Tanmay Gupta

• Visualized 10 years of natural disaster data that occurred in USA between 2004-2015 identifying most prevalent disaster in Texas and Alabama
• Chose Chevron as energy industry to detect different disaster types affecting the refinery in various states
• Created calculated fields to detect top 4 disasters impacting corporation in energy industry with Hurricane ranked first

• Performed a project simulation for a computer peripheral manufacturer company in order to design a new innovative printer.
• Effectively ran 4 different scenarios based on unique scopes, schedule, budgeting, cost overview, schedule, team process and resources.
• Implemented various target decisions based on the project completion requirements each week like skill level, outsourcing and prototype to estimate the total simulation score at the end of each scenario.
• Identified the… Show more

• Performed a project simulation for a computer peripheral manufacturer company in order to design a new innovative printer.
• Effectively ran 4 different scenarios based on unique scopes, schedule, budgeting, cost overview, schedule, team process and resources.
• Implemented various target decisions based on the project completion requirements each week like skill level, outsourcing and prototype to estimate the total simulation score at the end of each scenario.
• Identified the processes like team stress, over budget and schedule lagging problems and re-run the different scenarios for best score results and continuous improvement.
Show less

• Performed data retrieval and data preprocessing for an extracted dataset in R by replacing the missing values of the attributes
• Imported libraries like Rpart, Rattle, Rserve for decision tree and k-means clustering to identify the correct number of clusters
• Created parameters and calculated fields for variables like age, gender, siblings, passenger to visualize across different clusters
• Identified top 2 clusters having the highest survivability rate for each unique variable by… Show more

• Performed data retrieval and data preprocessing for an extracted dataset in R by replacing the missing values of the attributes
• Imported libraries like Rpart, Rattle, Rserve for decision tree and k-means clustering to identify the correct number of clusters
• Created parameters and calculated fields for variables like age, gender, siblings, passenger to visualize across different clusters
• Identified top 2 clusters having the highest survivability rate for each unique variable by plotting bar charts
Show less

• Extracted data from 1961-2017 and performed analysis on vanilla production worldwide, identified insights and trends of Area Harvested vs Yield within the data data flow data integrity
• Identified the top production area to be Madagascar, but officially its Indonesia as per the data, the reason being harvesting which lasts from May till August in Madagascar after which rainy season brings cyclones
• Performed visualization on bean price summary with actual price and used forecast… Show more

• Extracted data from 1961-2017 and performed analysis on vanilla production worldwide, identified insights and trends of Area Harvested vs Yield within the data data flow data integrity
• Identified the top production area to be Madagascar, but officially its Indonesia as per the data, the reason being harvesting which lasts from May till August in Madagascar after which rainy season brings cyclones
• Performed visualization on bean price summary with actual price and used forecast indicator to predict estimated price of crop
Show less

• Defined a workforce measure to identify sum of populations of states for ad-hoc data exploration , KPIs using Power BI
• Analyzed how population of cities and workforce of people change as we present animated visualization of trend in the population work-force from 1790 to 1990 in the US with New York having the highest population of 8Mdata

• Collected and analyzed the weather data of Texas on a day along with climate situations and wind speed using Python script
• Designed a python code in Jupyter notebook via BeautifulSoup to translate web-scrape data into a csv file
• Constructed graphical visualization to correlate the average and normal temperature of different locations using MS Excel

• Efficiently predicted the future churn of 20k employees based on 10 major attributes using the data sets of an IT company
• Performed exploratory analysis for analyzing different patterns, outliers, trends and using Pandas, Matplotlib, Seaborn for descriptive statistics and visualization
• Executed cluster analysis to identify the churn by grouping into the overall satisfaction and performance of the company
• Built and evaluated model performance using Gradient Boosting model which… Show more

• Efficiently predicted the future churn of 20k employees based on 10 major attributes using the data sets of an IT company
• Performed exploratory analysis for analyzing different patterns, outliers, trends and using Pandas, Matplotlib, Seaborn for descriptive statistics and visualization
• Executed cluster analysis to identify the churn by grouping into the overall satisfaction and performance of the company
• Built and evaluated model performance using Gradient Boosting model which identified a prediction accuracy rate of 92% Show less

• Extracted and merged the daily prices of bitcoin from 2014 to 2018 from St. Louis Fred data containing S&P 500, London market price for gold, the US/Euro exchange rate and West Texas intermediates spot price of oil
• Performed time series analysis and forecasted the trend for next 30 days using ARIMA modelling
• Determined the interrelation between the stocks using Granger causality tests and interpret them
• Worked on designing attractive graphical and visualization UI to deploy… Show more

• Extracted and merged the daily prices of bitcoin from 2014 to 2018 from St. Louis Fred data containing S&P 500, London market price for gold, the US/Euro exchange rate and West Texas intermediates spot price of oil
• Performed time series analysis and forecasted the trend for next 30 days using ARIMA modelling
• Determined the interrelation between the stocks using Granger causality tests and interpret them
• Worked on designing attractive graphical and visualization UI to deploy time series models
Show less

A web application was developed which emphasized on providing a platform in the form of a website where people were advised by in order to lead a healthy lifestyle.
Website Designing titled as “Diet Doyen” (Medical Home Dietitian)
Connecting to local host server using Client Side Scripting, ADO. Connectivity, Console.
Technologies: ASP.NET & HTML were used.