Academic Projects

Capstone Project - Predicting Success of Bank Telemarketing

Conducted comprehensive data cleaning and preprocessing tasks using both R and Python. This involved data mining, exploratory data analysis, handling missing values, removing duplicates, addressing outliers, and transforming data into a suitable format for analysis and modeling.
Employed advanced techniques to select relevant features that have a significant impact on the success of bank telemarketing calls. Utilized domain knowledge and statistical analysis to engineer new features that capture valuable information from the available data.
Developed and implemented machine learning models, including Decision Trees (DT), Random Forest (RF), and Perceptron, to predict the success of bank telemarketing calls.
Conducted thorough evaluation and interpretation of the machine learning models' performance. Utilized metrics such as accuracy, precision, sensitivity, specificity, G-mean, and area under curve (AUC) to assess model effectiveness. Interpreted model outputs and provided insights into the factors contributing to the success or failure of bank telemarketing calls.

Designed and developed a fully functional Codenames game using Python and its libraries (SpaCy, Gensim). Implemented text pre-processing, game logic, user interfaces, and interactive features to provide an engaging gaming experience.
Leveraged transformers and pre-trained models (GloVe) in the project to enhance the game's functionality and user experience. Utilized NLP techniques to generate clues, analyze word associations (Cosine Similarity), and provide intelligent hints to players.
Focused on user-centric design and user experience (UX) principles to create an intuitive and visually appealing interface for the game. Implemented user-friendly functionalities such as input validation, game progress tracking, and clear instructions to enhance usability and ensure a seamless gaming experience.