Developed a model which is designed to detect fraudulent transactions, utilizing a Logistic Regression algorithm to significantly enhance performance. The model was meticulously trained on a dataset of financial transactions, resulting in a remarkable accuracy of 93.5% on the training data and 91.37% on the testing data. This level of precision underscores the model's reliability and robustness in identifying fraudulent activities.
To achieve these results, I employed powerful data analysis libraries such as NumPy, Pandas, and Scikit-learn. These tools were integral in data preprocessing, feature selection, and model training, allowing for the effective handling of large datasets and the extraction of meaningful patterns. Additionally, the implementation of optimization techniques ensured that the model not only performed well on the training data but also generalized effectively to new, unseen data.
This project showcases my ability to develop and deploy machine learning models in a web-based environment, leveraging data science techniques to address real-world challenges in fraud detection.
To access the dataset used for Fraud Transaction Detection Model : https://drive.google.com/file/d/1Sg1eOryfcTIA-o_8inw1fQ9rrP51l3uj/view?usp=sharing