Fathimathul Susna Shoukkathali
Focus : Data Analytics, Deep Learning
Specifications : Data Analytics, Computer Vision
Skills
SQL, DBMS, Django, Python, Java, Excel, PowerBI, Data Visualization,Hobbies
Dancing, Travelling
Summary
Passionate Data Analyst from Kerala with proven Machine Learning and Analytics Skills. Currently a final year Master’s student in CHRIST (Deemed to be University) Lavasa. Seeking an opportunity to learn.
Education History
Internships
Artificial Intelligence and Machine Learning Product related testing and support
Pixel Level Segmentation on Autonomous Driving vehicles.
As a part of this Internship, I learned about Machine Learning Models and worked on different Regression, Classification, and Clustering projects.
Projects
In this project, I collaborated with a team of three to perform pixel-level scene segmentation on images of autonomous vehicles. We developed a model capable of accurately segmenting images into distinct objects, enhancing the understanding of the driving environment.
The 'Hungry Student' web application, built using the Django framework, facilitates food ordering with integrated frontend and backend components. It leverages an SQL server for data management, enabling users to browse menus, place orders, and manage accounts through an intuitive interface designed for ease of use and efficiency.
This research focuses on developing hybrid models using self-supervised learning (SSL) with an early fusion approach.Models can acquire meaningful representations from unlabeled data by utilizing SSL, which lessens the need for annotated datasets.The objective is to integrate data features early on for more effective processing and better performance, combining different neural architectures to improve accuracy in tasks like sentiment analysis and emotion recognition.
In the Aurasync project, I am working alongside a team member to integrate multimodal approaches thus providing more accurate emotion detection system. The aim is to combine various data types, By combining different input sources such as speech and facial expressions, the system will be better able to recognize and comprehen a wider range of emotional states. Through this our goal is to raise the process overall robustness and accuracy in detecting emotions.