Fatemeh Jamshidi

Fatemeh Jamshidi

Assistant Professor , Computer Science, College of Science

Current and Former Lab Members

Current Graduate Students (Thesis/Project Supervision)

Name Thesis/ Project Title Description Project Link

Bill Kim

Emotion Extraction from K-Pop Korean Songs

Uses AI to analyze lyrics, vocal tone, and musical elements to reveal the emotional depth in K-Pop, deepening fan connections to the music.

 

Trung Vu

Communicating Flood Risk and Mitigation using Augmented Reality

Since Hurricane Katrina in 2005, the National Flood Insurance Program has faced deep financial strain, losing over $1.4 billion annually. Climate change, urbanization, and rising sea levels have intensified flood impacts—especially for socially and economically vulnerable communities. In cities like New Orleans, centuries of racial and economic injustice have pushed marginalized groups into flood-prone areas with limited protection.Our project addresses these inequities by developing an interactive web portal and augmented reality (AR) tool that assesses property-level flood risks and suggests optimal mitigation strategies. Using GIS data and spatial models, the tool visualizes patterns of environmental vulnerability and helps residents, professionals, and policymakers make informed, equitable decisions.By scaling this solution globally, we aim to confront the disproportionate impacts of climate change on the Global South and marginalized populations worldwide—advancing resilience through data-driven environmental justice.

 

Keita Katsumi

Personalized Music Recommendations Using Deep Learning and User Behavior Analysis

This project develops a personalized music recommendation system that integrates deep learning with user behavior analysis. It leverages neural networks to process audio features (e.g., tempo, genre, rhythm) and combines them with behavioral data such as listening history and interaction patterns to model user preferences.The system is trained on a custom dataset generated by Dr. Jamshidi’s team in collaboration with iPalPiti, specifically curated to promote young performers. By merging content-based and behavior-based insights, the system aims to overcome common challenges like popularity bias and cold-start issues, ultimately delivering fair, personalized music recommendations and supporting emerging talent.

 


Current Undergraduate Students (Senior Project Supervision)

Name Project Title Description

Dhruv Bhatnagar

Generative Music AI for Game

Real-time adaptive AI soundtracks that change with gameplay, player actions, and moods—creating fresh, non-repetitive musical experiences.

Farid Vakili

Machine Learning and Generative AI for Personalized Sound and Visual Therapy in Palliative Care through Extend Reality

Palliative care aims to provide relief from the symptoms and stress of serious illness, improving the quality of life for both patients and their families. Integrating therapeutic elements such as music, tonal sounds, and engaging visuals has proven to offer calming and comforting effects, reducing pain perception and enhancing emotional well-being. Recent advancements in machine learning (ML) and generative AI offer opportunities to tailor these therapeutic interventions to individual needs. When coupled with immersive Extended Reality (XR) environments, patients can potentially experience highly personalized and adaptive care, improving both emotional and physiological outcomes.This research proposal outlines a framework to develop and evaluate a data-driven, AI-powered therapeutic platform that uses generative algorithms to deliver custom sounds, music, and visual stimuli within an immersive XR environment for palliative care patients.

Hans Jeremy

Adaprive Game NPC AI

Artificial intelligence and machine learning has become quite popular recently from continuous advancements in recent years. With these advancements, this technology could be implemented in an industry that can unlock new possibilities: games. The purpose of this project is to implement current machine learning techniques and methods to advance game AI to create an adaptive non-playable character. The research involves finding and collecting different data the player can show to the AI, which include the performance and emotions of the player. The AI will then learn from the different models and methods of machine learning and adapt accordingly to either challenge or help the player, depending on the NPC’s goals and the player’s condition.


Alumni – Graduate Students

Name Thesis/Project Title Description Project Link

Dhruvi H. Choksi

VR Prototype for Reducing Motion Sickness

Explores motion sickness in VR using Spider-Man-style swinging, Iron Man jetpacks, and immersive mechanics to create a more comfortable experience.

 

Amir Mohideen Basheer Khan

MindPool: VR Mind-Mapping Tool

Turns physical space into a limitless brainstorming canvas with holographic notes, AI integration, and internet connectivity.

 

 

Rubayet Mujahid

  



 

Integrating Amesim with Unity to visualize flight Path

This project presents an innovative integration of Amesim's robust simulation capabilities with Unity's immersive 3D visualization environment to facilitate the analysis of aircraft flight paths. The core challenge addressed is the disconnect between sophisticated engineering simulation and accessible, interactive visualization, particularly for users without deep technical expertise. The developed system employs a Python-based interface to mediate communication: user-defined flight parameters from a web-based interface trigger Amesim simulations; the resulting CSV data, detailing flight dynamics, is then processed and rendered as a dynamic 3D flight path in Unity. This methodology enables intuitive exploration of various flight scenarios and aims to enhance the understanding of complex aircraft behaviors. The research contributes a streamlined workflow for aerospace simulation and visualization, offering a valuable tool for educational, training, and preliminary design analysis by bridging detailed simulation with user-friendly graphical representation.