keerthi-sreenivas Goto Github PK

data-science-beginners-projects

This is a repository is for posting my projects while I learn data science through practical projects and datasets. Feel free to follow along on check them out.

deep-learning

dscc202-402-spring2022

course materials for Data Science at Scale

elon_musk-doge-tweet_analysis

The crypto currency Dogecoin’s price has seen great fluctuations in april and may of 2021. Our study tries to observe the trends in twitter sentiments and the significant impact of Elon Musk’s twitter activity on Dogecoin over a series of two events. Two such event’s we analyzed is when, after Elon Musk claimed via Twitter that “SpaceX is going to put a literal Dogecoin on the literal moon” and later called himself “DogeFather” as he announced pub- licly that he would be hosting the Saturday Night Live on 8 May 2021. By doing sentiment analysis using Textblob on relevant tweets before Musk’s tweet , during the period between his tweet and his actual appearance on the show and after the show , we found that Tweet volume , twitter sentiments and Google trends collectively play as predictor variables of price direction. By using these insights, a per- son is able to make better informed purchase and selling decisions related to Dogecoin/ or other crypto currencies when they are being talked about.

hugo-profile

A highly customizable and mobile first Hugo template for personal portfolio and blog.

keerthi-sreenivas.github.io

learningsparkv2

This is the github repo for Learning Spark: Lightning-Fast Data Analytics [2nd Edition]

opengl

Basic codes in open gl

python_programming

solubility-prediction-data-analytics

statistics_for_efficient_marketing_campaign

In this project we have used a marketing campaign dataset which consists of Segments of customers, the number of orders they placed and the number of different products they bought. Our study aims to find insights between these variables and inferences for better marketing. Our approach consists of 3 hypothesis. The first hypothesis involves testing whether Number of Web Orders(Q) placed by customers of different marital status(C) is different to focus advertising on a particular group and we found that mean of web orders placed by different marital groups is same[Parametric]. We also performed a second test to see which age group consumes more amount of wine and found out that customers whose age is above 45 consume more wine than whose age is below 45[Non-Parametric]. Second hypothesis tests to see if the number of children(C) a customer has and amount of sweet products(C) he/she purchases is dependent, and we found that these two variables are actually dependent on each other. Lastly, the third hypothesis is to check if amount of wines purchased and age are linearly related. Further we do a linear regression between to predict amount of wines purchased from meat purchases, multiple regression between amount of wines purchased and kidHome and educational status. We then perform boot strapping approach on the same multiple regression equation and also an additional model which could be predictors to wines purchased. As additional work, we have included Model fitting using AIC values(forward,backward and stepwise) and some exploratory data analysis.