Hemant Dinkar Patil's repositories
Predictive-Marketing-Analytics
Assessed brand loyalty patterns and price elasticity metrics to provide insights for brand’s market growth. Recommended strategies for the top 3 brands based on competitor analysis, customer profiling and customer retention through RFM analysis and multinomial logit.
Data-Visualization-NBA-analysis-using-Tableau-
Performed cluster analysis and segmentation on several metrics and made several visualizations in the form of charts, graphs, storyline showing different statistics related to players and teams in National Basketball Association (NBA) to determine whether La Lakers are still the team to bet on.
Python-Market-basket-Analysis
Designed a prototype of a market basket analytics system, look at the products the customer has in their online shopping cart and recommend another product.
Python-Recommender-system-for-hospitals-based-on-Medicare-ratings-and-patient-surveys
Compared Hospital dataset and developed a SQL database to parse and load each of the datafile into database using sqlite and performed statistical analytics to determine hospital ranking.
American-Heart-Association
Developed a predictive model using R, which is used to predict the top walkers and team captains. Advised on opportunities to expand and local event strategies, to boost yearly donations by 70%.
Data-Warehouse-Design-and-Strategy
Developed a scalable business data warehousing design for a bank
Machine-Learning
Predict the survival of the passenger on the famous Titanic
Python-Optimum-Flight-Schedule
Developed an optimized flight schedule considering the different arrival, departure times, minimum ground times, number of gates at the respective airports using python libraries such as pandas, numpy etc.
Recommender-System---Pearson-Correlation-Movie-Recommender
Basic recommender system build using the Pearson correlation coefficient to recommend a-like movies taking in consideration the user ratings.
The-Simpsons
Designed and built dashboards to reflect the descriptive statistics, content of episodes, popularity, ratings and views