Agricultural information retrieval in geographically distributed networks

As the rapid development of information technology and agricultural applications, agricultural information has been widely distributed geographically in a region. It is observed that agricultural production and trading activities tend to be localized in many cases. Effective retrieval of geographically distributed information can facilitate agricultural production as well as trading. In this paper, an agricultural information retrieval approach is proposed for geographically distributed networks. The network structure of the unstructured peer-to-peer network is introduced. A topic-aware agent reputation model that combines individual- and social-level reputations is adopted to distinguish different information retrieval performance on various topics. A location-based network structure optimization method is proposed to gather geographically small distance agents together. A hybrid query routing approach is presented that combines topic-aware agent reputation scores and location-based ratings. Simulation experiments are conducted for the evaluation of the proposed approach. The results of the experiments show good performance of it.