【摘要】： 本體(Ontology)的概念起源于哲學(xué)領(lǐng)域，20世紀(jì)90年代以來(lái)，人們將本體的概念引入人工智能、知識(shí)工程和檢索領(lǐng)域，從而使本體概念的內(nèi)涵也隨之發(fā)生了變化，在這些領(lǐng)域中，本體研究是關(guān)于知識(shí)概念表示和知識(shí)組織體系方面的研究。本體通常是指一套有關(guān)某一學(xué)科或某一領(lǐng)域的術(shù)語(yǔ)詞表，以及術(shù)語(yǔ)之間關(guān)系的規(guī)范和說(shuō)明。近年來(lái)，關(guān)于本體的研究與應(yīng)用呈現(xiàn)出加速度發(fā)展的趨勢(shì)，基礎(chǔ)理論更為完善，應(yīng)用領(lǐng)域更為廣泛，技術(shù)產(chǎn)品更為成熟．與國(guó)外涌現(xiàn)的研究項(xiàng)目和研究成果相比，國(guó)內(nèi)關(guān)于本體的研究尚存在較大差距，國(guó)內(nèi)關(guān)于本體應(yīng)用的研究尚處于起步階段，尚未見(jiàn)到有關(guān)國(guó)內(nèi)構(gòu)建具有推理功能的學(xué)科或領(lǐng)域本體系統(tǒng)的報(bào)導(dǎo)。 本課題以公安執(zhí)法領(lǐng)域中描述案情的中文文本為研究對(duì)象，探討采用自然語(yǔ)言理解中的文本分類(lèi)方法，實(shí)現(xiàn)根據(jù)文本的內(nèi)容來(lái)判別文本類(lèi)別，，給出相應(yīng)的法律建議。首先在對(duì)語(yǔ)義檢索部分，我們提出了最短匹配算法，針對(duì)檢索的結(jié)果采用了一種基于本體的語(yǔ)義信息檢索模型，避免了傳統(tǒng)向量空問(wèn)模型在語(yǔ)義處理方面的缺陷。根據(jù)不同關(guān)鍵字之間的組合利用本體的推理機(jī)制，采用Jena開(kāi)發(fā)包作為起推理工具。利用不同關(guān)鍵字之間的語(yǔ)義關(guān)聯(lián)等方面來(lái)充分考慮，得到與每個(gè)涉案人員相關(guān)一些的相關(guān)法律條款。并初步構(gòu)建了一個(gè)公安法律文本自動(dòng)分類(lèi)系統(tǒng)。
[Abstract]:The concept of ontology (Ontology) originated in the field of philosophy. Since 1990s, people have introduced the concept of ontology into the fields of artificial intelligence, knowledge engineering and retrieval. Ontology research is about knowledge concept representation and knowledge organization system. Noumenon usually refers to a set of glossary of terms relating to a discipline or field, as well as the specification and explanation of the relationship between terms. In recent years, the research and application of ontology show the trend of acceleration development, the basic theory is more perfect, the application field is wider, the technology product is more mature. There is still a big gap in the research of ontology in China, and the research on ontology application is still in its infancy. There is no report on the construction of ontology system with reasoning function in China. Taking the Chinese text describing the case in the field of public security law enforcement as the research object, this paper discusses the method of text classification in natural language understanding, realizes the classification of the text according to the content of the text, and gives the corresponding legal advice. In the part of semantic retrieval, we propose the shortest matching algorithm, and adopt an ontology based semantic information retrieval model to avoid the defects of the traditional vector space query model in semantic processing. According to the combination of different keywords, using the reasoning mechanism of ontology, the Jena development kit is used as the reasoning tool. By taking full account of the semantic correlation between different keywords, we can get some relevant legal provisions related to each person involved in the case. And has constructed a public security law text automatic classification system preliminarily.
【學(xué)位授予單位】：南昌大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2007
【分類(lèi)號(hào)】：D631.1

【引證文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前3條

1 謝彩云;基于法律領(lǐng)域的本體學(xué)習(xí)方法的研究與應(yīng)用[D];南昌大學(xué);2012年

2 張傳輝;公安知識(shí)管理系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];電子科技大學(xué);2012年

3 唐果;基于語(yǔ)義領(lǐng)域向量空間模型的文本相似度計(jì)算[D];云南大學(xué);2013年

本文編號(hào)：2361083