Market Expansion using Data Science

• Project Category: Business Use-case

In this project, I have assumed a business use case for a hypothetical sports retail company. The company wants to expand its operations and setup its first retail store in Toronto, Ontario. The task is to identify locations in Toronto that are most suitable for the company to setup their first retail store.

For the same, I have performed analysis based on location data gathered through web scraping and using the Foursquare API. Finally, using clustering algorithms, I have identified 4 prime locations in which the company can setup their first retail store.

GitHub Repository | Kaggle Notebook

KPMG Data Analytics Consulting Virtual Internship

• Project Category: Business Use-case

This project is a collection of the work performed as part of the Data Analytics Consulting Virtual Internship offered by KPMG through InsideSherpa (now Forage). It involves data quality assessment, data preparation for analysis, data analysis and insights presentation.

The project involves data preparation done using Tableau Prep, as well as implementation of a linear regression model to predict the profits that the company may earn by targeting new customers, as identified through the analysis of provided and prepared data. Finally, insights are presented through a data visualisation dashboard created using Tableau Desktop.

GitHub Repository | Tableau Public

Tweety - # Scraper for Twitter

• Project Category: Application

Tweety is a Django-based web application that can be used to create a tweets dataset based on hashtags. With the help of snscrape, Tweety scrapes Twitter for tweets that contain the hashtag given by the user. The result is a dataset of all tweets containing that hashtag, as well as a dataset of all users who posted those tweets. The dataset can be downloaded as a .zip file and used in Data Science and NLP projects.

This project is currently under development. Stay tuned.

Spotify EDA & Recommendation System

• Project Category: Recommendation System

In this project, I have developed a Content-based recommendation System for Spotify. For the same, I have used a dataset of 600,000 tracks and 1.1M artists on Spotify, available on Kaggle. Based on the song given, the recommendation system recommends more songs as well as artists that the user might also like.

The recommendation system uses information describing each song/track, such as its audio characteristics (instrumentalness, tempo, etc.) and genre information, to determine other similar songs. The same is done for artists as well. Using the Vector Space Model, each song and artist is represented as a vector. Similarity is then determined using cosine similarity measure. The most similar songs and artists are recommended.

Application Website | GitHub Repository | Kaggle Notebook

Lionel Messi Career Analysis - EDA & Machine Learning

• Project Category: Sports Analytics

Through this Data Science project, I have analysed a number of football datasets available to get a deeper insight into Lionel Messi's football career at FC Barcelona.

This project consists of an Exploratory Data Analysis of football events dataset and other football related datasets to analyse the football career of Lionel Messi at FC Barcelona based on various crucial statistics. The project also involves a logistic regression model to predict expected goals for players. Finally, the project also aims to explore future prospects for Lionel Messi.

GitHub Repository | Kaggle Notebook

#FarmersProtest Tweets Sentiment Analysis

• Project Category: Natural Language Processing

In this project, I have performed sentiment analysis based on tweets related to the ongoing farmers' protest in India. This tweets dataset was collected by scraping Twitter for tweets that contain the hashtag "#FarmersProtest", using the Python library snscrape. The dataset contains nearly 850,000 tweets from 150,000 Twitter users.

This project is carried out in 2 parts. In the first part, I have performed some data pre-processing and transformation on the raw data. The first part also includes an Exploratory Data Analysis for the available data. In the second part of the project, I have performed sentiment analysis for the tweets.

Kaggle Notebook (Part 1)

Experience

Education

Skills & Tools

Python Programming

Machine Learning

Apache Airflow

PostgreSQL

Tableau

Git

Microsoft Excel

Data Analysis

Probability & Statistics

Projects

Projects



• Market Expansion using Data Science
  Data Science project involving data extraction and wrangling, as well as analysis of location data to provide suggestions and solutions for an assumed business use-case.
See project


---

• KPMG Data Analytics Consulting Virtual Internship
  Data analysis, modelling and insight presentation through dashboards done as part of the Data Analytics Consulting Virtual Internship provided by KPMG through InsideSherpa (now Forage).
See project


---

• Tweety - # Scraper for Twitter
  Django-based web application to generate tweets dataset based on trending hashtags provided by user.
This project is currently under development.


---

• Spotify EDA & Recommendation System
  Content-based Recommendation System for Spotify, developed to recommend more songs and artists that a user might like based on given song.
See project


---

• Lionel Messi Career Analysis - EDA & Machine Learning
  EDA and Machine Learning project to analyse the football career of Lionel Messi based on various crucial statistics.
See project


---

• #FarmersProtest Tweets Sentiment Analysis
  Sentiment analysis performed for tweets available in the tweets dataset collected based on the trending hashtag "#FarmersProtest", available on Kaggle.
See project


---

• Regional Language Weather Chatbot
  Flask-based weather chatbot application developed to address users' weather related queries in Gujarati language (one of the many regional languages of India).
See project


---

• PhotoTrans
  Flask-based application developed to extract text in any language from images of objects, such as road signs or warnings, and translate the extracted text into English.
See project


---

• Digits Recogniser
  Machine Learning project to identify hand-drawn digits from 0 to 9 based on data about pixel values for the image of each hand-drawn digit.
See project


---





Publications



• Word Embedding Based Answer Evaluation System for Online Assessments
  Research paper published in the International Journal of Scientific and Engineering Research (ISSN 2229-5518) discussing a word embedding based approach to automatically evaluate answers submitted in online examinations.
See publication


---

Certifications

• IBM Data Science Professional Certificate
  Issued by Coursera, authorized by IBM (June 2021)
See credential


---

• Tableau Data Scientist
  Issued by Tableau Software (July 2020)
See credential


---

• Tableau Analyst
  Issued by Tableau Software (July 2020)
See credential


---

• Tableau Desktop Specialist
  Issued by Tableau Software (June 2020)
See credential


---

• Virtual Experience Program Participant
  Issued by Forage (May 2020)
See credential


---

Achievements

Test Scores



• Graduate Record Examinations (GRE)
  Verbal Reasoning: 160/170
  Quantitative Reasoning: 168/170
  Analytical Writing Assessment: 4.5/6.0
  
  Test date: 27 July, 2021



---

• International English Language Testing System (IELTS)
  Overall band score: 8.0
  Listening: 9.0
  Reading: 8.5
  Writing: 7.5
  Speaking: 7.0
  
  Test date: 31 July, 2021



---





Sports



• Winner, Regional Level Subroto Football Tournament, Ahmedabad, Gujarat, India (2016)



---

• Represented Gujarat state at International Subroto Football Tournament, New Delhi, India (2016)



---





Competitions



• Represented Navrachana International School, Vadodara at National Level CBSE Science Exhibition, New Delhi, India (2011)



---