Avantika Nair

Highly motivated and detail-oriented candidate passionate about leveraging data to develop actionable project solutions and utilizing data mining and data visualization to create meaningful insights. Proficient technical aptitude and knowledge of Python for data analytics and data visualization. Keen interest in applying Machine Learning and Artificial Intelligence in future projects.

Identification of Addictive Medications (focus on Machine Learning Model), Winter 2024, 

• Analyzed a dataset comprising 250,000+ entries and 58 attributes with the Pandas package, conducting detailed exploratory data analysis to extract crucial insights on medication usage, side effects, chemical compounds, and functions
• Created comprehensive graphs utilizing Seaborn and Matplotlib Packages to illustrate the patterns and relationships among addictive and non-addictive medications
• Constructed, trained, and cross-validated three classification models utilizing the SciKit-Learn Package to predict medication addiction probabilities with 90% accuracy; providing healthcare providers and patients with critical insights to prevent potential dependencies
• Identified Logistic Regression as the optimal model with 96.4% accuracy; crafted and analyzed a detailed confusion matrix to validate results further and refine model parameters
• Project Code: https://github.com/4vantikanair/Project/blob/main/Medication_and_Addiction.ipynb

Analysis of Olympic Athlete History (focus on EDA), Summer 2021

• Conducted comprehensive exploratory data analysis utilizing Pandas package on a dataset of 250,000 entries and 15 attributes to identify key factors influencing gold medal attainment
• Applied Seaborn for data visualization of results in Jupyter Notebook
• Explored basic machine learning by creating a classification tree model using results through the Scikit-learn package and achieved an accuracy of around 65%
• Project Code: https://github.com/4vantikanair/OlympicsProject.git