Mylie Mudaliyar's Projects
Credit Fraud Detection of a highly imbalanced dataset of 280k transactions. Multiple ML algorithms(LogisticReg, ShallowNeuralNetwork, RandomForest, SVM, GradientBoosting) are compared for prediction purposes.
My cs50 codes (2022) which I had submitted. Feel free to refer : )
Disney Plus Hotstar Clone using html, css and js.
E-commerce Online Luxurious Furniture Store Website
An Interactive Sales Dashboard on MS Excel of a Local Coffee Beans Shop with 1000 purchases.
The feature engineering techniques discussed are - dimensionality reduction(pca), scaling(standard scaler, normalizer, minmaxscaler), categorical encoding(one hot/dummy), binning, clustering, feature selection. These are techniques performed on a dataset consisting of Californian House Prices.
A very well known game called flappy bird coded in java.
A datascience project analysing the effect of poverty and international trade among and within 196 countries on global carbon footprint.
Lab Exercises done for course CSE-3002 Internet and Web Programming
A full stack development of this a social networking website using HTML, CSS, JS, AJAX, PHP and MYSQL; and consists several user functionalities
A repo for all the relevant code notebooks and datasets used in my Machine Learning tutorial videos on YouTube
Using machine learning models to predict various major diseases and deploying in Streamlit.
Responsive food website made using html, css and javascript.
This is a Restaurant Recommendation System made as a part of SmartIntenz Externship Program. This project involves tf-idf matrix for recommendation and is deployed using flask.
Machine Learning Model to predict Salary of Software Developers and deployed in Streamlit. The model was trained on a real time dataset of 2022 published by Stack Overflow Survey with over 73k entries..
A basic bank/atm system using java core and oops concepts (inheritance,polymorphism,etc) wherein one can check balance , withdraw and deposit money.
Small Projects using html, css and javascript.
A methodology integrating web scraping and K-means clustering to systematically categorize and prioritize responses. Custom web scrapers extract com-ments from YouTube videos, overcoming limitations of traditional screen scraping. Sentiment analysis using TextBlob and HuggingFace provides in-sight into overall video sentiment.
This project employs ML algorithms and NLP techniques, including TF-IDF vectorization, to build a TED Talk recommender. By analyzing talk metadata and computing cosine similarity, it offers personalized recommendations, enhancing user engagement with TEDx content.
This is a responsive travel website made using html, css and javascript.
A university admission predictor using scores such as GRE, TOEFL, CGPA, LOR ratings, SOP ratings, College ratings, and work experience. This model is deployed using StreamLit.
All the resources I am using to learn deep learning and neural networks.
Stocks Visualization of top tech companies (Apple, Google, IBM, Tesla, Amazon, sp500) from years 2012-2020.
A very basic web scraping project of faculty details from a university website using Beautiful Soup. The details get saved in a csv file.