SUSMITHA PRIYADARSHINI PATLURI

As a recent graduate with a Master's degree in Business Analytics, I am enthusiastic about applying my analytical skills and knowledge to contribute to a dynamic organization. I am eager to learn and grow while delivering data-driven insights for strategic decision-making.

CALORIES BURNT PREDICTION

Utilizing the available dataset, a prediction model has been constructed to enable users to assess the calories burned during physical activities. Beyond exercise duration, several critical factors such as age, height, weight, heart rate, and body temperature contribute significantly to the accuracy of calorie calculations. The project involves importing necessary libraries, cleaning and understanding the data, followed by visualization. Appropriate model selection is crucial, with training and testing phases executed subsequently. Evaluation metrics like root-mean-squared error (RMSE) and mean-squared error (MAE) play key roles in regression, which involves predicting continuous values. Linear Regression and Decision Trees are common regression algorithms. Notably, XGBoost models include important components such as RMSE and MAE, each serving a vital function in quantifying prediction accuracy.

EFFECTIVENESS OFDATA CLEANING

In the context of a growing reliance on data-driven decision-making, data quality and accuracy have become paramount. Data cleansing, or data cleaning, is the process of improving data integrity by detecting and rectifying errors and inconsistencies. It aims to ensure that data meets business requirements and standards through a combination of people, technology, and processes. With the proliferation of data, maintaining its quality is challenging, and while various approaches exist to clean and enhance data, the adaptation of these methods to the unique demands of big data, as well as considering data value and domain expertise, remains a persistent challenge.

AIRLINE PASSENGER SATISFACTION PREDECTIVE ANALYSIS

In the competitive landscape of the airline industry, prioritizing passenger satisfaction is crucial, and customer feedback serves as a key performance indicator. This project aims to assess the significance of various features in contributing to passenger satisfaction. By leveraging data from customer satisfaction surveys, the notebook seeks to predict the pivotal attributes for a content passenger, such as seat comfort, check-in service, food and drinks, inflight entertainment, and cleanliness. Utilizing a Random Forest model, the process involves generating random samples, constructing decision trees, collecting votes from multiple trees, and deriving a prediction based on the most popular feature attributes, enabling airlines to identify crucial elements for improvement and enhanced customer experience.