Tamilselvan S Email and Phone Number

Project Title: Machine Learning for IT Service Management (ITSM) EnhancementPredicting High-Priority Tickets: The goal is to predict high-priority tickets (Priority 1 and 2) to enable preventive measures and improved customer satisfaction.Forecasting Incident Volume: ML will forecast incident volumes quarterly and annually, aiding resource planning for efficient incident handling.Auto-Tagging Tickets: Auto-tagging tickets with correct priorities and departments reduces reassignments, leading to faster incident resolution and operational efficiency.Predicting RFC and Misconfigurations: The project aims to predict Request for Change (RFC) and identify potential misconfigurations, enabling proactive measures to prevent service disruptions.

Project Title: Rice Leaf Disease Prediction Using CNN:The primary objective of this project is to build a deep learning model, specifically a Convolutional Neural Network (CNN), that can accurately classify and detect diseases or abnormalities in rice plant leaves based on images.

Project Title: FIFA 20 Data Analysis(Real Time Project)Description:This project involves the comprehensive analysis of football player data from the FIFA video game series (FIFA 15 to FIFA 20). The objective is to gain insights into player attributes, performance trends, positional skills, and to compare machine learning models for predictive purposes.Data Analysis and Exploration: This involves data loading, cleaning, and exploratory analysis to understand the dataset's structure and characteristics.Football Skills Clustering: Players are clustered based on their attributes to identify groups of similar players with distinct skills.Specific Questions Analysis: Specific questions related to player statistics are addressed, such as identifying the top countries producing players, exploring age vs. performance, and comparing earnings based on player positions.Model Comparison: Various machine learning models are compared in terms of performance on the dataset, with a recommendation for the best model for potential predictive tasks.Challenges Faced:Challenges encountered during data preprocessing, analysis, and modeling are documented, along with the techniques and strategies used to address them. This section reflects on lessons learned and insights gained from overcoming challenges.

Automobile Imports Price Prediction:Data Analysis Report: In Task 1, we performed in-depth data analysis on a dataset containing car attributes. Through exploratory data analysis (EDA), statistical analysis, and data visualization, we uncovered patterns and insights within the data, laying the foundation for predictive modeling.Predictive Model for Car Prices: Task 2 involved creating a predictive model using data science techniques and machine learning algorithms. The model accurately predicts car prices and helps in understanding how different attributes impact prices, allowing for informed business decisions.Datasets:The dataset contains various attributes such as make, fuel type, engine specifications, and more, which were used to build the predictive model.Result:The XGBRegressor with hyperparameter tuning outperformed the initial model, reducing mean absolute error and root mean squared error, and increasing the R-squared score, indicating better predictive accuracy. The final model choice is the XGBRegressor with hyperparameter tuning.

Project Title: Portuguese Bank Marketing Data Analysis and Predictive ModelingData Analysis Report: Comprehensive analysis of the provided bank marketing dataset, exploring various attributes and their relationships.Predictive Model: Development of a predictive model using machine learning algorithms to predict whether a customer will subscribe to a term deposit based on their characteristics.Challenges Faced: A detailed report on the challenges encountered during the data analysis and modeling process, along with the techniques used to overcome them.