top of page

                             Projects

Get ready to be blown by my Project Session! It's a carefully curated selection of projects that showcase my creativity, technical skills, and passion for innovation.

Each project is a testament to my dedication, crafted with meticulous attention to detail and a deep user experience.

My portfolio spans diverse disciplines, from data science and machine learning, NLP to Generative AI to demonstrating my versatility as a professional. With a relentless focus on innovative solutions and a user-centric approach, my projects are designed to make a lasting impact and deliver meaningful experiences. So, what are you waiting for?

Dive into my portfolio and witness the culmination of my expertise and commitment to excellence in every endeavor!

Enhancing Search Engine Relevance for Video Subtitles

Welcome to the Enhancing Search Engine Relevance for Video Subtitles project!

 

This project is dedicated to refining the search experience by improving the relevance of video subtitles. In the fast-evolving landscape of digital content, effective search engines are crucial for connecting users with relevant information. This project focuses on developing an advanced search engine algorithm to efficiently retrieve subtitles based on user queries, leveraging natural language processing and machine learning techniques to enhance the relevance and accuracy of search results.

 

Skills Used:

  • Data Preprocessing: Reading and cleaning subtitle documents, decoding files, and managing data subsets.

  • Vectorization: Utilizing BOW/TFIDF for keyword-based representation and BERT-based “SentenceTransformers” for semantic embeddings.

  • Document Chunking: Dividing large documents into manageable chunks with overlapping windows to preserve context.

  • Database Management: Storing embeddings in a ChromaDB database for efficient retrieval.

  • Similarity Calculation: Using cosine distance to measure similarity between query and document embeddings.

  • Natural Language Processing (NLP): Enhancing search relevance through advanced text representation techniques.

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

Sentiment Analysis Project: TripAdvisor Hotel Reviews

Welcome to the Sentiment Analysis Project focused on TripAdvisor hotel reviews!

 

This project aims to understand customer sentiments—whether positive, negative, or neutral—regarding hotel experiences. By analyzing these sentiments, we gain valuable insights into customer satisfaction and dissatisfaction, allowing for the identification of pain points and features that contribute to overall experiences.

​

The project involves several key steps:

  1. Data Preprocessing: Text cleaning to remove special characters, punctuation, and stopwords, and text normalization through lemmatization to standardize text.

  2. Data Visualization: Utilizing Matplotlib and the wordcloud library to visualize sentiment trends and key insights such as positive sentiment dominance, room quality enhancement, customized dining experiences, and efficient operations.

  3. MLFlow: Using MLFlow for experiment tracking and model registry. This includes creating experiments, defining naming conventions, logging parameters, metrics, and artifacts, and comparing results to iterate and optimize models. Key metrics analyzed include accuracy, efficiency, scalability, and interpretability across models like Naive Bayes, Decision Tree, Logistic Regression, and Random Forest, with Logistic Regression identified as the best model.

  4. Deploying a Flask Web Application on AWS Cloud: Setting up an AWS account and resources, preparing the Flask application, configuring the EC2 instance, and deploying the application. This involves defining HTML code structure for the web application, including metadata, styling, and functional routes for predictions.​

Skills Used:

  • Data Preprocessing: Text cleaning and normalization techniques

  • Data Visualization: Matplotlib, wordcloud

  • Model Evaluation: MLFlow for tracking experiments, comparing models (Naive Bayes, Decision Tree, Logistic Regression, Random Forest)

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

GenAI App - Interactive Chatbot for Data Science Interviews

Welcome to the GenAI App project! This innovative application is designed to assist aspiring Data Scientists in preparing for interviews through an interactive chatbot powered by OpenAI. Addressing common challenges such as finding relevant content and overcoming confusion, the GenAI App provides an efficient and user-friendly solution for interview preparation.

 

Key Features:

  • User-Friendly Interface: A streamlined interface to facilitate the interview preparation process.

  • Interactive Chatbot: An AI-driven chatbot that offers comprehensive answers to Data Science interview questions.

  • Content Relevance: Ensures users receive up-to-date and pertinent information tailored to their needs.

  • Efficient Preparation: Helps users prepare quickly and effectively, reducing confusion and frustration.

How It Works: Users input their interview questions or topics of interest, and the chatbot uses OpenAI's advanced capabilities to generate detailed responses. It provides step-by-step guidance on approaching questions, along with tips and best practices. Interactive conversations allow users to clarify doubts, explore concepts, and deepen their understanding of Data Science topics.

