My Name is Mohammadsadegh Saberian,

I am an undergraduate student in computer engineering at Department of Computer Engineering at Sharif University of Technology , Tehran, Iran. I am interested in Application of Machine Learning and Deep Learning in various engineering fields, such as Signal processing and Bioinformatics (mainly Structural Bioinformatics) for solving real-world problems.

I am currently working on a project aimed to predict genetic diseases caused by mutations in protein sequence. To this end I am designing a deep neural network for extracting informative features from the protein's tertiary structure.

Address: Department of Computer Engineering, Sharif University of Technology, Azadi Avenue, Tehran, Iran, 11155-11365.

My CV can be found here.

Research Experience

Genetic Diseases Prediction

Department of Stem Cells and Developmental Biology, Royan Institute for Stem Cell Biology and Technology, Tehran, Iran.

Under Supervision of Dr. Ali Sharifi-Zarchi and Dr. Razieh Karamzadeh.

Feb. 2017 - Present

This project is aimed to predict genetic diseases caused by mutations in protein sequence. To this end I am designing a deep neural network for extracting informative features from the protein's tertiary structure. A presentation of this project can be found here .

Network Traffic Identification

Information, Network and Learning Lab (INL) , Department of Computer Engineering, Sharif University of Technology, Tehran, Iran.

Under Supervision of Prof. Mahdi Jafari Siavoshani

May 2016 - Sep. 2017

I trained various models, including auto-encoders (AE), stacked auto-encoders (SAE), convolutional networks (CNN) to identify the end-user application, e.g. Skype, from the network traffic. The final model improved the system accuracy from 91% to 96%. (Keras + TensorFlow)

You can find more details here.

Indoor Positioning System

Shetab Saman Peyvand Co., Tehran, Iran.

(June 2017 - Sep. 2017)

Designed and trained a neural network model using a combination of Recurrent and Convolutional networks (LSTM + CNN) for an indoor positioning system. The model uses several of mobile sensors such as magnetometer. I trained and tested different network architectures and the final model improved the baseline accuracy by 30%. (Keras + TensorFlow)

Skills

Programming Languages

  • R
  • Matlab
  • Python
  • C++
  • Verilog

Structural Bioinformatics Tools

  • VMD
  • Modeller
  • NAMD

Machine Learning and Deep Neural Network tools and languages

  • R
  • Keras
  • Tensorflow
  • Scikit-learn

Education

Sharif University of Technology, Tehran, Iran.

B.S. in Computer Engineering, GPA (up to now): 17.90/20 (3.62/4).

Sep. 2013 - Expected 2018

Related Courses:

  • Signals and Systems (20/20)
  • Data Analysis (20/20)
  • High-dimensional Data Modeling and Analysis (Current Semester)
  • Data Transmission (20/20)
  • Computer Measurements and Control (20/20)
  • Stochastic Processes (20/20)
  • Network Coding (19.5/20)
  • Computer Networks (20/20)
  • Engineering Probability and Statistics (20/20)
  • Digital System Design (19/20)
  • Microprocessors (18/20)
  • VLSI Design (17.5/20)
  • Numerical Analysis (19/20)
  • Advanced Programming (C++) (20/20)

Shahid Hashemi-nezhad (NODET), Mashhad, Khorasane-e-razavi, Iran.

High School Diploma in Mathematics and Physics,

GPA: 19.30/20.

Sep. 2009 - July 2013

Selected Projects

  • Clustering Stages of Alzheimer’s Disease Based on Gene Analyzing using R. I used diffrenet feature selection methods for finding informative features. The results can be found here.

  • Analysing universities around the world regarding several criteria such as citation. The data was crawled from here using Python. The results can be found here.

  • Analysing world countries based on World Development Indicators (WDI) data and clustering similar countries using R. The results can be found here.

  • Second-Hand Cars Prices Prediction using R. The results can be found here.
  • Analysis of Trends in International Mathematics and Science Study (TIMSS) data using R. The results can be found here.

  • Analysis of Iran Football league (Persian Gulf Pro League) data. The data was crawled from the official website using R. The results can be found here.

  • Design and implementation of a Tempurature Control and Alarm System based on Arduino Uno, WIFI module and GSM module with a Python-based GUI. A technical report can be found here.

  • Design and implementation of a Security Alarm System based on Arduino Uno, WIFI module and GSM module. A technical report can be found here.

  • A Simulation of computer network using GNS3.
  • A comprehensive study on EXT4 FileSystem on Ubuntu 12.04. A technical report can be found here.
  • Network on Chip (NoC), implemented in Verilog, synthesizable on FPGA. A technical report can be found here.
  • Icompiler, designed for compiling a given language (LULU), implemented in C++.
  • Great Little War game, implemented in C++ using QT.
  • Billiard game, implemented in C using GTK+.

Article

  • An article about James William Cooley, co-developer of Fast Fourier transform (FFT) [In Farsi]. The article can be found here.

Awards and Honors

Ranked 2nd Fanavard Data Mining Challenge.

Dec. 2016

This was a nationwide competition with nearly a hundred participating teams. The challenge was to detect goods smuggling using transportation records, e.g. time and routes, across the country. We developed a scoring system based on popularity of the routes and deviations from the expected routes. This helped us to detect if there was any suspicious activity during a transportation record.

Ranked 3rd in GPA.

Sep. 2015 - Present

Among Hardware Engineering Students, Computer Engineering Department, Sharif University of Technology.

Member of National Organization for Development of Exceptional Talents, NODET.

Sep. 2005 - Present

Ranked 124th in the Iranian National University Entrance Exam.

Jun. 2013

The Iranian National University Entrance Exam has almost 100,000 participants.

Teaching Experience

Publications