{"title":"基于内容相似度的多文档摘要自动评价分析","authors":"Li-qing Qiu, Bin Pang","doi":"10.1109/ICDS.2008.9","DOIUrl":null,"url":null,"abstract":"We introduce an automated evaluation method based on content similarity, and construct a vector space of words, on which we compute cosine similarity of automated summaries and human summaries. The method is tested on DUC 2005 data, and produces acceptable results, which may avoid some shortcomings of n-gram. We also test the effects of stopwords and stemming.","PeriodicalId":422080,"journal":{"name":"Second International Conference on the Digital Society","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Analysis of Automated Evaluation for Multi-document Summarization Using Content-Based Similarity\",\"authors\":\"Li-qing Qiu, Bin Pang\",\"doi\":\"10.1109/ICDS.2008.9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We introduce an automated evaluation method based on content similarity, and construct a vector space of words, on which we compute cosine similarity of automated summaries and human summaries. The method is tested on DUC 2005 data, and produces acceptable results, which may avoid some shortcomings of n-gram. We also test the effects of stopwords and stemming.\",\"PeriodicalId\":422080,\"journal\":{\"name\":\"Second International Conference on the Digital Society\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-02-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Second International Conference on the Digital Society\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICDS.2008.9\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Second International Conference on the Digital Society","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDS.2008.9","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Analysis of Automated Evaluation for Multi-document Summarization Using Content-Based Similarity
We introduce an automated evaluation method based on content similarity, and construct a vector space of words, on which we compute cosine similarity of automated summaries and human summaries. The method is tested on DUC 2005 data, and produces acceptable results, which may avoid some shortcomings of n-gram. We also test the effects of stopwords and stemming.