Display Accessibility Tools

Accessibility Tools

Grayscale

Highlight Links

Change Contrast

Increase Text Size

Increase Letter Spacing

Readability Bar

Dyslexia Friendly Font

Increase Cursor Size

Innovative Research in Causal Relationship Graph Extraction by CMSE Students

The MSU Department of Computational Mathematics, Science and Engineering (CMSE) continues to push the boundaries of data science research through its course CMSE 495 - Experiential Learning in Data Science. This undergraduate course operates more like an internship, where students collaborate in teams to deliver a product or service to a community partner. One of the standout projects from this course has recently been published, showcasing the innovative work of CMSE students—a rare achievement for an undergraduate class.

The research paper, authored by a CMSE graduate student and multiple students from both the master’s program and the CMSE 495 course, presents a novel approach to extracting causal relationship graphs from text using machine learning models. The master’s students involved are part of the MSDS Master's of Science in Data Science program, a joint collaboration between the departments of Statistics, CMSE, and Computer Science and Engineering (CSE). This interdisciplinary program equips students with a robust foundation in data science, combining statistical theory, computational techniques, and practical applications.

Social science has long sought to understand and model the collective intelligence underlying humanity’s most pressing problems. A key tool in this endeavor is the causal relationship graph, which represents the increase or decrease of relationships between concepts. These graphs are crucial for encoding complex systems and understanding the dynamics within them.

The project focuses on developing methods to extract these graphs from textual data. The quality of the extracted graphs is measured using both computational algorithms and manual input from humans. The study’s findings emphasize the necessity of developing more specialized measures that align better with human judgment and capture the complexities of causal relationship graph extractions. This research highlights the potential for machine learning models to enhance our understanding of complex social systems and improve the accuracy of causal relationship representations.

The collaboration between CMSE and TwoSix Technologies exemplifies the power of academic and industry partnerships in driving innovation and providing students with invaluable real-world experience. As CMSE 495 continues to offer students opportunities to work on impactful projects, the department remains committed to fostering an environment where education, research, and professional development intersect. This project is just one of many that highlight the potential of our students to contribute meaningfully to the field of data science and beyond.