Project Details

Methodology:
1.Data Preprocessing:Clean and preprocess the dataset, handling missing values and encoding categorical features. Perform feature scaling for numerical variables.
2.Model Development:Utilize Random Forest and XGBoost algorithms for training predictive models. Evaluate models using performance metrics like RMSE, MAE, and R².Optimize hyperparameters using GridSearchCV for better accuracy.
3.Deployment:Build an interactive Streamlit web application where users can input relevant details to get an estimated insurance premium prediction.
4.Technologies & Libraries Used: Python, Pandas,NumPy for data handling,Scikit-learn for preprocessing and model training ,XGBoost,Random Forest for prediction,Matplotlib,Seaborn for visualization
Streamlit for deployment
This project can help insurance companies offer personalized premium plans and assist individuals in understanding their expected costs based on medical and financial attributes.

Project Details

This ML-driven Credit Risk Assessment project predicts loan default probability using key financial indicators like DPD (Days Past Due), loan amount, tenure, age, credit utilization ratio, loan type, loan purpose, FOIR (Fixed Obligation to Income Ratio), LTV (Loan-to-Value), and bureau score. Advanced feature selection techniques (SHAP values, recursive feature elimination) ensure the most impactful variables are used, optimizing model performance. Cutting-edge ML models like XGBoost, Random Forest, and Logistic Regression are trained with hyperparameter tuning to enhance accuracy. The final model is deployed via Streamlit, providing an interactive web app where lenders can input borrower details and receive real-time credit risk scores. This solution empowers financial institutions with data-driven insights, enabling better lending decisions, reducing default rates, and improving risk management strategies. 🚀📊

Project Details

In a project,I tackled several net sales issues by leveraging advanced SQL techniques.I identified discrepancies and optimized the accuracy of financial data.By utilizing SQL joins,I integrated data from multiple tables to create a unified view of sales,returns,and discounts.SQL views were designed to streamline access to clean, consolidated datasets for reporting purposes.I also implemented triggers to automate updates and maintain data consistency,ensuring real-time accuracy across sales-related tables. For in-depth financial analytics,I employed SQL window functions like ROW_NUMBER,RANK,and SUM to calculate running totals, rank sales by region,and analyze trends over time.These window functions were key to breaking down data into segments for better insights into revenue patterns.My solutions improved net sales visibility, provided actionable insights,and enhanced financial reporting efficiency for the company,contributing to data-driven decision-making and long-term profitability.

Project Details

AtliQ Hardware, a rapidly growing consumer electronics company, faces limitations with Excel for data analysis, particularly affecting their performance in the Latin American market. To address these challenges, the company plans to implement Power BI, a powerful business analytics tool, to enhance their data-driven decision-making. Power BI will enable AtliQ to efficiently collect, process, and visualize large datasets from multiple sources, transforming raw data into actionable insights. The solution will integrate data from various business functions, allowing real-time dashboards and interactive reports to monitor market trends, sales performance, and financial metrics. By leveraging Power BI’s advanced analytical capabilities, AtliQ aims to improve forecasting, optimize operations, and regain profitability in Latin America. The goal is to move beyond the limitations of Excel, streamline data workflows, and empower teams to make well-informed, timely decisions.

Live Interactive Dashboard

Project Details

Disease Prediction Using Machine Learning and Deep Learning is a system that predicts the disease based on the information provided by the user. It also predicts the disease of the patient or based on the information he/she enters into the system and provides accurate results based on it.The system is built using Python language along with its suitable modules such as TensorFlow and Keras. Therefore, disease prediction systems that use machine learning algorithms assist in cases which require a second opinion to get accurate results .A sequential CNN model is used for the prediction of the chosen diseases.