zlacker

Location: Dearborn, MI

  Remote: Yes (EST but flexible)

  Willing to relocate: No
  

Technologies: Data science, Machine Learning, Deep Learning

  Résumé/CV: 
    Throughout my career, I am proud to have accomplished the following:
 

* Developed and applied advanced Machine Learning techniques, including XGBoost optimization, to analyze and interpret large and complex oncology patient data sets, resulting in improved understanding of the effectiveness of tumor-targeted drugs and personalized treatment recommendations.

* Processed and analyzed over 1 billion clinical trial records with exceptional results by utilizing a combination of Exploratory Data Analysis(EDA), advanced feature engineering techniques and statistical analysis with data visualization, resulting in data-driven decision-making.

* Implemented a comprehensive Machine Learning approach for pre-processing large and complex clinical data sets, including data quality control,normalization, and dimensionality reduction, and utilized linear models to accurately predict and effectively communicate insights through interactive data analytics dashboards.

* Applied advanced statistical methods such as hypothesis testing and regression analysis to identify key factors influencing drug effectiveness on oncology patients, using XGBoost optimization to enhance model performance and accuracy.

* Improved model accuracy through rigorous training and validation strategies, including k-fold cross-validation and optimal test-train splitting methods, ultimately reducing error loss.

* Experienced in optimizing machine learning models in the field of life sciences through hyperparameter tuning and model selection, and proficient in evaluating model performance using key metrics such as accuracy, precision, recall, F1-score, AUC-ROC, and confusion matrix.

* Highlighted expertise in machine learning and deep learning through repeated top finishes in Kaggle data science competitions, including the successful completion of several challenging projects.

* Developed a highly accurate and relevant personalized book recommendation system using big data techniques and advanced data analysis methods like Popularity, Correlation, Content-Based Filtering, and Collaborative Filtering, resulting in improved customer experience and increased sales.

*Successfully identified key factors influencing drug effectiveness on patients through exploratory data analysis and feature engineering, utilizing XGBoost techniques to optimize model performance and accuracy.

  Email: imshiv@umich.edu
  LinkedIn: https://www.linkedin.com/in/shiva-00mi/