Embedded Files
Project Video.mp4Advancing Real Estate Analytics: A Machine Learning Approach for Price Prediction and Personalized Recommendation
Advancing Real Estate Analytics: A Machine Learning Approach for Price Prediction and Personalized Recommendation
Created a Machine Learning-powered Real Estate Price Prediction and Recommendation System by leveraging data extracted from the widely-used real estate platform, https://www.99acres.com/. This undertaking required managing a challenging dataset riddled with numerous missing values. It involved implementing sophisticated data cleaning techniques and applying machine learning methods for predictive modeling. Additionally, the system tailors recommendations to individual customer interests, providing a personalized experience based on their preferences.
Link: Project Link
Machine Learning-Powered Pair Trading Strategy for High-Frequency Trading
Machine Learning-Powered Pair Trading Strategy for High-Frequency Trading
This project, executed by Tipu Sultan in the context of CISC 5352: Machine Learning in Finance, utilizes MATLAB and machine learning techniques to develop a pair trading strategy for high-frequency trading. Focusing on JPMorgan Chase & Co. (JPM) and BlackRock, Inc. (BLK), the project employs linear, polynomial regression, and Long Short-Term Memory (LSTM) networks to forecast cumulative profit and loss (P&L). Rigorous hyperparameter tuning enhances the LSTM model's accuracy, demonstrating its effectiveness in predicting stock movements. Results highlight the potential of LSTM networks in financial forecasting, offering valuable insights for real-world trading strategy deployment.
Link: Project Link
Email Spam Classifier_.mp4Email Spam Classifier
Email Spam Classifier
Developed an Email Spam Classifier by employing a range of machine learning algorithms such as Logistic Regression, SVM, and Naive Bayes. Demonstrated exceptional accuracy in spam classification, particularly with the Multinomial Naive Bayes model. The project encompassed meticulous data cleaning, exploratory data analysis (EDA), text preprocessing, and comprehensive model evaluation, underscoring proficient expertise in the realms of machine learning and classification.
Link- Project Link
multinomial_Regression.mp4Customer Spending Analysis for Ecommerce Platform
Customer Spending Analysis for Ecommerce Platform
This ML project utilized Multiple Linear Regression to analyze customer data for an Ecommerce platform. Key findings revealed a positive correlation between average session length and yearly spending, emphasizing the importance of user engagement. Time on the app and website were assessed for their impact on spending behavior. Additionally, a strong positive correlation was identified between the length of customer membership and yearly spending.
The insights gained provide actionable recommendations for platform optimization, including refining the mobile app experience, enhancing website features, and fostering customer loyalty for increased revenue. This project exemplifies the efficacy of ML in decoding customer spending patterns to inform strategic business decisions.
Link- Project Link
logistic.mp4Click Prediction with Logistic Regression
Click Prediction with Logistic Regression
This Logistic Regression project focused on predicting ad clicks using a simulated advertising dataset. Features included user information such as daily time spent on site, age, area income, daily internet usage, and more. The analysis aimed to create a model that predicts whether a user would click on an ad based on these features. The practical application of Logistic Regression highlights its relevance in understanding and predicting user behavior for targeted advertising strategies.
Movie Recommendation system
Movie Recommendation system
Engineered a sophisticated Movie Recommendation Model employing advanced techniques such as Content-Based and Collaborative Filtering Approaches. This model adeptly furnishes personalized movie recommendations by leveraging intricate algorithms that analyze both content-related features and collaborative user behavior. The implementation ensures a tailored cinematic experience by seamlessly aligning with the nuanced preferences of the audience.
Link- Project Link
Students' Performance in Exams
Students' Performance in Exams
This project is centered on the examination of students' academic performance through the application of linear regression. The primary objective is to extract meaningful insights into their scholastic achievements, employing statistical techniques to discern patterns and correlations within the data.
Link- Project Link
Page updated
Google Sites
Report abuse