Presentation | 2023-03-02 [Poster Presentation] Clothes recommendation system based on user's latent interests extracted from vectors of clothes images Daishi Tanaka, Michiharu Niimi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | As the fashion and apparel industry is becoming more and more e-commerce, if you want to buy clothes online such as e-commerce sites, you have to find clothes that suit your taste from a huge number of clothes. It would be convenient if there was a recommendation system that presents recommended products to the user. One of sightseeing spot recommendation systems use word2vec to vectorize words and calculate the vectors to discover latent interests that users have not noticed before. In this paper, we apply this idea to a clothing image-based recommendation system for purchasing clothes. Firstly, the feature space of the clothing image is defined based on three evaluation items, and then the images are vectorized in the space. A piece of clothes in the real world is correspond to a point in the feature space. If a person has multiple clothes, there are multiple vectors, and the average vector of the multiple vectors is defined as latent interest. Clothing recommendation in this research means to recommend clothes in the real world that are similar to latent interests in the feature space. In order to evaluate this recommendation method, we collected clothes images on the Internet and conducted experiments. We confirmed that the recommendation is more efficient than the recommendation that presents clothes images at random. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | recommendation system / clothes image vector / latent interests |
Paper # | EMM2022-75 |
Date of Issue | 2023-02-23 (EMM) |
Conference Information | |
Committee | EMM |
---|---|
Conference Date | 2023/3/2(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Fukue culture hall |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Ryoichi Nishimura(NICT) |
Vice Chair | Kotaro Sonoda(Nagasaki Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.) |
Secretary | Kotaro Sonoda(Kaishi Professional Univ.) / Masatsugu Ichino(Chiba Univ.) |
Assistant | Tomoko Kajiyama(Hiroshima City Univ.) / Shieyuki Sakazawa(Osaka Inst. of Tech.) |
Paper Information | |
Registration To | Technical Committee on Enriched MultiMedia |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Poster Presentation] Clothes recommendation system based on user's latent interests extracted from vectors of clothes images |
Sub Title (in English) | |
Keyword(1) | recommendation system |
Keyword(2) | clothes image vector |
Keyword(3) | latent interests |
1st Author's Name | Daishi Tanaka |
1st Author's Affiliation | Kyushu Institute of Technology(KIT) |
2nd Author's Name | Michiharu Niimi |
2nd Author's Affiliation | Kyushu Institute of Technology(KIT) |
Date | 2023-03-02 |
Paper # | EMM2022-75 |
Volume (vol) | vol.122 |
Number (no) | EMM-412 |
Page | pp.pp.44-49(EMM), |
#Pages | 6 |
Date of Issue | 2023-02-23 (EMM) |