模具经验性知识的搜索条件预处理方法

发布时间：2018-11-06 17:36

【摘要】：随着科学技术的发展,企业对知识越来越重视,而获取知识时搜索引擎无疑是较好的选择；而对于企业所需的专业知识,垂直搜索引擎无疑是较好的选择,但垂直搜索引擎也具有其弊端,例如垂直搜索引擎的起步较晚,使得绝大多数领域并没有垂直搜索引擎,无法满足大多数领域用户的需求同时垂直搜索引擎的开发成本也比较高,所以如何在通用搜索引擎上较为准确地搜索到所需要的专业知识就显得尤为重要。在互联网上的知识大多是以非结构化的形式存在的,而通用搜索引擎对非结构化信息的处理能力较低,同时对专业知识的挖掘深度不够,难以准确的搜索到所需要的专业知识。模具经验性知识对企业的发展至关重要,能够帮助企业提高快速解决问题的能力,管理水平,创新能力等；所以本文以模具经验性知识为研究对象,进行了相关研究,以期能够对模具企业获取这类知识起到一点帮助。针对以上问题,本文提出了一套对搜索条件的预处理方案,使得用户的搜索条件经过预处理后,转换成新的规范的搜索条件,在通用搜索引擎上能够更快更准确的搜索到所需要的专业知识。本文的主要工作内容有： 1、对通用搜索引擎的工作原理进行了研究,掌握其搜索的一般规律,使得后期建立的语义框架具备理论基础,能够更好更准确的为用户服务；对通用搜索引擎所存在的问题进行了学习研究。 2、对模具经验性知识的特征进行了深入调研,分析了模具经验性知识所具有的基本特征,并结合实验室师兄师姐以前的研究成果,确定了模具经验性知识的基本特征。 3、根据通用搜索引擎的工作原理及模具经验性知识的基本特征,提出了合适的模具经验性知识搜索条件语义框架,并通过实验验证了用户的搜索条件经过语义框架转换后得出的新的搜索条件,能够更准确的搜索到用户所需要的知识。 4、学习分析了前人所研究的分词算法,结合模具经验性知识的基本特征和所建立的语义框架,提出了一种改进型的分词算法,实现了对搜索条件的分词提取 5、针对前文所研究的内容,设计开发了一个搜索条件的预处理系统,实现了对搜索条件的语义框架匹配,中文自动分词及词典的智能学习等功能。
[Abstract]:With the development of science and technology, enterprises pay more and more attention to knowledge, and search engine is undoubtedly a better choice when acquiring knowledge. But the vertical search engine has its disadvantages, such as the late start of the vertical search engine, which makes most fields do not have a vertical search engine. At the same time, the development cost of vertical search engine is also high, so it is very important to search the required professional knowledge accurately in general search engine. Most of the knowledge on the Internet exists in the form of unstructured, but the ability of general search engine to deal with unstructured information is low, at the same time, the depth of professional knowledge mining is not enough, it is difficult to accurately search the required professional knowledge. The experiential knowledge of mould is very important to the development of the enterprise, which can help the enterprise to improve the ability to solve the problem quickly, the management level, the innovation ability and so on. Therefore, this paper takes the empirical knowledge of mould as the research object, carries on the related research, in order to be able to obtain this kind of knowledge to the mold enterprise to play a little help. In view of the above problems, this paper puts forward a set of preprocessing scheme of search condition, which makes the user's search condition be converted into a new standard search condition after preprocessing. Search for the required expertise faster and more accurately on a generic search engine. The main contents of this paper are as follows: 1. The working principle of general search engine is studied and the general rules of search are grasped so that the semantic framework established in the later period has theoretical foundation and can serve users better and more accurately; The problems of general search engine are studied. 2. The characteristics of mould experience knowledge are investigated deeply, the basic characteristics of mould experience knowledge are analyzed, and the basic characteristics of mould experience knowledge are determined in combination with the previous research results of lab brother and elder sister. 3. According to the working principle of the general search engine and the basic characteristics of the experiential knowledge of the mould, a suitable semantic framework for the search condition of the experiential knowledge of the mould is put forward. The new search conditions are verified by experiments, which can more accurately search the knowledge that users need. 4. Learning and analyzing the word segmentation algorithms studied by predecessors, combining with the basic characteristics of the knowledge of mould experience and the semantic framework established, an improved word segmentation algorithm is proposed, which realizes the segmentation of search conditions. A preprocessing system of search conditions is designed and developed to match the semantic framework of search conditions, Chinese automatic word segmentation and the intelligent learning of dictionaries.
【学位授予单位】：广东工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.3

【相似文献】