Projects
WomenTechWomenYes (WTWY) — Exploratory Data Analysis (EDA)
The WomenTechWomenYes (WTWY) foundation is striving to achieve gender equality in the tech world and to encourage more women to participate in this field. In line with this mission, WTWY is preparing to host an annual gala. This gala event will not only raise awareness but also serve as an opportunity to increase participant numbers and boost donations. Click to read the full article.
Building a Box Office Revenue Prediction Model Using Web Scraped Rotten Tomatoes Data
In today's data-driven world, big data analytics and machine learning are revolutionizing many industries, including the film industry. In this post, I'll share my experience of scraping data from the Rotten Tomatoes website and using this data to create a box office revenue prediction model.
Building a Streamlit Interface Model to Detect Vehicle Insurance Fraud
Insurance fraud has become a significant issue for both insurance companies and customers. Fraudulent cases can cost insurance companies billions of dollars and lead to increased premiums for innocent customers. Therefore, finding an effective method to detect fraud is of critical importance. With this goal in mind, I have developed a model that detects insurance fraud using vehicle insurance data and present it with a user-friendly interface through Streamlit.
Twitter Sentiment Analysis with a Streamlit Interface (NLP Project)
Natural Language Processing (NLP) is a powerful tool for extracting meaning and analyzing sentiment from text data. In this project, I developed a sentiment analysis model using Twitter data and presented it through a user-friendly Streamlit interface.
Key Features
-
Data Collection: Tweets were gathered using the Twitter API and processed with standard NLP techniques such as cleaning, stemming, and stop-word removal.
-
Model: A machine learning model was trained to classify sentiments as positive, negative, or neutral using NLP methodologies.
-
Interface: A Streamlit dashboard was created featuring real-time sentiment analysis, visualizations, and custom input fields for user queries.
This project combines NLP techniques with sentiment analysis, making it easier to derive meaningful insights from social media data.
Hybrid Book Recommendation System
This project combines collaborative and content-based filtering techniques to create a hybrid recommendation system for books. Using user preferences and book metadata, the system provides highly personalized suggestions.
Key Features
-
Collaborative Filtering: Analyzes user ratings to recommend books liked by similar users.
-
Content-Based Filtering: Uses book genres, authors, and descriptions to find similar books.
-
Scalable Design: Built to handle large datasets efficiently.
This hybrid approach ensures accurate and diverse recommendations, catering to varied user tastes.
Note: You can access all the codes of the projects by clicking on the github icon below.
Click for more information