• My Research experience and objectives lie at the intersection of Machine learning, NLP, deep learning and information retrieval. I have extensive experience in Python for implementing statistical, machine learning and NLP. EDUCATION
University of Washington, Seattle, WA Mar 2018 - Jun 2019
• Master’s degree in Computer Science and Systems, specialization in Data Science. GPA: 3.9
• Fully paid international graduate scholarship recipient Iran university of Science and Technology, Tehran, Iran Sep 2011 - May 2016
• Bachelor’s degree in Computer Science. GPA: 3.66 EXPERIENCE
University of Washington, Graduate Research Assistant, Seattle, WA Mar 2018 -present
• An artificial intelligence-powered software that conducts conversation, answer questions.
• Cleaned Amazon(Q,A) dataset and applied TF-IDF
• Applied information retrieval methods such as BM25, Glove Embeddings. Samsung R&D, Machine learning Scientist and data analyst, Tehran, Iran Sep 2013 -Aug 2017
• Designed and developed a song recommender system for Samsung Tune which is a smart speaker.
• Implemented a scalable collaborative filtering method using Spark to provide service for millions of requests.
• Made a batch-training model to process sensor data in Smart Home project. Sadra Consulting Company, Software Engineer, Tehran, Iran Jun 2014 - Mar 2015
• Responded to study website behavior patterns using site metric tools to analyze and optimize business results and user experience for the customers.
Iran University of Science and Technology, Software Engineer, Tehran, Iran May 2014 - Mar 2016
• Detecting epileptic seizure to alarm patients before the attach to run away from dangerous places. Artificial Intelligence Lab, Software Engineer, Tehran, Iran Mar 2013 - Mar 2016
• Using of Dynamic Synapse Neural Networks for noisy signals.
• Processing a DSNN has been developed for EEG signals classification.
• A genetic algorithm learning method with different fitness functions has been used to optimize the neural network. PROJECTS
Personalized movie recommender, Movie recommend system
• A movie recommender system was designed and developed to rank and recommend movies.
• The ranking mechanism was implemented as a multi-label classification problem using a multi-layer perception. Predicting User’s Preference for a Movie, predicting whether a user likes a movie or not.
• Analyzed and cleaned 1M Movielens dataset.
• Used different recommendation techniques that can predict movie tastes such as collaborative filtering, content-based filtering and Hybrid filtering.
• Proposed a model that enhances the collaborative filtering that helps to make recommendations or predictions based on user’s demographic information.
• Used K-means clustering to make the clusters of users that are similar to each other in terms of their demographic information and solved cold start problem when a new user without any watched history comes. Novel recommendation engine, predicting what a user will buy next.
• The user and item representation were learned through the multi-label classifier optimizing what a user will buy next.
• Developed the model by Keras and suggests a personalized vector for each user and consequently recommends a similarity among the user with similar taste.
• Solved cold start problem for newly opened stores and outperforms the currently available models. No-Show Prediction, predicting whether or not a patient will be a “no-show”.
• Applied different machine learning model like SVM, XGB and logistic regression. Sepsis Prediction predicting sepsis in hospitalized with machine learning.
• Analyzed, cleaned and solved the problem of unbalanced mimic data and applied different ML algorithms. Breast Cancer Prediction, predicting Drug Induced live injury (DILI)
• Predicting models of Drug Induced live injury (DILI) using MCF7 (breast cancer).
• Analyzed and cleaned data and applied different algorithms and feature selection (t-test & PCA) SKILLS
• Programming Languages: Python(expert), R, Java,C, MySQL, HTML, LaTex
• Analytics: Scikit-learn, Graph-lab, Pandas, Numpy, Glmnet, Scipy, Earth, Keras, TensorFlow
• Cloud Computing Services: AWS, Docker
• Have used tools such as NumPy, Glmnet and Pandas for analytics.
• Scikit-learn and caret for machine learning (Regression model, Classification, Random forest)