Dr. Hassan Fathivavsari

Churn Prediction based on Machine learning (ML) algorithms (XGBoost, RF) using data provided by IPM institute in Iran. - Deploying best model into production using AWS tools (SageMaker) - Exploring Apache Spark and Big data Analysis with Hive. - Kaggle project: Chest X-ray Abnormalities detection: in this competition, we classify common thoracic lung diseases and localizing critical findings. The problem is approached by Convolutional Neural Network (CNN).

Implementation of Deep Neural Network with multivariate Regression problem to predict the flux and the shape of the stellar objects programming with Python. Achieved a 15% more accurate prediction than previous studies. - Data management on a robust astronomical dataset using SQL and Python in a limited time frame.

Employed ML algorithms (XGBoost, SVM, KNN) and clustering (K- means clustering) to predict emission lines of massive stars. - Analyzed high-throughput dataset (> 100 GB). Gained expertise in applying exploratory analysis (data visualization, PCA) and pattern recognition (Ensemble averaging, Linear regression). - Improved data mining process by developing a new pipeline, resulting in a 20% decrease in time needed to infer insights from customers.

Updated company webpage regularly [html]. - Developed a platform to monitor quality condition of product (Water purification filters) for the EasyWell costumers.

Updated company webpage regularly [html]. - Reported and followed up and maintenance of the mathematics department IT facilities to maintenance team.

Teaching assistant of undergraduate and graduate courses of physics and engineering. Designed and evaluated exams. - Tutoring of > 30 graduate and undergraduate students in Computer Physics courses, which includes weekly problem set designing and debugging students codes.

Data analysis, Machine learning pattern recognition, spectroscopic analysis