The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-2/W7
https://doi.org/10.5194/isprs-archives-XLII-2-W7-137-2017
https://doi.org/10.5194/isprs-archives-XLII-2-W7-137-2017
12 Sep 2017
 | 12 Sep 2017

AMATCHMETHOD BASED ON LATENT SEMANTIC ANALYSIS FOR EARTHQUAKEHAZARD EMERGENCY PLAN

D. Sun, S. Zhao, Z. Zhang, and X. Shi

Keywords: Earthquake, Plan, Latent Semantic Analysis (LSA), Keywords Extraction, Match, Vector Space

Abstract. The structure of the emergency plan on earthquake is complex, and it’s difficult for decision maker to make a decision in a short time. To solve the problem, this paper presents a match method based on Latent Semantic Analysis (LSA). After the word segmentation preprocessing of emergency plan, we carry out keywords extraction according to the part-of-speech and the frequency of words. Then through LSA, we map the documents and query information to the semantic space, and calculate the correlation of documents and queries by the relation between vectors. The experiments results indicate that the LSA can improve the accuracy of emergency plan retrieval efficiently.