Information Retrieval Technology Based on Knowledge Graph
Ce Wang, Hongzhi Yu, Fucheng Wan · Proceedings of the 2018 3rd International Conference on Advances in Materials, Mechatronics and Civil Engineering (ICAMMCE 2018) · 2018
In the big data environment with the rapid development of the Internet, the dependence on information search is becoming stronger and stronger.At present, full text search based on keywords has been difficult to satisfy people's search needs.In this case, an information retrieval method based on knowledge graphs is proposed.Through a self-supervised open Chinese relation extraction method, the knowledge of knowledge graphs is extracted from large-scale unstructured data in the Internet, and the knowledge graphs are constructed based on related domain knowledge bases.Based on the knowledge graph, information retrieval is performed through the calculation of semantic similarity.Using this technology for information retrieval, the efficiency and accuracy of the retrieval results will be greatly improved, and it has a very good application value in the field of information retrieval and smart recommendation. PrefaceWith the rapid spread of the Internet, the rapid increase in the volume of digital information data has brought us a wealth of valuable information data.Although these data have been classified and managed, effective information has been retrieved from thousands of data for search engines.It is also a great challenge.In the age of big data, the massiveness, heterogeneity, dynamics, and diversity of Web data have become the major challenges facing information retrieval.The traditional information retrieval is to index the content of the main webpage through the keyword searched by the user and feed back the relevant webpage link to the user based on the keyword in the matching user's search request.This search mode brings great convenience to Internet information retrieval.However, this model has a big drawback.That is, the results returned by the search engine are in a single form.It is impossible to directly provide accurate information based on the user's search request.The user still needs to continue to search for the required information in the web page according to the provided link.In response to this problem, we propose an information retrieval technology based on knowledge graphs.This technology implements entities by further performing entity information mining on Web page content and through an open-source relational extraction method based on self-supervised learning in machine learning.The extraction of the synonymous relationship, the upper-lower relationship, and the attribute relationship between (concepts), the construction of knowledge graphs through the relationship between entities.The information retrieval based on the knowledge graph is to construct the relationships between the entities and index these data and relationships.At the same time, the use of the entity-based retrieval tools in the semantic aspects is timely.The application of the knowledge graph in the information retrieval makes the search engine more efficient.A good understanding of the needs of users, and can provide users with more intelligent, accurate, humane results.