Benefits:

  • Time-Saving: Instant access to relevant interview material saves time.

  • Confidence Building: Clear guidance and explanations boost interview confidence.

  • Personalized Assistance: Adapts to users' preferences and learning styles for effective preparation.

  • Accessibility: Available anytime, anywhere, ensuring users have access to valuable resources whenever needed.

Skills Used:

  • AI and Machine Learning: Leveraging OpenAI's capabilities

  • Chatbot Development: Designing and implementing an interactive chatbot

  • User Experience (UX) Design: Creating a user-friendly interface

  • Natural Language Processing (NLP): Generating relevant and accurate responses

  • Software Development: Integrating and deploying the chatbot application

GenAI App - AI Code Reviewer App

Welcome to the GenAI App project, an AI-powered code analyzer designed to assist coders, software developers, and data scientists in identifying and fixing bugs efficiently. By providing detailed feedback on code submissions, this application aims to streamline the debugging process and offer alternative, more efficient solutions.

User Interface: The app features a user-friendly interface built with Streamlit, allowing users to easily input their Python code for analysis.

​

Code Analysis Output:

  • Mistakes and Bug Descriptions: The AI identifies mistakes, bugs, and areas for improvement in the submitted code, providing detailed explanations of the issues.

  • Corrected Code Snippets: For each identified issue, the app presents corrected code snippets, demonstrating how to resolve the problems.

  • Alternative Solutions: The analyzer suggests alternative approaches to solving code problems, emphasizing more efficient coding practices.

Benefits:

  • Efficiency: Automates the code review process, saving users valuable time and effort in debugging.

  • Learning Tool: Offers detailed feedback and suggested solutions to help users enhance their coding skills and understanding of Python best practices.

  • Streamlined Workflow: Provides a seamless and efficient workflow for code review and debugging tasks through its intuitive interface and AI-powered analysis.

Skills Used:

  • AI and Machine Learning: Leveraging AI for code analysis and feedback

  • Web Development: Building a user-friendly interface using Streamlit

  • Software Debugging: Identifying and correcting code issues

  • Alternative Solutions: Suggesting efficient coding practices and solutions

​

Enhancing Search Engine Relevance for Video Subtitles

Welcome to the Enhancing Search Engine Relevance for Video Subtitles project!

 

This project is dedicated to refining the search experience by improving the relevance of video subtitles. In the fast-evolving landscape of digital content, effective search engines are crucial for connecting users with relevant information. This project focuses on developing an advanced search engine algorithm to efficiently retrieve subtitles based on user queries, leveraging natural language processing and machine learning techniques to enhance the relevance and accuracy of search results.

 

Skills Used:

  • Data Preprocessing: Reading and cleaning subtitle documents, decoding files, and managing data subsets.

  • Vectorization: Utilizing BOW/TFIDF for keyword-based representation and BERT-based “SentenceTransformers” for semantic embeddings.

  • Document Chunking: Dividing large documents into manageable chunks with overlapping windows to preserve context.

  • Database Management: Storing embeddings in a ChromaDB database for efficient retrieval.

  • Similarity Calculation: Using cosine distance to measure similarity between query and document embeddings.

  • Natural Language Processing (NLP): Enhancing search relevance through advanced text representation techniques.

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

Sentiment Analysis Project: TripAdvisor Hotel Reviews

Welcome to the Sentiment Analysis Project focused on TripAdvisor hotel reviews!

 

This project aims to understand customer sentiments—whether positive, negative, or neutral—regarding hotel experiences. By analyzing these sentiments, we gain valuable insights into customer satisfaction and dissatisfaction, allowing for the identification of pain points and features that contribute to overall experiences.

​

The project involves several key steps:

  1. Data Preprocessing: Text cleaning to remove special characters, punctuation, and stopwords, and text normalization through lemmatization to standardize text.

  2. Data Visualization: Utilizing Matplotlib and the wordcloud library to visualize sentiment trends and key insights such as positive sentiment dominance, room quality enhancement, customized dining experiences, and efficient operations.

