發(fā)布時(shí)間：2019-01-01 12:23

【摘要】：本文主要研究利用人工免疫系統(tǒng)進(jìn)行Web文本挖掘的方法。 第1章首先對(duì)選題背景進(jìn)行了介紹,然后介紹了免疫學(xué)的發(fā)展歷程和主要研究?jī)?nèi)容。因?yàn)槊庖邔W(xué)與人工免疫系統(tǒng)是密切相關(guān)的,對(duì)二者之間的內(nèi)在聯(lián)系進(jìn)行了歸納。本文的研究?jī)?nèi)容就是建立在免疫學(xué)與人工免疫系統(tǒng)的關(guān)系基礎(chǔ)上。 第2章對(duì)自然免疫系統(tǒng)進(jìn)行了較詳細(xì)的介紹,主要包括免疫系統(tǒng)的基本組成、機(jī)制和原理。人工免疫系統(tǒng)所依據(jù)的免疫學(xué)原理主要包括免疫網(wǎng)絡(luò)理論,克隆選擇和陰性選擇原理。人工免疫系統(tǒng)正是建立在免疫學(xué)理論和免疫系統(tǒng)機(jī)制基礎(chǔ)上。 第3章就集中介紹了人工免疫系統(tǒng)的二進(jìn)制模型。許多是為了研究免疫系統(tǒng)機(jī)制而開發(fā)的。后來出現(xiàn)的模型逐漸轉(zhuǎn)到工程領(lǐng)域。本章重點(diǎn)介紹了最早由Farmer提出的微分方程,基于基因庫的模型和協(xié)同進(jìn)化算法模型。協(xié)同進(jìn)化算法模型在本文進(jìn)行擴(kuò)展,應(yīng)用到Web文本挖掘。 第4章主要對(duì)Web文本挖掘技術(shù)進(jìn)行了詳細(xì)討論,Web文本挖掘技術(shù)是涉及多個(gè)技術(shù)領(lǐng)域的交叉領(lǐng)域。包括許多較為復(fù)雜的技術(shù)方法,從特征抽取到模型建立,以及模型評(píng)價(jià)方法等等。這一章與前三章尤其是第2和第3章結(jié)合起來,形成本文第5章所給出的Web文本分類模型理論與技術(shù)基礎(chǔ)。 第5章給出了基于協(xié)同進(jìn)化算法的免疫陰性選擇模型,并與傳統(tǒng)方法進(jìn)行了比較,給出了比較結(jié)果。表明人工免疫系統(tǒng)做為Web文本分類方法是可行的。雖然還有許多不足之處。
[Abstract]:This paper mainly studies the method of Web text mining using artificial immune system. The first chapter introduces the background of the topic, and then introduces the development of immunology and the main research contents. Because immunology and artificial immune system are closely related, the relationship between them is summarized. The research of this paper is based on the relationship between immunology and artificial immune system. Chapter 2 introduces the natural immune system in detail, including the basic composition, mechanism and principle of immune system. The immune principles of artificial immune system mainly include immune network theory, clone selection and negative selection. Artificial immune system is based on immunological theory and immune system mechanism. Chapter 3 focuses on the binary model of the artificial immune system. Many have been developed to study the mechanisms of the immune system. The models that emerged later gradually moved to the field of engineering. This chapter focuses on the differential equations proposed by Farmer, gene pool based model and coevolutionary algorithm model. The co-evolutionary algorithm model is extended in this paper and applied to Web text mining. In chapter 4, the Web text mining technology is discussed in detail. Web text mining technology is a cross domain involving many technical fields. It includes many complicated technical methods, from feature extraction to model building, and model evaluation methods and so on. This chapter is combined with the first three chapters, especially the second and third chapters, to form the theoretical and technical foundation of Web text classification model given in Chapter 5. In chapter 5, the immune negative selection model based on coevolutionary algorithm is presented, and the results are compared with the traditional method. The results show that the artificial immune system is feasible for Web text classification. Although there are many shortcomings.
【學(xué)位授予單位】：哈爾濱工程大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2005
【分類號(hào)】：TP18;R392

【引證文獻(xiàn)】

相關(guān)博士學(xué)位論文 前1條

1 宋吉廣;基于升力反饋的全航速減搖鰭研究[D];哈爾濱工程大學(xué);2012年

相關(guān)碩士學(xué)位論文 前3條

1 包暉;基于免疫算法的木馬檢測(cè)技術(shù)研究[D];河南工業(yè)大學(xué);2010年

2 周利霞;鐵譜圖像識(shí)別的理論與方法研究[D];浙江大學(xué);2006年

3 趙麗;木馬檢測(cè)方法的研究與實(shí)現(xiàn)[D];蘭州理工大學(xué);2008年

，

本文編號(hào)：2397559