Analysis and prediction of 'Yuru-Chara' mascot popularity using visual and auditory features
| Autor: | Yuri Nakasato, Toshihisa Tanaka |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Information retrieval
business.industry Computer science Public sector 02 engineering and technology 010501 environmental sciences Machine learning computer.software_genre Object (computer science) 01 natural sciences Popularity Visualization Market research Mascot 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business computer Tourism 0105 earth and related environmental sciences |
| Zdroj: | ICME Workshops |
| DOI: | 10.1109/icmew.2017.8026310 |
| Popis: | Yuru-chara is a special category of Japanese mascots to promote local tourism and other locally based economic drivers that are generally created by local governments or other arms of the public sector. Reports have indicated that the popularity of yuru-charas can strongly affect local economies. This paper presents a framework for analyzing and predicting the popularity of yuru-charas with a limited size of data set (< 2,000). As each yuru-chara has its own name, and this name contributes to its popularity, we use both visual and auditory features from the name of each one to analyze and determine the popularity of a given yuru-chara. During this analysis, we found that bright colors positively contribute to the popularity of a yuru-chara. Popular yuru-charas also have names with strong first-half accents. Furthermore, experimental results show that the combination of visual and auditory features improves the accuracy of popularity-related predictions. This implies that in the case of a single object without background, even though the availability of data is limited, appropriate features works well in the analysis and prediction of the popularity. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|