A Semantic Extension Method for Retrieval Restricted Items of Remote Sensing Products Based on Web Mining

Abstract

The invention discloses a web mining-based remote sensing product search limited item semantic extension method. The method mainly comprises the following steps: extracting brief introduction content and time information on the basis of a network search result so as to form an extract chapter; extracting time words and space words in the chapter, and labeling basic time expressions corresponding to the time words and the space words; normalizing the labelled basic time expressions; counting the normalized basic time expressions and the space words, ad taking the time and space words which appear at high frequency as extension results. In allusion to the problem that the time and space information input by users has fuzziness, reference performance and dynamics in remote sensing product search, the web mining-based remote sensing product search limited item semantic extension method is disclosed, so that integrated time and space information is positioned, the semantic extension carried out on user input information is realized, the user demands can be correctly understood, and the correctness and timeliness of the search are improved.

Description

technical field [0001] The invention belongs to the field of remote sensing data processing and information extraction, and relates to a semantic extension technology for remote sensing product retrieval limited items based on network mining technology. Background technique [0002] Semantic expansion of restricted items in remote sensing products refers to the semantic expansion of the time and place information contained in the restricted items in the retrieved content when retrieving remote sensing products. Among them, the limited item refers to the time and space limited information contained in the search content of the remote sensing product that the user queries, and is divided into explicit and implicit space-time information. The explicit space-time information refers to the time and space directly displayed in the search content. Spatial information, while implicit spatio-temporal information means that the search content does not directly display the time and spa...

AI Technical Summary

Problems solved by technology

However, limited by the principle of ontology construction, the query expansion through ontology is extremely professional, and it is not convenient for ordinary users to search. However, with the publicization of remote sensing products and services, the domain characteristics of remote sensing products are gradually weakening, and the heterogeneity of remote sensing products and services The nature and dynamic characteristics are becoming more and more obvious, and user semantic extension solely relying on the ontology idea cannot meet the requirements of retrieval precision and recall

Images