I am a passionate Data Analyst with a strong focus on machine learning and data science. My expertise spans data analysis, model building, and deep learning techniques. With extensive experience in Python, SQL, Excel, and Power BI, I strive to extract meaningful insights and actionable conclusions from data. I have hands-on experience in building machine learning models using advanced algorithms and frameworks such as TensorFlow, Keras, PyTorch, Scikit-Learn, along with a deep understanding of the key principles of supervised and unsupervised learning.
New Brunswick Institute for Research and Data Training (NB-IRDT), Fredericton, New Brunswick
May 2024 β August 2024
- Conducted in-depth data analysis and modeling on large healthcare datasets using Stata, extracting actionable insights for healthcare research support.
- Developed a comprehensive dictionary of derived variables to ensure clarity and ease of use across different programming languages for future research.
- Collaborated with senior researchers to create statistical models that aided in better understanding the data and identifying trends and patterns relevant to healthcare.
Infosys, Chennai, India
Jan 2020 β Nov 2021
- Worked as a Systems Engineer, focusing on automating testing processes and utilizing Python, Java, and SQL in an Agile environment.
- Developed automated testing frameworks using Selenium WebDriver and integrated with Jenkins for continuous integration.
- Enhanced software solutions and ensured the functionality of machine learning algorithms through rigorous testing and deployment.
June 2023 β September 2023
This project involves using a convolutional neural network (CNN) in pervasive healthcare to detect skin diseases from images. Various Keras optimizers were compared, and the most accurate one was employed to predict skin abnormalities, enhancing detection efficiency.
Link to Publication
This paper compares various supervised machine learning techniques, including Decision Tree, Random Forest, KNN, Naive Bayes, SVM, Logistic Regression, and a neural network, to the "mammographic masses" dataset. Using 10-fold cross-validation, the study aims to identify the most accurate model, with data preprocessing, normalization, and hyperparameter tuning applied to improve performance.
Link to Publication
Master of Science in Computer Science
University of New Brunswick (UNB)
Graduated in 2024
- Focus on Machine Learning, Deep Learning, and Data Analysis
- π§ Machine Learning & Neural Networks: Experience designing and implementing deep learning models for real-world applications.
- π Data Analysis: Expertise in data extraction, cleaning, and visualization using Python, SQL, and Power BI.
- π οΈ Algorithms: Strong foundation in supervised and unsupervised learning techniques.
- π Insights Discovery: Skilled at finding actionable insights from complex datasets.
- π¬ Exploring advanced machine learning techniques for better predictive modeling.
- π Developing custom dashboards to visualize key insights.
- βοΈ Enhancing skills in big data processing and cloud-based tools.
- πΌ LinkedIn
- π GitHub
- βοΈ Email: sharon110699@gmail.com