counting of bikes for each hour Building Models and evaluating models like… Show more Tags: Linear ,Lasso, Ridge Regressions, Random Forest, Gradient Boost, XG Boost, EDA,, Multicollinearity, Grid Search Optimization, R2-Score, Adj-R2- Score, MSE, RMSE,MAE A stable supply of rental bikes becomes a major concern and a count is required at each hour for the stable supply of rental bikes with optimal prices. Arrange the Bike for the public at the right time as it less the waiting time. Counting of bikes for each hour Building Models and evaluating models like Linear regression, regularization techniques like LASSO, RIDGE, and Random forest regressor, Gradient boost regressor, XG boost Regressor, and CAT boost regressor. Best predictions are obtained with a random forest model with an accuracy of 0.875 is 87%. Best predictions are more effective on High bike availability with lower prices as per seasons to get the Best business outcome To Improve The Customer Experience.

• Airbnb Data Analysis

  • Jan 2022 - Feb 2022

  Tags: Private room, Brooklyn, Manhattan, Central Park, beautiful, spacious, subway, Williamsburg, Bushwick, times square, Scatter plot, Boxplot,Violin plot Airbnb is one of the top Hotel Booking wants to analyse the Data of Newyork city to understand the demand and supply of Different listings. To analyze this Data we plot the Bar plot, Pie Chat, Boxplot, Violin plot, Strip plot, Violin plot,Scatter plot. Private Room, followed by Entire type, the Shared room is the lowest… Show more Tags: Private room, Brooklyn, Manhattan, Central Park, beautiful, spacious, subway, Williamsburg, Bushwick, times square, Scatter plot, Boxplot,Violin plot Airbnb is one of the top Hotel Booking wants to analyse the Data of Newyork city to understand the demand and supply of Different listings. To analyze this Data we plot the Bar plot, Pie Chat, Boxplot, Violin plot, Strip plot, Violin plot,Scatter plot. Private Room, followed by Entire type, the Shared room is the lowest bookings. Manhattan has more bookings of Entire room compared to Private room. Other boroughs have more bookings of Private rooms compared to the Entire room. Entire Homes are the most expensive, followed by Private, Shared rooms, Manhattan is the most expensive borough for all types of rooms. Now you can make wiser decisions to make your profit or to save your money Provide the Best Customer Experience.