论文部分内容阅读
一、词汇的规范化处理对检索效率的影响情报检索语言的规范化处理的目的是使情报检索具有较高的检索效率。衡量情报检索效率的标准一般来说,主要有以下三个方面.1.检全率:检出的、与提问相关的文献在整个被检索系统的相关文献中所占的比例。2.检准率:检出的、与提问相关的文献在全部检出文献中所占的比例.3.成本效益问题:语词的规范化处理费用(包括所花费的人力、物力、财力及所占的空间和时间等)与检全率,检准率提高之间的比例。现在似乎有这样一种看法,即语词的规范化处理越严格越好.作者认为并不尽然。
First, the normalization of the word processing efficiency of the retrieval efficiency Information retrieval language standardization processing is to make the information retrieval has a high search efficiency. Generally speaking, there are three aspects to measure the efficiency of information retrieval: 1. Check-up rate: the proportion of the documents related to the questions detected and relevant to the retrieved system. 2.Precision rate: the proportion of the detected and question-related documents in all the detected documents.3.Cost-benefit problem: The normalized processing cost of the words (including the manpower, material and financial resources and their share The space and time, etc.) and the detection rate, the rate of increase the accuracy of the ratio. Now there seems to be a view that the more standardized norms of words processing the better, the author does not think it is.