  3. MLFlow: Using MLFlow for experiment tracking and model registry. This includes creating experiments, defining naming conventions, logging parameters, metrics, and artifacts, and comparing results to iterate and optimize models. Key metrics analyzed include accuracy, efficiency, scalability, and interpretability across models like Naive Bayes, Decision Tree, Logistic Regression, and Random Forest, with Logistic Regression identified as the best model.

  4. Deploying a Flask Web Application on AWS Cloud: Setting up an AWS account and resources, preparing the Flask application, configuring the EC2 instance, and deploying the application. This involves defining HTML code structure for the web application, including metadata, styling, and functional routes for predictions.​

Skills Used:

  • Data Preprocessing: Text cleaning and normalization techniques

  • Data Visualization: Matplotlib, wordcloud

  • Model Evaluation: MLFlow for tracking experiments, comparing models (Naive Bayes, Decision Tree, Logistic Regression, Random Forest)

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

GenAI App - Interactive Chatbot for Data Science Interviews

Welcome to the GenAI App project! This innovative application is designed to assist aspiring Data Scientists in preparing for interviews through an interactive chatbot powered by OpenAI. Addressing common challenges such as finding relevant content and overcoming confusion, the GenAI App provides an efficient and user-friendly solution for interview preparation.

 

Key Features:

  • User-Friendly Interface: A streamlined interface to facilitate the interview preparation process.

  • Interactive Chatbot: An AI-driven chatbot that offers comprehensive answers to Data Science interview questions.

  • Content Relevance: Ensures users receive up-to-date and pertinent information tailored to their needs.

  • Efficient Preparation: Helps users prepare quickly and effectively, reducing confusion and frustration.

How It Works: Users input their interview questions or topics of interest, and the chatbot uses OpenAI's advanced capabilities to generate detailed responses. It provides step-by-step guidance on approaching questions, along with tips and best practices. Interactive conversations allow users to clarify doubts, explore concepts, and deepen their understanding of Data Science topics.

Benefits:

  • Time-Saving: Instant access to relevant interview material saves time.

  • Confidence Building: Clear guidance and explanations boost interview confidence.

  • Personalized Assistance: Adapts to users' preferences and learning styles for effective preparation.

  • Accessibility: Available anytime, anywhere, ensuring users have access to valuable resources whenever needed.

Skills Used:

  • AI and Machine Learning: Leveraging OpenAI's capabilities

  • Chatbot Development: Designing and implementing an interactive chatbot

  • User Experience (UX) Design: Creating a user-friendly interface

  • Natural Language Processing (NLP): Generating relevant and accurate responses

  • Software Development: Integrating and deploying the chatbot application

GenAI App - AI Code Reviewer App

Welcome to the GenAI App project, an AI-powered code analyzer designed to assist coders, software developers, and data scientists in identifying and fixing bugs efficiently. By providing detailed feedback on code submissions, this application aims to streamline the debugging process and offer alternative, more efficient solutions.

User Interface: The app features a user-friendly interface built with Streamlit, allowing users to easily input their Python code for analysis.

​

Code Analysis Output:

  • Mistakes and Bug Descriptions: The AI identifies mistakes, bugs, and areas for improvement in the submitted code, providing detailed explanations of the issues.

  • Corrected Code Snippets: For each identified issue, the app presents corrected code snippets, demonstrating how to resolve the problems.

  • Alternative Solutions: The analyzer suggests alternative approaches to solving code problems, emphasizing more efficient coding practices.

Benefits:

  • Efficiency: Automates the code review process, saving users valuable time and effort in debugging.

  • Learning Tool: Offers detailed feedback and suggested solutions to help users enhance their coding skills and understanding of Python best practices.

  • Streamlined Workflow: Provides a seamless and efficient workflow for code review and debugging tasks through its intuitive interface and AI-powered analysis.

Skills Used:

  • AI and Machine Learning: Leveraging AI for code analysis and feedback

  • Web Development: Building a user-friendly interface using Streamlit

  • Software Debugging: Identifying and correcting code issues

  • Alternative Solutions: Suggesting efficient coding practices and solutions

​

Enhancing Search Engine Relevance for Video Subtitles

Welcome to the Enhancing Search Engine Relevance for Video Subtitles project!

 

This project is dedicated to refining the search experience by improving the relevance of video subtitles. In the fast-evolving landscape of digital content, effective search engines are crucial for connecting users with relevant information. This project focuses on developing an advanced search engine algorithm to efficiently retrieve subtitles based on user queries, leveraging natural language processing and machine learning techniques to enhance the relevance and accuracy of search results.

