Skip to content
View yildize's full-sized avatar

Block or report yildize

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
yildize/README.md

Hey

Hey there it's Erdogan. I am interested in AI, Machine Learning, Deep Learning.

Some of my Notes:

Some of my Projects:

Movie Recommendation App

  • Machine learning based movie recommendation app. Users can combine different movies and get suggestions for the provided combination. I trained a collaborative filtering model and obtain embedding vectors for about 60k movies. Recommendations relies on these embedding vectors. Designed a simple front-end with angular and backend with flask.
  • deployed app
  • github page

Joke Recommender

  • Again I trained a collaborative filtering model on some jokes. This recommendation system works a little different than the above. I ask users to rate some jokes first, and based on these jokes, system creates a taste profile for the user (embedding vector of the most similar user based on the ratings). The recommendations is done based on this taste profile.
  • Binder App: Binder
  • github page

Collaborative Filtering Notebook

Tabular Models Notebook

NLP Notebook

"Beauty" Detector

  • It is a computer vision application. Basically the app tries to detect "beauty". This project is done only for educational purposes and fun.
  • I collected data using google search. Old version does a binary classification and assign a probability for "beauty" based on the given image.
  • So it was basically a binary image classificiation model.
  • Old Version Model Training
  • Later I improved this app, added it some other features. Two different model works on it. One is a multiclass image classifier (based on gender and "beauty" level). Another is an image regressor (age prediction).
  • Binder App: Binder
  • github page

Price Predictions

Survey App (Angular)

  • It is a simple survey app that I've created for a friend that allows users to create surveys and share it with others. I used Angular. It uses Firebase to store the data
  • deployed app page
  • github page

Some Websites

Crypto Price Compare Bots

  • Some web crawlers that compares crypto prices on different crypto markets and send notification e-mails.

Pinned Loading

  1. README README Public

  2. Movie-Cocktail-Project Movie-Cocktail-Project Public

    A machine learning based movie recommendation app.

    Jupyter Notebook

  3. DeepLearningSpecialization-AndrewNg DeepLearningSpecialization-AndrewNg Public

    Jupyter Notebook

  4. MachineLearning-CourseContents-AndrewNg MachineLearning-CourseContents-AndrewNg Public

    MATLAB

  5. Predict-Future-Sales Predict-Future-Sales Public

    Mağaza-ürün satış rakamı tahmini

    Jupyter Notebook 2

  6. Survey-App-Angular Survey-App-Angular Public

    TypeScript