Associative retrieval in spatial big data based on spreading activation with semantic ontology

The opportunities associated with big data have helped generate significant interest, and big data analytics has emerged as an important area of study for both practitioners and researchers. For example, traditional cause-effect analysis and conditional retrieval fall short in dealing with data that are so large and complex. Associative retrieval, on the other hand, has been identified as a potential technique for big data. In this paper, we integrate the spreading activation (SA) algorithm and the ontology model in order to promote the associative retrieval of big data. In our approach, constraints based on variant weights of semantic links are considered with the aim of improving the spreading-activation process and ensuring the accuracy of search results. Semantic inference rules are also introduced to the SA algorithm to find latent spreading path and help obtain results which are more relevant. Our theoretical and experimental analysis demonstrate the utility of this approach.