 

Skills Used:

  • Data Preprocessing: Reading and cleaning subtitle documents, decoding files, and managing data subsets.

  • Vectorization: Utilizing BOW/TFIDF for keyword-based representation and BERT-based “SentenceTransformers” for semantic embeddings.

  • Document Chunking: Dividing large documents into manageable chunks with overlapping windows to preserve context.

  • Database Management: Storing embeddings in a ChromaDB database for efficient retrieval.

  • Similarity Calculation: Using cosine distance to measure similarity between query and document embeddings.

  • Natural Language Processing (NLP): Enhancing search relevance through advanced text representation techniques.

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

Sentiment Analysis Project: TripAdvisor Hotel Reviews

Welcome to the Sentiment Analysis Project focused on TripAdvisor hotel reviews!

 

This project aims to understand customer sentiments—whether positive, negative, or neutral—regarding hotel experiences. By analyzing these sentiments, we gain valuable insights into customer satisfaction and dissatisfaction, allowing for the identification of pain points and features that contribute to overall experiences.

​

The project involves several key steps:

  1. Data Preprocessing: Text cleaning to remove special characters, punctuation, and stopwords, and text normalization through lemmatization to standardize text.

  2. Data Visualization: Utilizing Matplotlib and the wordcloud library to visualize sentiment trends and key insights such as positive sentiment dominance, room quality enhancement, customized dining experiences, and efficient operations.

  3. MLFlow: Using MLFlow for experiment tracking and model registry. This includes creating experiments, defining naming conventions, logging parameters, metrics, and artifacts, and comparing results to iterate and optimize models. Key metrics analyzed include accuracy, efficiency, scalability, and interpretability across models like Naive Bayes, Decision Tree, Logistic Regression, and Random Forest, with Logistic Regression identified as the best model.

  4. Deploying a Flask Web Application on AWS Cloud: Setting up an AWS account and resources, preparing the Flask application, configuring the EC2 instance, and deploying the application. This involves defining HTML code structure for the web application, including metadata, styling, and functional routes for predictions.​

Skills Used:

  • Data Preprocessing: Text cleaning and normalization techniques

  • Data Visualization: Matplotlib, wordcloud

  • Model Evaluation: MLFlow for tracking experiments, comparing models (Naive Bayes, Decision Tree, Logistic Regression, Random Forest)

  • Web Application Deployment: Flask, AWS EC2 setup, HTML/CSS for web design

GenAI App - Interactive Chatbot for Data Science Interviews

Welcome to the GenAI App project! This innovative application is designed to assist aspiring Data Scientists in preparing for interviews through an interactive chatbot powered by OpenAI. Addressing common challenges such as finding relevant content and overcoming confusion, the GenAI App provides an efficient and user-friendly solution for interview preparation.

 

Key Features:

  • User-Friendly Interface: A streamlined interface to facilitate the interview preparation process.

  • Interactive Chatbot: An AI-driven chatbot that offers comprehensive answers to Data Science interview questions.

  • Content Relevance: Ensures users receive up-to-date and pertinent information tailored to their needs.

  • Efficient Preparation: Helps users prepare quickly and effectively, reducing confusion and frustration.

How It Works: Users input their interview questions or topics of interest, and the chatbot uses OpenAI's advanced capabilities to generate detailed responses. It provides step-by-step guidance on approaching questions, along with tips and best practices. Interactive conversations allow users to clarify doubts, explore concepts, and deepen their understanding of Data Science topics.

Benefits:

  • Time-Saving: Instant access to relevant interview material saves time.

  • Confidence Building: Clear guidance and explanations boost interview confidence.

  • Personalized Assistance: Adapts to users' preferences and learning styles for effective preparation.

  • Accessibility: Available anytime, anywhere, ensuring users have access to valuable resources whenever needed.

Skills Used:

  • AI and Machine Learning: Leveraging OpenAI's capabilities

  • Chatbot Development: Designing and implementing an interactive chatbot

  • User Experience (UX) Design: Creating a user-friendly interface

  • Natural Language Processing (NLP): Generating relevant and accurate responses

  • Software Development: Integrating and deploying the chatbot application

GenAI App - AI Code Reviewer App

Welcome to the GenAI App project, an AI-powered code analyzer designed to assist coders, software developers, and data scientists in identifying and fixing bugs efficiently. By providing detailed feedback on code submissions, this application aims to streamline the debugging process and offer alternative, more efficient solutions.

