[2Q6-GS-11] AI and Society:

[2Q6-GS-11-01] Detection of Attribute Changes Based on the Order of Rental Properties

〇Tetta Noshiro¹, Hikaru Saito¹, Kiyone Kambe¹, Sato Wataru¹, Yuki Yamagishi^1,2,3, Airi Tanaka³, Shigeki Aoki⁴, Masahiro Hashimoto⁵ (1. Shizuoka Institute of Science and Technology, 2. RIKEN, 3. Shizuoka City, 4. MaOI, 5. Hosei University)

Keywords:Land Prices, Geographic Information, Rank Sum Test

Generally, by referencing search results from rental information sites or the official land prices, one can estimate the rent levels of various regions to a certain extent. Further, if one wishes to estimate more refined rent levels or the characteristics of properties that affect prices, there are supervised learning approaches based on prices. However, it is presumed that comparing and considering the geographical limitations with rent and attributes such as layout still involves considerable effort, such as how property features change with the distance from a station or how altering the rent can change the layout. This study aims to facilitate decision-making in such comparative considerations by utilizing the ordinality of distances from stations and rents, as well as the categorical attributes of layouts, etc., within the framework of rank-sum testing for information visualization. Additionally, the proposed analytical method demonstrates its applicability not only to attributes originally in categorical values but also to discretized numerical attributes such as exclusive area sizes and ages of buildings.