/Artificial-Intelligence-Course-Projects

Artificial Intelligence course projects/ University of Tehran/ Fall 2022 - 2023

Primary LanguageHTMLMIT LicenseMIT

Artificial Intelligence Course Projects

CA0 - An Introduction to ML Tools

In this project, We learnt to use pandas, matplotlib and some other famous libraries used in ML and AI. This project included the Titanic's passengers dataset to predict which passengers could survive and which couldn't. It didn't contain any complicated AI models and the prediction was done through calculating probabilities and using some statistics.

CA1 - Search

This project was about a problem called "DZ Day". The problem is about a person who wants to search a graph and accomplish his goal; Our job was to help him to achieve his goal in the minimum time. This included searching the states space with different methods such as BFS, IDS, and A*.

CA2 - Games and Genetics

This project was actually two parts. The first part was about the Equation Problem, which is about some numbers and operators to be put in an order such that the equation is satisfied. We were asked to solve this problem using a genetic algorithm.
The second part was about a game called Sim. the purpose of this project was to implement a bot that will play the game against a random opponent (Someone who choose his next move randomly)using minimax trees.

CA3 - Naive Bayes

This project was about classifying the news articles extracted from Asr Iran into their relevant categories. It included preprocessing the text using Hazm library and classifying the news with a self-implemented Naive Bayes classifier.

CA4 - ML

In this project we were asked to predict the diabetes dataset with different models in Scikit-Learn library. This was done with Decision Tree, K-NN, Logistic Regression, and Random Forest models.

CA5 - Neural Networks and Deep Learning

In this project we implemented a neural network to classify pictures of Arabic numbers. We were given a semi-complete code and were asked to implement the TODO parts. We also used Keras and Tensorflow libraries to predict CIFAR-10 dataset.