Dejitaru Akaibu Gakkaishi
Online ISSN : 2432-9770
Print ISSN : 2432-9762
Feature: Sport Digital Archives
Application of AI in the Analysis and Classification of Data Models for the Construction of Sports Digital Archive System
Sumiko YAMAMOTO
Author information
JOURNAL OPEN ACCESS

2020 Volume 4 Issue 3 Pages 280-283

Details
Abstract

Based on the "Sports Digital Archive Frameworkt" of the Sports Agency, Toppan Printing has collaborated with seven organizations that hold sports-related materials since fiscal year 2017 up to 2019, and has released a virtual environment, "public demonstration platform" to demonstrate and prove the usefulness of the sports digital archive. We report on issues in system construction through analysis of data models and data structures of sports materials, construction of dictionary data such as names of people, games, events, shapes, and classifications, a interoperability with other systems. In addition, we report the results of the labor-saving experiment using AI for the labor-intensive classification work, which demonstrated an 80% agreement of results with those of manual classification.

Content from these authors

この記事はクリエイティブ・コモンズ [表示 4.0 国際]ライセンスの下に提供されています。
https://creativecommons.org/licenses/by/4.0/deed.ja
Previous article Next article
feedback
Top