Sayan De's Projects
Developed an ad-click prediction model using Logistic Regression to forecast user interactions with online advertisements. Analyzed features such as user behavior and demographics to predict click-through rates. The project includes data preprocessing, model training, and evaluation to enhance ad targeting and marketing strategies.
This project uses machine learning to predict AIDS virus infection with 95% accuracy. By applying logistic regression and random forest algorithms, it involves data preprocessing, feature selection, model training, and evaluation. Comparing these models will identify the most effective method, aiding in early detection and treatment strategies.
Description: Explore the dynamic landscape of Airbnb listings in New York City through our interactive Power BI dashboard. Gain valuable insights into trends, pricing dynamics, occupancy rates, and more.
Analyze Amazon sales metrics effortlessly with Tableau. Visualize sales amount, quantity, geography, shipping modes, and more for informed decision-making.
The Anime Insights project aims to delve into the world of anime through data analysis, leveraging Python libraries such as Pandas, Matplotlib, and Seaborn. Anime has become a global phenomenon, with a diverse range of genres, styles, and themes captivating audiences worldwide. This project seeks to uncover insights into characteristics of anime
This project aims to build an advanced book recommendation system by integrating collaborative filtering, content-based filtering, and machine learning. It offers tailored suggestions based on user preferences and interactions, using EDA for insights and cosine similarity and SVD for precise recommendations.
Utilizing ML techniques, the project builds a predictive model for housing prices, leveraging diverse features like location, size, amenities, and neighborhood details. Using a rich dataset, it aims to deliver a precise and insightful tool for real estate professionals.
Built a cancer detection model using Support Vector Machine (SVM) classifiers. Utilized SVM to analyze patient data and classify cancerous vs. non-cancerous cases with high accuracy. Includes data preprocessing, model training, and evaluation to support early detection and improve diagnostic outcomes.
Analyze trends yearly, vaccine distribution, infections, recoveries, deaths. Empower decisions with real-time data on Tableau dashboard.
Autoencoder, ModelCheckpoint
The Smart Crop Recommendation System utilizes machine learning to suggest optimal crops based on soil and environmental factors like Nitrogen, Phosphorus, Potassium, Temperature, Humidity, pH, and Rainfall, aiding farmers in maximizing yield and sustainability.
Leveraging diverse customer data, we analyze interactions to uncover trends and anomalies. Through rigorous data cleaning and advanced techniques like clustering, we segment customers for personalized strategies, enhancing competitiveness in today's market