Projects
As a Data Scientist, I have worked on various projects across different domains. I have analyzed data, identified insights, and recommended solutions. Here are some of the projects I have worked on:
JARVIS: The Data Mining Robot
JARVIS: The Data Mining Robot is a virtual robot built for scraping the Corporate Action data which is available on the web. Though this data is available free of cost on the web there are some data vendors who charge high because it's difficult to find the right data for the right company in the right format. That is where Jarvis comes into action, it surfs the web based on the user requirements, finds the right data, and then scraps the data which is delivered to the user. Jarvis is not only a scraper but also an assistant, it assists its user with the stock prices of the target companies based on the Corporate Action. As we know Corporate Actions have a significant impact on the stock prices of the companies, making the stock price a dependent variable for the Corporate Actions enabling Jarvis to perform predictive actions based on the user requirements to deliver prominent results.
Semi Automated Machine Learning
Machine learning techniques have deeply rooted in our everyday life. However, since it is knowledge- and labor-intensive to go pursue good learning performance, humans are heavily involved in every aspect of machine learning. To make machine learning techniques easier to apply and reduce the demand for experienced human experts, application of automated machine learning has emerged as a hot topic with both industrial and academic interest. In this project, we try to implement some of the most common machine learning algorithms which actually perform decent on most kinds of data. This is a platform where even amateurs can utilise the power of machine learning algorithms to obtain quick regression and classification results. In simple words we can call it as a plug and play platform for simple machine learning problems
Review Classification
With the flourishing of the web, online review is becoming a more and more useful and important information resource for people. As a result, automatic review mining and summarizing have become a hot research topic recently. “Text Classification” is one of the most important NLP tasks. It is the process of classifying text strings or documents into different categories, based upon the contents of the strings. Some examples of text classification are Understanding audience sentiment from social media, Detection of spam and non-spam emails, Auto-tagging of customer queries, and Classifying blog posts into different categories. This project focuses on a specific domain – Movie Reviews. A multi-knowledge-based approach is used, which integrates statistical analysis and movie knowledge. The dataset used in this project is Movie Review Dataset, which is retrieved from the IMDB movie reviews dataset. The dataset is comprised of thousands of positive and negative movie reviews. The dataset is cleaned up and pre-processed in several ways and all reviews in the dataset are converted to English language. The experimental results show the effectiveness of the multi-knowledge-based approach in movie review mining and summarizing. Thus enabling the audience to have a clear and summarized end review of the movie.
Supermarket Billing System
Engineered a robust software solution utilizing JAVA and SQL Server, revolutionizing the billing process in supermarkets; achieved significant time savings by generating bills quickly, while seamlessly storing details in a centralized database.
Dynamic Pricing Using Uber and Lyft Cab Prices
Guided a team project focused on creating a machine learning model to predict real-time fares for ride-sharing services, utilizing historical data from Uber and Lyft and applying a combination of regression and ensemble methods for improved precision.
Kaggle Machine Learning and Data Science Survey Analysis
Led the development and execution of a comprehensive data science survey analysis, uncovering impactful insights on demographics, education, and industry trends from 2018 to 2021; informed strategic initiatives that drove revenue growth, optimized marketing campaigns, and improved customer satisfaction.
E-Commerce Platform Database Schema and Implementation
Engineered a highly efficient e-commerce platform, utilizing advanced SQL techniques to optimize schema for 50+ product categories; implemented stored procedures for automated data manipulation, resulting in dynamic management of a 100+ item product catalog and improved operational efficiency by 40%.
E-Commerce Customer Segmentation and Recommendation
Implemented customer segmentation using data science techniques to analyze buying patterns, resulting in tailored product recommendations for each segment. This strategic approach contributed to an increase in revenue.
Brazilian E-Commerce Data Analytics and Reporting System
Led the successful development of a MySQL-powered data analytics platform, expertly handling a vast dataset comprising more than 10,000 data points. Utilized advanced SQL queries and Key Performance Indicators to extract valuable insights, enabling data-driven decision-making and facilitating business growth.