An Effective Approach to Predicting Plant Species in an Ecological Environment

摘要

Using data mining method in the ecosystem modeling, the correlation between a plant and certain ecological environment can be discovered. This is helpful for the reconnaissance and the exploitation of rare plants. Based on the semantic proximity, a mining method to evaluate the fuzzy association degree is given in this paper. Inverse document frequency (IDF) weight function has been adopted in this investigation to measure the weights of ecological environments in order to superpose the fuzzy association degrees. To implement the method, the "growing window" and the proximity computation pruning are introduced to reduce both I/O and CPU costs in computing the fuzzy semantic proximity between time-series. Extensive experiments on real datasets are conducted, and the results show that the mining approach is reasonable and effective.