An intelligent insect search system based on observation of the insect's structure

摘要

In today's ecological environment, non-experts need insect search systems to identify insect species and to obtain its u-Learning contents. Even though non-experts have been assisted by the naieve web search system based on scientific names, it is very difficult for them to use the system effectively because they are laypersons of insects' scientific names. To assist them more effectively, the ISBC (Insect Search based on Biological Classification) method and ISOBC (Insect Search through Observation based on Biological Classification) method had been proposed. However, the ISBC method requires their time-consuming efforts to search insects because it is due to biological classification based on insects' Order-Family-Species and their scientific names. On the other hand, the ISOBC also gives them troublesomeness to observe insects because it is due to the sequence of biological classification. To overcome such difficulties, we propose a new model, the ISOIS (Insect Search based on Observation of the Insect's Structure) method. It is based on natural observation in the sequence of insect's structure. In addition to that, it is equipped with inference through similarity measure according to the observation attributes by the sequence of biological classification to improve user satisfaction. Finally, we compare it with the ISBC and ISOBC methods. In order to compare the priorities among these three insect search systems, using the AHP method, we derive three evaluation criteria for user satisfaction and three sub-evaluation criteria for each evaluation criterion. In the empirical survey results, we found the order of priorities was ISOIS, ISOBC, and ISBC. This shows that the ISOIS system proposed in this study is comparatively superior to the ISBC and ISOBC systems in usage and quality.