- Data-Science-Course-Projects-S2024
- Intro
- Assignment 0: Web Scraping and Introductory Data Analysis
- Assignment 1: Statistics Analysis Tools
- Assignment 2: Investigating Open-Ended Questions
- Assignment 3: Spotify Dataset Analysis
- Assignment 4: Loss Functions and Regression
- Assignment 5: Feature Engineering Techniques
- Assignment 6: Dimensionality Reduction and Unsupervised Learning
- Assignment 7: Semi-Supervised Learning and LLMs
- Project
This repository contains the assignments for the Data Science course at the University of Tehran. The assignments are completed by the group of students in the course and are used to demonstrate the skills and knowledge they have gained throughout the course. They are completed using the python programming language and are focused on data analysis, visualization, and machine learning. The assignments are completed using real-world data sets and are designed to give students hands-on experience with the tools and techniques used in data science. Apart from the assignments, the repository also contains the final project for the course, which is a group project that involves working on a real-world data science problem from start to finish. In the final project we were supposed to gather data, clean it, analyze it, and build a machine learning model to solve a specific problem.
In this assignment, we were tasked with scraping data from a website and performing some basic data analysis on it. The website we scraped was Etherscan and the data we scraped was the transaction history of the Ethereum blockchain. We used the requests
and BeautifulSoup
libraries to scrape the data and then used pandas
and matplotlib
to perform some basic data analysis on it. Then we analyze different sampling methods and their effects on the data analysis process.
In this assignment, we were tasked with using different statistical analysis tools. First, we used monte carlo simulation to estimate the value of pi and outcome of a mensch game. Then we analyzed the Central Limit Theorem (CLT) and its effects on the distribution of sample means. Finally, we implemented two different hypothesis tests. The first was a for outcomes of flipping a coin and the second was for checking if the GPA of students is affected by the amount of job they have, this one is based on a real-world dataset.
In this assignment, we were tasked with investigating open-ended questions. For this purpose we were given two datasets, one containing the information about passengers of the sunken RMS Luistania
and the other containing the information about the salaries statistics of scientists. We used these datasets to answer some open-ended questions.
In this assignment, we were tasked with analyzing a dataset containing information about songs on Spotify. We used this dataset to gain some interesting insights about the songs and their features.For this part we used the pyspark
library to analyze the dataset. We also used the matplotlib
and seaborn
libraries to visualize the data.
In this assignment, we were tasked with implementing some basic loss functions and using them to train a simple linear regression model. We implemented the Mean Squared Error (MSE) and Mean Absolute Error (MAE) loss functions and used them to train a simple linear regression model on a dataset containing information about house prices. We also implemented the R^2 score to evaluate the performance of the model. The dataset used in this assignment is the Diabetes
dataset from the sklearn
library. We also used Ordinary Least Squares (OLS) to train the model and compared the results with the results obtained using the loss functions we implemented.
In this assignment, we were tasked with implementing some feature engineering techniques and using them to train a simple linear regression model. The given dataset contains information about outcome of shoots in football matches. Then we were tasked with regressing the price and horsepower of cars with a different dataset. Then we trained a Multivariate and a K-Fold model to do the regression.
In this assignment, we were tasked with implementing some dimensionality reduction techniques and using them to train a simple K-Means and DBSCAN clustering model. We implemented the Principal Component Analysis (PCA) dimensionality reduction technique and used it to reduce the dimensionality of a dataset containing information about the diabetics data. We then used the reduced dataset to train a K-Means and DBSCAN clustering model to cluster the songs based on their features.
In this assignment, we were tasked with implementing some semi-supervised learning techniques and using them to train a simple Label Propagation model. We implemented the machine learning methods of label propagation and also used LLMs for label propagation and used the results to train a model on a dataset containing information about the IMDB review data. We then used the trained model to predict the labels of the unlabeled data points. For LLM model we used Phi-3 model and for the label propagation we used the SelfTraining
method in the sklearn
library.
For this phase we chose to scrape We Buy Car website to gather data about the cars they have for sale. The gathered data includes the car's make, model, year, price, and much more.
In this phase we cleaned the data we gathered in the previous phase. We removed any duplicates, missing values, and outliers. We also converted the data into a format that is suitable for analysis. We also performed some exploratory data analysis to gain some insights about the data.
In this phase we used the cleaned data to train a regression model to predict the price of the cars. We used different regression techniques such as Neural Networks, SVM, Decision Tree, and other methods as well. We also used different feature engineering techniques to improve the performance of the model. Finally, we evaluated the performance of the model using different evaluation metrics such as the