Data Scientist

About

A seasoned Data Scientist with a rich academic background and extensive professional experience, I specialize in leveraging advanced analytical techniques and data-driven insights to drive meaningful business outcomes. My expertise spans a wide range of technical proficiencies, including Python (with libraries like TensorFlow, PyTorch, Scikit-Learn, Pandas, NumPy, Matplotlib) and data visualization tools like Tableau. Proficient in SQL and version control with Git, I have a proven track record of implementing effective data solutions in various industries. At the Royan Institute, my data analysis contributed to saving $1 million, showcasing my ability to deliver significant financial impacts. My soft skills in problem-solving, effective communication, teamwork, and time management complement my technical abilities, making me a well-rounded professional. I am continually seeking opportunities to collaborate on projects that challenge my skills and contribute to meaningful advancements in data science.

Technical Skills:

Python (TensorFlow, PyTorch, Keras, Scikit-Learn, Scipy, Pandas, NumPy, Matplotlib)
SQL
Data Visualization (Tableau)
Version Control (Git)

Education

Degree	Institution & Year
Ph.D. in Software Engineering	The University of Western Ontario, Canada (2021-Present)
Master of Engineering Science in Software Engineering	The University of Western Ontario, Canada (2019-2021)
Master of Science in Industrial Engineering	Amirkabir University of Technology (Tehran Polytechnic), Iran (2012-2014)
Bachelor of Science in Applied Mathematics	Ferdowsi University, Mashhad, Iran (2007-2012)

Professional Experience

Research Assistant at Western University (2019-Present)

Established a Clustered Federated Learning approach with CNNs to address the challenges of non-IID data in sentiment analysis applications.
Formulated a Federated Learning model using LSTM and Transformer networks to assess imbalanced dataset impacts on sentiment analysis training.
Developed a transfer learning method using CNNs and wavelet decomposition for human activity recognition, refining accuracy by re-training for individuals.
Introduced a personalized CNN-based approach with signal processing for sensor data-driven human activity recognition, enhancing individual-specific accuracy.
Created CNN and FFNN models using leave-one-subject-out validation for accurate human activity recognition from sensor data for new individuals.

Software Developer Software Developer at Algobot (Cryptocurrency software) (Apr-Aug 2022)

Developed a scanner for the Cryptocurrency market to filter cryptos with high momentum
Developed a bot to draw static and dynamic supports and resistances automatically and predicting future prices
Examined various strategies with backtests and real-time tests to find a good trading strategy

Data Scientist at Xlscout (Sep-Dec 2021)

Found U.S. patents’ solutions and related sections in each patent using machine learning techniques.
Summarized the main idea for the solution of each patent using extractive and abstractive models such as Bert, Pegasus, and so on.

Senior Data Analyst at Royan Institute (Jan-Jun 2017)

Applied different classification methods such as C4.5 decision tree, logistic regression, radial basis function neural networks, and feed-forward neural networks to predict do we have to freeze the umbilical cord blood stem cells or not
Predicted quality of cord blood for newborns using ensemble supervised and unsupervised machine learning techniques and saving $1 Million for the Royal Institute using our machine learning approach

Senior Data Analyst at Sina Bank (Nov 2016 - Feb 2017)

Extracted strong and frequent rules between the bank’s web services to improve the recommender system in the Customer Relationship Management (CRM) section
Used data mining techniques, in particular, association rules to create and improve the recommendations for the banking industry

Data Analyst at Traffic Department of Mazandaran Province (Jul 2016 - Nov 2016)

Identified the factors affecting traffic violations using the Logistic Regression (LR) method, such as gender, age, education level, city, type of vehicle, driver’s experience certificate, and wearing glasses and evaluation of the model by the Hosmer and Lemeshow test and McFadden’s pseudo R2 index
Predicted the potentially offending drivers using a novel hybrid classification approach; the Multi-Layered Perceptron Neural Network trained by the Genetic Algorithm and Levenberg–Marquardt algorithm

Data Analyst at Saderat Bank of Iran, (Feb 2014 - May 2014)

Predicted Customer Churn Using Machine Learning Techniques, in particular, Support Vector Machine (SVM), Naive Bayes, and Neural Networks
Predicted multi-level Customer Churn Using Neural Networks; identifying customers who are prone to churn before completely leaving the organization
Estimated the loss of Saderat Bank due to customer churn using supervised machine learning techniques and a proposed loss function

Research and Publications

I have made contributions to multiple research projects and publications, primarily focusing on machine learning, deep learning, and federated learning, as they intersect with natural language processing, customer relationship management, and human activity recognition. Several of my contributions include:

“Personalized Models for Human Activity Recognition with Wearable Sensors: Deep Neural Networks and Signal Processing” (Springer Applied Intelligence) Link
“Deep Neural Networks for Human Activity Recognition with Wearable Sensors: Leave-one-subject-out Cross-validation for Model Selection” (IEEE Access) Link
“A Case Study of Fintech Industry: A Two-Stage Clustering Analysis for Customer Segmentation in the B2B Setting” (Journal of Business-to-Business Marketing) Link
“Investigating the performance of technical indicators in electrical industry in Tehran’s Stock Exchange using hybrid methods of SRA, PCA and Neural Networks” (Conference on Thermal Power Plants) Link
“Gene selection from microarray expression data: A Multi-objective PSO with adaptive K-nearest neighborhood” (arXiv) Link
“Customer Churn Prediction Using a New Criterion and Data Mining; A Case Study of Iranian Banking Industry” (International Conference on Industrial Engineering and Operations Management) Link
“Customer Churn Prediction Using a Meta-Classifier Approach; A Case Study of Iranian Banking Industry” (International Conference on Industrial Engineering and Operations Management) Link
“Productivity change and its determinants: Application of the Malmquist index with bootstrapping in Iranian steam power plants” (Utilities Policy) Link
“Dynamic changes in CO2 emission performance of different types of Iranian fossil-fuel power plants” (Energy Economics) Link
“Assessing productivity changes using the bootstrapped Malmquist index: the case study of the Iranian construction industry” (International Conference on Industrial Engineering and Operations Management) Link

Soft Skills:

Problem-Solving
Effective Communication
Teamwork
Time Management

Contact

Feel free to explore my repositories and connect with me for collaborations or discussions in my fields of interest.