大学サイト   学部サイト   学科サイト   EnglishPages
 
旧尾家・塚本研究室です.尾家祐二 現・九州工業大学学長の情報は こちら.
   
研究業績


A Generation Gap Model for a Human-based Evolutionary Algorithm Using a Tag Cloud
著者
  Masatomo Azumaya and Kei Ohnishi
 
雑誌名/会議名
  Joint 8th International Conference on Soft Computing and Intelligent Systems and 17th International Symposium on Advanced Intelligent Systems (SCIS&ISIS2016), pp.880-850, Sapporo, Japan, August 25-28, 2016. 2017 年 3 月
 
アブストラクト
  We previously proposed a human-based evolutionary algorithm (human-based EA) using a tag cloud and newly propose a generation gap model for the human-based EA in this paper. In the previously proposed human-based EA, people create candidate solutions and the created candidate solutions are displayed as tags in the tag cloud. Tags representing candidate solutions are evaluated by people and tags with higher fitness are displayed using larger fonts in the tag cloud. However, the human-based EA can display only tags less than or equal to a given number in the tag cloud, and if new tags are created after tags of the given number were created, they never appear in the tag cloud. Therefore, we propose a generation gap model for solving this problem. The proposed generation gap model chooses tags with fitness over a given threshold from among the present generated tags and brings them to a next generation as they are. Meanwhile, the remainder are replaced by tags in a queue for storing created but ever undisplayed tags. We carried out the experiment of using the human-based EA including the generation gap model and the experimental results showed that almost all participants could create better tags compared to tags created alone.
 
キーワード
 
 
記述言語
  English
 
 

web_admin [ at ] infonet.cse.kyutech.ac.jp /  Copyright (c) 2016 Oie & Tsukamoto Laboratory, All Right Reserved.