Home // INTELLI 2016, The Fifth International Conference on Intelligent Systems and Applications // View article

Species Pattern Analysis in Long-Term Ecological Data Using Statistical and Biclustering Approach

Authors:
Hyeonjeong Lee
Miyoung Shin

Keywords: long-term ecological data, association mining, visualization, species-set, species abundance pattern

Abstract:
Analyzing long-term ecological data and appropriate visualization techniques are important for understanding biodiversity mechanisms and predicting effects of environmental changes. In this study, we applied an unconventional approach of finding species pattern, the tendency of species abundance monthly and annually in long-term ecological data, by using statistical and biclustering methods. We tended to find out the similarity between each species after summarizing long-term dataset, and then visualized a correlation matrix and network, which exhibit significant statistical association with each other. For detecting species sets frequently appearing together or showing similar variation in abundance, we also employed a clustering based association mining. For experiments, we used weekly abundance butterfly data from the Environmental Change Network (ECN) in the UK. We could find out how often sets of species show the repeated pattern in long-term species abundance data. The approaches we have described can enable researchers to gain insight of many other relationships like between various species and environmental factors. In addition, combining our methods with detailed analyses or assumptions, such as genetic associations between species and functional subsystems may especially be effective in further analysis.

Pages: 30 to 32

Copyright: Copyright (c) IARIA, 2016

Publication date: November 13, 2016

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-518-0

Location: Barcelona, Spain

Dates: from November 13, 2016 to November 17, 2016