![](assets/images/image01.jpg?v=48fa8826)
Ehsan Mokhtari
Data Scientist | M.Sc. Data Science Student at Sapienza University of Rome | Passionate about Data Science, Machine Learning, Deep Learning and Computer Vision
About me
Hi, I'm Ehsan.
A Data Science Master's student at Sapienza, merging Computer Engineering with skills in Data Science, Machine Learning, and Software Development. Fueled by self-study projects, courses and books, I'm eager to contribute and thrive in the tech landscape.
Experience
Software Engineer Internship
Rural Cooperatives Organization of Iran
Oct 2020 / Dec 2020
Publications
Color Image Encryption Algorithm based on 3D Chaotic Economic Map
5th international conference on applied research in computer, electrical and information technology, Avicenna international community college LLC, Georgia, Sep 2020
Education
M.Sc. of Data Science
Sapienza Università di Roma
B.Sc. of Computer Engineering
University of Tabriz
Interests
Data Science
Machine Learning
Deep Learning
Computer Vision
Projects
Graph Analysis
Analyzing academic paper datasets, constructing citation and collaboration graphs; then focusing on the most connected component (top 10,000 papers with the highest citations); and finally dividing the analysis into Backend and Frontend components for extracting insights.
Building ML model to predict shot success in the Kobe Bryant shot selection dataset using various algorithms
This project employs machine learning algorithms to predict Kobe Bryant's shot success in the NBA, covering data preprocessing, EDA, feature engineering, and algorithm implementation, with a final step exploring a voting ensemble for enhanced accuracy.
Recommendation system on "Netflix audience behaviour - UK movies" dataset
Implementing our version of the LSH algorithm, which will take as input the user's preferred genre of movies, find the most similar users to this user, and recommend the most watched movies by those who are more similar to the user.
Building different NN models to predict cryptocurrency price
Applied API data to forecast cryptocurrency prices through diverse neural networks.
PunkyPoops NFT Collection
Implementing 8750 unique, programmatically-generated NFTs with distinct metadata for each using Node.js, analysing metadata for assigning price for each NFT according to its attributes using Python, storing on IPFS decentralized data-storage platform, deploying on Polygon blockchain as ERC-721 tokens using smart contracts and Solidity, publishing on opensea.io marketplace.
Real-time Face Detection and Recognition, Mask detection, Digit Detection
Developing real-time computer-vision programs for face recognition, digits and covid-19 mask detection using Machine-learning and Image-processing.
Text (both English and Persian) Tokenizer with TF and IDF Scores Calculation from scratch
Developing a program for tokenizing English and Persian texts (including words, numbers and links), sorting tokens and calculating Term-Frequency and Inverse-Document-Frequency scores for each.