User Interface: The app features a user-friendly interface built with Streamlit, allowing users to easily input their Python code for analysis.

​

Code Analysis Output:

  • Mistakes and Bug Descriptions: The AI identifies mistakes, bugs, and areas for improvement in the submitted code, providing detailed explanations of the issues.

  • Corrected Code Snippets: For each identified issue, the app presents corrected code snippets, demonstrating how to resolve the problems.

  • Alternative Solutions: The analyzer suggests alternative approaches to solving code problems, emphasizing more efficient coding practices.

Benefits:

  • Efficiency: Automates the code review process, saving users valuable time and effort in debugging.

  • Learning Tool: Offers detailed feedback and suggested solutions to help users enhance their coding skills and understanding of Python best practices.

  • Streamlined Workflow: Provides a seamless and efficient workflow for code review and debugging tasks through its intuitive interface and AI-powered analysis.

Skills Used:

  • AI and Machine Learning: Leveraging AI for code analysis and feedback

  • Web Development: Building a user-friendly interface using Streamlit

  • Software Debugging: Identifying and correcting code issues

  • Alternative Solutions: Suggesting efficient coding practices and solutions

​

IPL  Data Analysis

This project utilizes Streamlit to create an interactive web application for analyzing IPL data. Key objectives include mastering data processing and visualization in Python, gaining insights into IPL trends, and building machine learning and neural network models for score predictions. 

Skills Used:​

  • Data Cleaning: Python

  • Data Analysis: Python, Pandas

  • Data Storage: CSV file handling

  • Data Visualization: Matplotlib, Seaborn

Health Insurance Exploratory Data Analysis

Welcome to the Insurance Insights Project! This repository is dedicated to uncovering transformative insights from a health insurance dataset. Health insurance in India is an emerging sector driven by the rise in the middle class, higher hospitalization costs, expensive healthcare, digitization, and increased awareness. Insurance companies face the challenge of setting accurate insurance premiums with limited information about the insured population. This project involves performing an Exploratory Data Analysis (EDA) to identify significant variables related to charges, explore data distribution, and provide actionable insights for data-driven business decisions. Key findings include the positive correlation between age and charges, the higher charges for smokers, the direct correlation between BMI and charges, and the impact of the number of children on charges. These insights help tailor insurance offerings, optimize risk and coverage, and ensure fair premium pricing.

Skills Used:​

  • Data Cleaning: Python

  • Data Analysis: Python, Pandas

  • Data Storage: CSV file handling

  • Data Visualization: Matplotlib, Seaborn

YouTube Scraping and Analysis Project

Welcome to the YouTube Scraping and Analysis project! This repository is dedicated to exploring the dynamics of YouTube success through the analysis of revenue, subscribers, and view counts. Leveraging Python's Beautiful Soup and Requests libraries, we conducted a comprehensive analysis to uncover valuable insights into YouTube channel performance. The project involved extracting data from YouTube channels, focusing on key metrics such as revenue, subscribers, and view counts, and ensuring data quality through robust cleaning techniques. Our analysis highlighted correlations and key factors influencing YouTube channel success, providing a deeper understanding of what drives performance on the platform.

Skills Used:

  • Web Scraping: Beautiful Soup, Requests

  • Data Cleaning: Python

  • Data Analysis: Python, Pandas

  • Data Storage: CSV file handling

  • Data Visualization: Matplotlib, Seaborn,Powre BI

Data Science Job Market Analysis

Welcome to the Glassdoor Data Science Job Analysis repository! This project empowers you to delve into the realm of data science job opportunities by seamlessly scraping, cleaning, and analyzing data from Glassdoor. By analyzing salary estimates, examining company sizes, and exploring geographical distribution, the project aims to provide professionals with actionable insights to negotiate competitive compensation, align with their preferred work environments, and make informed decisions regarding relocation or remote work. 

Skills Used:

  • Web Scraping: Beautiful Soup, Requests

  • Data Cleaning: Python

  • Data Analysis: Python, Pandas

  • Data Storage: CSV file handling

  • Data Visualization: Matplotlib, Seaborn

bottom of page