Naveen Krishna

The project's goal is to create a system that can forecast a user's risk of developing heart-related problems down the road. Python-based Django was used to develop the complete system. The UCI repository provided the dataset, while the Pandas and NumPy tools were used for data preprocessing. Additionally, the project has an easy-to-use interface that lets people register or log in under a hospital and view their previous or anticipated outcomes.

There are three main modules in the system:
1. Users

2. Super Admin

3. Hospitals

Using the testing dataset, the predictive model yielded an accuracy of 89%.

The project developed a hostel management system for a university to manage and allocate rooms or villas to students during admission. It includes four modules: Super Admin, University Staff, House Owners, and Students. University staff manually register students, who can then browse available rooms by street name and request accommodation. Requests are first approved by staff and then forwarded to house owners, who can accept or reject them. Upon acceptance, the student is allocated the room. House owners must complete a verification process with property documents to register onto the site. A detailed analysis of the web site is integrated with the admin panel of the site. The analysis shows the amount of students accomodated around the city, and many more.

AuraSync is a multimodal emotion detection system that integrates both computer vision and natural language processing (NLP) models to provide real-time emotion analysis. Developed as a web application using Django and SQLite, the project uses a CNN model to analyze facial expressions and an NLP model to assess sentiment from speech-to-text conversion. These models work together to capture and evaluate a user's emotions throughout the day, generating a comprehensive emotional report. AuraSync includes a user-friendly interface and aims to provide deeper insights into a user's emotional state for personal well-being or broader industry applications.

My work focusses on image captioning, which uses deep learning methods to extract descriptive text from photos. The main goal of the project is to create an automated system that connects computer vision with natural language processing (NLP) by producing precise and insightful captions for photographs. The research specifically attempts to produce Malayalam captions by using models that comprehend the image's visual information as well as the subtleties of the target language.

The goal of this project's analysis is to comprehend how socioeconomic issues affect students' academic achievement. Finding patterns and relationships between factors such as parental education, family size, study time, and grades is done through the use of exploratory data analysis, or EDA. These correlations are better illustrated by visualizations like scatter plots and bar charts. In order to forecast student grades after EDA, a variety of machine learning models are used, including XGBoost, Decision Tree, Support Vector Machine, and Linear regression. With an R-squared of 0.216802, XGBoost demonstrated the highest performance, whereas Linear Regression provided meaningful insights. In accordance with SDG 4, the analysis seeks to offer practical suggestions for enhancing educational outcomes and promoting inclusive and equitable schooling.