Hewlett Packard Enterprise Data Science Institute

The Department of Computer Science and the Hewlett Packard Enterprise Data Science Institute hosted the Summer AI Student Showcase on August 2. The event brought together graduate research teams from across the University of Houston to share innovative projects in artificial intelligence (AI), machine learning and data science. 

Six teams were invited to present, and three winners were chosen by the judging panel. The projects spanned a wide range of applications, showcasing the versatility and potential of AI in addressing complex challenges.  

Following the success of the Data Science Showcase for undergraduate students, the AI Showcase offered a platform for graduate students to present their research and engage with peers and faculty. Nouhad Rizk, director of undergraduate studies for the UH Department of Computer Science, and Claudia Neuhauser, director of the HPE DSI, led the organizing efforts. 

“We are pleased to see the level of dedication that these students have brought to this showcase,” said Rizk. “This event underscores the important contributions of our graduate students and gives them a place to promote their research and its applications.” 

First Place 

Team Lead: Saikiran Anugam, Department of Engineering Data Science 

Team Members: Hariharan Annadurai, Department of Chemical and Biomolecular Engineering 

Project: Anugam and Annadurai investigated the relationship between chemical compositions and mechanical properties— such as tensile and yield strength—  of industrial steels using machine learning and neural network models. Their research offers insights for optimizing manufacturing processes and improving material performance, with advanced models and robust error analysis enhancing the reliability of their findings. 

Second Place  

Team Lead: Divija Kalluri, Computer Science Department(NSM) 

Team Members: Charan Gajjala Chenchu, Computer Science Department  

Project: Kalluri and Chenchu proposed a method to improve X-ray interpretation accuracy using the DenseNet deep learning architecture. By combining segmented image features with diagnostic prompts and processed report data, they trained a large language model to generate concise medical reports. This system enhances radiologists' efficiency by automating image interpretation and report generation, providing a reliable tool for clinical diagnostics. 

Third Place 

Team: Rabimba Karanjai, Computer Science Department 

Project: Karanjai’s AI-Powered Education Diary leverages advanced large language model technology to generate personalized code examples and study materials grounded in authoritative sources like textbooks and lecture notes. Continuous user feedback will refine the content, resulting in an intuitive, multi-device application that enhances personalized learning.