Correcting Broken Data and AI Analysis... Dong-Eui Institute of Technology Offers Hands-On Data Training

Direct Analysis of Busan Transportation Corporation Ridership Data

Creating Reports and Learning Content Using Generative AI

"Hands-On Practice in Verifying AI-Generated Results"

The practice session went beyond simply entrusting data to AI; students directly corrected erroneous data and verified the analysis results themselves.

DIT AI Hub Center (Director: Jo Subum) of Dong-Eui Institute of Technology (President: Kim Youngdo) announced on October 7 that it had conducted the "AI Data Analysis Practice" session using public data and generative AI for students over two days, September 16 and 17.

This training was organized as part of the 2026 AID Transition-Focused College Support Project. It was conducted with a practical focus so that students could organize and analyze real-world data using generative AI and then apply the results to their major assignments. A total of 37 students from the Department of Health Administration participated in the sessions, with 17 in the first round and 20 in the second round.

Students participating in the AI data analysis practice.

Students used ridership data by station and time from Busan Transportation Corporation. They first carried out data preprocessing, such as fixing broken encodings and sorting out missing values and duplicates.

Afterward, using generative AI tools like Gemini and ChatGPT, they aggregated and compared data by station, time slot, and day of the week. They also organized the analysis results into tables and graphs and created analysis reports in HTML format that included their own interpretations.

They learned methods for not simply accepting the results produced by AI at face value. Students designed the analysis procedures step by step using AI, while also learning verification processes to check for unsupported assertions, causal errors, and logical leaps.

To enable students to continue analyzing data independently after the practice, they made use of starter prompts and checklists that could be applied to a variety of datasets.

The students’ outputs went beyond just analysis reports. They also gained experience by creating question banks and educational game apps using AI, which they could further develop into major assignments and capstone projects.

In a survey conducted immediately after the training, "creating a question bank with AI," "restoring corrupted files," and "making games with AI" were cited as some of the most helpful activities.

Choi Kiwoong, a first-year student (class of ’26) in the Department of Health Administration, said, “I always felt data analysis was difficult, but working with generative AI to organize actual public data and create an analysis report myself gave me confidence.”

He added, “Now that I’ve learned how to verify rather than blindly trust AI outputs, I will actively apply these methods to my future major assignments.”

Jo Subum, Director of DIT AI Hub Center, said, “We focused on enabling students not only to simply use generative AI, but to actually handle real-world data and develop the ability to verify the results themselves. We will continue to supplement practice data and examples tailored to each major, so students can build solid data analysis skills using AI.”

DIT AI Hub Center is operating AI utilization education and major-linked AI programs for students, faculty, and local industry employees through the AID Transition-Focused College Support Project.

This content was produced with the assistance of AI translation services.