Gajju4

ChurnSense-Smart-Telecom-Churn-Prediction-System

• Created a PowerBI dashboard for telecom churn analysis • Balanced imbalanced data using SMOTEENN, then built accurate Decision Tree (93%) and Random Forest (80%) models for churn prediction. • Deployed the models using Flask, making them accessible for real-time customer churn classification.

Classified-the-rice-samples-based-on-their-quality

Build a model based on Gaussian Naive Model from scratch(without using any library), that classify the rice samples based on their quality and then compare the model's accuracy with the model created using Sklearn library. .

Heart-Disease-Prediction-using-Support-Vector-Machine.

Created a model from scratch (without using any libraries) to predict whether a person have a heart diseases using support vector machine. and then compare the model's accuracy with model created using Sklearn library.

Model-to-Identify-which-cities-may-potentially-give-business-higher-profits

Model-to-predict-whether-a-student-gets-admitted-into-a-University

Obesity-Data-Categorization-Using-Single-Linkage-Divisive-Top-Down-Clustering-Technique

Estimates the obesity levels in individuals from the countries of Mexico, Peru and Colombia, based on their eating habits and physical condition.

portfolio

My portfolio website

Predict-the-Flight-Ticket-Price

Predict-whether-microchips-from-a-fabrication-plant-pass-quality-assurance-QA-

Short-Term-Rain-Forecasting-using-Decision-Tree-based-Learning-Model

Build a decision tree learning model from scratch (Without using Sklearn Library) to predict whether it will rain on the following day using the information from the present day. Got 81.03% accuracy. Also Build Model using Sklearn library and got 81.06% accuracy.

Wheat-seeds-classification-using-ANN

Build a Artificial neural Network (ANN) from scratch i.e. without using any libraries to classify Wheat seeds. Got 90.47% final testing accuracy. Also build model with Sklearn library that classify Wheat seeds and got 88.09% accuracy.

StreamForecaster-Real-time-Stock-Trend-Analysis-and-Prediction

• Developed an LSTM model for accurate stock trend predictions based on tickers with RSME value = 2.23 (14.19%). • Created an interactive Web App using Streamlit to provide data-driven insights into stock behaviour via machine learning techniques.