- Convolutional Neural Network to Recognize Persian Handwritten Digits using Python, Theano
Achieving 97% accuracy on a 20,000 sample dataset
Exporting network parameters into an iOS app and building a keyboard to recognize handwritten Persian digits.
- Convolutional Neural Network to recognize cystic fibrosis patterns in lung HRCT images using Python, Theano
- Convolutional Neural Network to recognize ILD Disease patterns in lung HRCT images in 3D Voxel Patches
- Social Mobile Banking App at Dotin, Fanap Co.
- Implementation of Neural Network to Recognize MNIST Handwritten Digits
- Design and Implementation of Plagiarism Detector Software
Undergraduate Final Project
Design and implementation of plagiarism detector software to detect and report plagiarism probability of a given document using Google Search Engine in C#.
- Hoortash | Social Network App in iOS
As my capstone project in iOS and to show the world my abilities in development iOS apps, I developed a social network app called Hoortash.
There are some features of this app:
- Instantly updates Feeds whenever a user posts, likes or comments on a post.
- Cached image system to improve app performance.
- 2-way authentication, one-step Login / Sign Up and Facebook Authentication
- Ability of leaving comment and like a post.
- Ability of editing user profile
- Building Recommender System based on Movie Ratings
From MovieLens 100k Dataset from GroupLens Research. This dataset consists of ratings on a scale of 1 to 5. The dataset has 943 users, and 1682 movies.
- Anomaly Detection Application to Detect Anomalous Behavior in Server Computers
The features measure the throughput (mb/s) and latency (ms) of response of each server. Dataset consists of 307 examples.
- Spam Filter using Support Vector Machine (SVM)
Implementation of a spam filter with Support Vector Machine (SVM) to classify spam (y=1) and non-spam(y=0) emails. The dataset is based on a a subset of the SpamAssassin Public Corpus, available at: http://spamassassin.apache.org/publiccorpus/
- Image compression using K-Means Algorithm