Summer 2020 Data Engineering Projects

The MIS Academic and Research Technologies Group (ARTG) Data Engineering Project Team worked with 13 MIS graduate students who created projects focused on data engineering, data science and visualization over the summer of 2020.

The students selected projects topics ranging from sports analytics to social justice.

We challenged these students to build from a foundation of Python, SQLite and Jupyter Notebooks; many layered on additional functionality with Tableau and git. Please see their excellent work below.

Telecom Customer Churn Analysis and Prediction

Customer attrition, also known as customer churn, customer turnover, or customer defection, is the loss of clients or customers which in turn leads to loss of revenue. The aim is to understand the factors that contribute significantly to customer attrition and predict customer behavior using predictive modelling. Once we analyze all relevant customer data we can develop focused customer retention programs.

See this project on Google Colaboratory

Project team:

Iti shri Kotiyal

Crimes in Boston Analyzed

The project is based on the "Crimes in Boston" dataset, which records the incidents in Boston from 2015 to 2018. The dataset was cleaned and analyzed to find various correlations. With the help of visualizations, we were able to deduce patterns and regions that contribute to the increasing crime rate in Boston's city.

See this project on Google Colaboratory

Project team:

Vedashree Dalvi

Stuti Khurana

Amit Kumar Saragadam

Analysis of accidents in USA

This project takes an in-depth look about the data on the accidents that occurred in the year 2016. It highlights some of the interesting aspects and draws inferences surrounding the accidents like the distribution of accidents based on cities, states, time zones, weather conditions and amenities.

See this project on Google Colaboratory

Project team:

Sai Nivedith Reddy Thuraka

US Police Crimes

Our project was inspired by the #BlackLivesMatter movement and the goal was to identify the various injustice acts such as police brutality happening in the United States. This helps identify crimes amongst the minority population as well as identify the various races which have been the victims.

See this project on Google Colaboratory

Project team:

Riya Shah

Sariska Sujaini

Fantasy Premier League Points Predictor

The aim of the project is to analyze player and team performance statistics from the English Premier League over the last few years and to predict the trends that would classify a player into various criterions defined on success. The idea is to identify the attributes that make a player successful and ultimately predicting whether a player deserves a spot on your Fantasy Football roster.

See this project on jupyter.org

Project team:

Anmol Sabharwal

Raghav Bhutra

Aashish Harneja

Subject matter expert

Sagar Patnaik

Subject matter expert

Detection of Fake Product Reviews Using NLP

This project utilizes 21,000 Labeled (Fake/ Not Fake) Amazon reviews submitted for different products to study if language elements and user information (verified purchase) can be used to verify authenticity of user reviews. Multiple supervised and unsupervised machine learning models like SVM, Naive Bayes, Decision Tree, Random Forest, Logistic Regression and Latent Dirichlet Allocation (LDA) are compared and evaluated for prediction accuracy. Finally, Gaussian Naïve Bayes model trained on user information and sentiment analysis of reviews proves to be the most promising one with 80% accuracy.

See this project on Google Colaboratory

Project team:

Pooja Tyagi

Demystifying Public Sentiments on Airline Reviews

Understanding people's emotions are essential for a company's success more than ever before since customers these days express their feelings openly in online reviews. Sentiment Analysis helps businesses understand the tone of customer reviews using NLP and ML techniques. In this project, we have performed a sentiment analysis on the Airline review tweets extracted from Twitter.

See this project on Google Colaboratory

Project team:

Madhuri Eluru

Madhvika Sehgal

Prakhar Jain

The ARTG Data Engineering Project Team

The MIS ARTG Data Engineering Project Team was responsible for providing guidance, feedback and review for students participating in this project, as well as managing communication channels, collaboration resources and providing project management. We are proud to have worked with these students on their work and hope the ideas and expertise we shared with them enhanced their experience.