The 2024 IEEE International Conference on Big Data (IEEE BigData 2024) is held in Washington D.C., USA. from December 15 to 18, 2024, and we presented our research results.

In this study, we analyzed the “university atmosphere” by applying a topic model to the visitors’ interests in urban areas based on search queries and GPS logs. By revealing the characteristics and school culture of a university, the so-called university atmosphere, it may be possible to contribute to the promotion of the university and the development of facilities for the university’s users. Therefore, we assume that the users or visitors affect universities’ atmospheres, and we identify the atmosphere based on what university’s visitors usually think about.

In our analysis, we use a method that reveals the visitors’ interests in each area by combining the interests revealed by search queries and the destinations revealed by GPS logs. Furthermore, by using a topic model which is developed for extracting topics that compose documents from document words, we extracted the atmosphere that composes a university from the visitors’ interests.

In our experiments, we used a dataset based on actual search queries and GPS Logs. We discussed the potential demand in the surrounding area based on the extracted atmosphere and the composition of each university’s atmosphere.

For the detailed contents of this research project, please see the corresponding project page.

— Presentation information —

Poster/Demo Session (Mon., Dec. 16, 2024, 10:00 – 10:30, 16:00 – 16:30 EDT) (Program)

Revealing Universities’ Atmosphere from Visitor Interests Using Search Queries and GPS Logs
Kaoru Miyanaga*, Soto Anno*, Kota Tsubouchi**, Masamichi Shimosaka*
*: Institute of Science Tokyo, **: LY Corporation