Evolving association streams

详细信息    查看全文

• 作者：Andreu Sancho-Asensio ; ^a ; ^{andreu.sancho@gmail.com" class="auth_mail" title="E-mail the corresponding author} ; ^{andreus@salleurl.edu" class="auth_mail" title="E-mail the corresponding author} ; Albert Orriols-Puig^a ; ^{aorriols@salleurl.edu" class="auth_mail" title="E-mail the corresponding author} ; Jorge Casillas^b ; ^{casillas@decsai.ugr.es" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Online learning ; Data stream ; Association rule ; Concept drift ; Genetic fuzzy systems
• 刊名：Information Sciences
• 出版年：2016
• 出版时间：20 March 2016
• 年：2016
• 卷：334-335
• 期：Complete
• 页码：250-272
• 全文大小：1652 K

文摘

The increasing bulk of data generation in industrial and scientific applications has fostered practitioners’ interest in mining large amounts of unlabeled data in the form of continuous, high speed, and time-changing streams of information. An appealing field is association stream mining, which models dynamically complex domains via rules without assuming any a priori structure. Different from the related frequent pattern mining field, its goal is to extract interesting associations among the forming features of such data, adapting these to the ever-changing dynamics of the environment in a pure online fashion-without the typical offline rule generation. These rules are adequate for extracting valuable insight which helps in decision making. This paper details Fuzzy-CSar, an online genetic fuzzy system designed to extract interesting rules from streams of samples. It evolves its internal model online, being able to quickly adapt its knowledge in the presence of drifting concepts. The different complexities of association stream mining are presented in a set of novel synthetic benchmark problems. Thus, the behavior of the online learning architecture presented is carefully analyzed under these conditions. Furthermore, the analysis is extended to real-world problems with static concepts, showing its competitiveness. Experiments support the advantages of applying Fuzzy-CSar to extract knowledge from large volumes of information.