Deep_Learning_Projects

1. Churn Modelling: In this project, dataset of churn's bank is used. The dataset contains information like customer id, credit score, gender, age, tenure, balance, if the customer is active, has a credit card, etc. The aim was to determine whether a customer of the bank would exit or close its bank account in the future based on past data of past customers. The dataset is given as churn_modelling.csv. An Artificial Neural Network is designed in order to achieve this goal. The final outcome tells whether a given customer is likely to exit the bank or not.

2. Image Recongnition: In this mini project, an Convolutional Neural Network moel that is able to detect various objects in images and also should be able to differentiate between different images of cats and dogs.

3. Google Stock Price Prediction: In this project, dataset containing past 10 years of Google's stock prices (i.e. both opening and closing) is used. The aim was to determine the future trends of this stock which would give rough idea to investors about its future prices. In order to achieve this goal, we designed an Recurrent Neural Network model which would take Google's past 10 years of stock prices as input and would tell the likely stock prices of the stock in upcoming time.

4. Fraud Detection: The aim of this project was to detect frauds in credit card applications. The dataset used contains information on customers applying for an advanced credit card. This is the data that customers provided when filling the application form. The task is to detect potential fraud within these applications. The final output would be a list of customers who cheated on their application.

5. Recomender System: In this project, the aim was to create a recommender system by working on a dataset which has exactly the same features as the Netflix dataset: plenty of movies, thousands of users, who have rated the movies they watched. The ratings go from 1 to 5, exactly like in the Netflix dataset. The final Recommender System will be able to predict the ratings of the movies the customers didn’t watch. Accordingly, by ranking the predictions from 5 down to 1, the Deep Learning model will be able to recommend which movies each user should watch. Two different models were created for this goal. The first model was created using complex Boltzmann Machines, and the second one was created using AutoEncoders.