【摘要】：隨著科技的發(fā)展，個(gè)性化信息檢索(Personalize Information Retrieval)作為人工智能的分支得到了快速的發(fā)展。伴隨著Internet技術(shù)的發(fā)展和普及，越來越多的用戶開始利用搜索引擎來檢索自己所需要的信息。盡管搜索引擎技術(shù)的發(fā)展已經(jīng)比較成熟，但用戶要想準(zhǔn)確、快速的檢索出自己感興趣的信息卻變得不是那么的容易。因此，個(gè)性化搜索引擎的出現(xiàn)，為人們提供了檢索服務(wù)。 首先，本文系統(tǒng)的研究了基于網(wǎng)絡(luò)代理的個(gè)性化搜索引擎中所涉及的關(guān)鍵技術(shù)，提出了一種基于客戶端及模糊網(wǎng)絡(luò)代理的信息檢索系統(tǒng)模型。該系統(tǒng)由用戶代理服務(wù)器、搜索引擎代理服務(wù)器、排序代理服務(wù)器三個(gè)模塊構(gòu)成。 其次，，介紹的該搜索系統(tǒng)的工作過程：用戶通過下載客戶端，在個(gè)人的PC機(jī)上進(jìn)行安裝。每當(dāng)用戶需要檢索信息時(shí)，通過PC客戶端進(jìn)行輸入查詢。并將查詢的結(jié)果通過客戶端提供給用戶。該客戶端主要與模糊網(wǎng)絡(luò)代理服務(wù)器進(jìn)行交互，信息交換，信息反饋，信息過濾等操作。利用模糊網(wǎng)絡(luò)構(gòu)建搜索引擎代理，充分的利用了模糊網(wǎng)絡(luò)的安全性，智能性，可適應(yīng)性等特點(diǎn)。模糊網(wǎng)絡(luò)代理中的排序代理模塊根據(jù)客戶端存儲(chǔ)的用戶興趣配置文件來進(jìn)行檢索結(jié)果的信息過濾，并將此次用戶興趣及時(shí)更新到客戶端的配置文件中，最終將檢索結(jié)果通過客戶端呈獻(xiàn)給用戶。由于每個(gè)代理都是基于模糊概念網(wǎng)絡(luò)的，所以在檢索速度上有很大的提升。通過客戶端提供的用戶配置文件去過濾檢索結(jié)果，這很大程度上提高了檢索結(jié)果的準(zhǔn)確性。 最后，我們?cè)谂判虼碇袘?yīng)用改進(jìn)的排序算法去進(jìn)行結(jié)果排序，這提高的搜索引擎的效率。三個(gè)模塊相互配合，從而實(shí)現(xiàn)了個(gè)性化、高效率、智能化的搜索引擎。
[Abstract]:With the development of science and technology, personalized information retrieval (Personalize Information Retrieval), as a branch of artificial intelligence, has been developed rapidly. With the development and popularization of Internet technology, more and more users begin to use search engines to retrieve the information they need. Although the development of search engine technology has been more mature, it is not easy for users to retrieve the information they are interested in accurately and quickly. Therefore, the emergence of personalized search engines, for people to provide retrieval services. Firstly, this paper systematically studies the key technologies involved in the personalized search engine based on network agent, and proposes a model of information retrieval system based on client and fuzzy network agent. The system consists of three modules: user proxy server, search engine proxy server and sort proxy server. Secondly, the working process of the search system is introduced: the user downloads the client and installs it on the personal PC machine. Each time the user needs to retrieve information, the input query is made through the PC client. The result of the query is provided to the user through the client. The client mainly interacts with fuzzy network proxy server, information exchange, information feedback, information filtering and so on. Using fuzzy network to construct search engine agent makes full use of the security, intelligence and adaptability of fuzzy network. The sorting agent module in the fuzzy network agent filters the information of the retrieval result according to the user interest profile stored by the client, and updates the user interest to the configuration file of the client in time. Finally, the retrieval results are presented to the user through the client. Because each agent is based on fuzzy concept network, the retrieval speed is greatly improved. The retrieval results are filtered through the user profile provided by the client, which greatly improves the accuracy of the retrieval results. Finally, we use the improved sorting algorithm to sort the results, which improves the efficiency of the search engine. The three modules cooperate with each other to achieve a personalized, efficient and intelligent search engine.
【學(xué)位授予單位】：哈爾濱理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王琦,唐世渭,楊冬青,王騰蛟;基于DOM的網(wǎng)頁主題信息自動(dòng)提取[J];計(jì)算機(jī)研究與發(fā)展;2004年10期

2 肖增良;樂曉波;周輝;;基于蟻群-遺傳算法的模糊Petri網(wǎng)參數(shù)優(yōu)化算法[J];計(jì)算機(jī)工程與應(yīng)用;2010年29期

3 何映思;鄧輝文;;一種帶權(quán)重的真值流推理算法[J];計(jì)算機(jī)科學(xué);2009年12期

4 伍方明;趙曉哲;郭銳;;模糊專家系統(tǒng)中量詞的推理方法[J];計(jì)算機(jī)工程;2009年19期

5 董茜;;模糊集上基于一般蘊(yùn)含算子的三I算法[J];計(jì)算機(jī)應(yīng)用與軟件;2010年09期

6 張曉華;王世倫;;網(wǎng)絡(luò)學(xué)習(xí)中認(rèn)知超載問題及其解決[J];中國(guó)現(xiàn)代教育裝備;2007年09期

7 孫堅(jiān);鄭恩輝;鄒超;劉長(zhǎng)東;;支持向量機(jī)和一類模糊推理系統(tǒng)的等效性及其應(yīng)用[J];控制與決策;2009年09期

8 章成志;王惠臨;;多語言文本聚類研究綜述[J];現(xiàn)代圖書情報(bào)技術(shù);2009年06期

9 黃德才;戚華春;錢能;;基于主題相似度模型的TS-PageRank算法[J];小型微型計(jì)算機(jī)系統(tǒng);2007年03期

10 萬樹平;;基于Vague集的多屬性群決策專家權(quán)重的確定[J];應(yīng)用數(shù)學(xué)與計(jì)算數(shù)學(xué)學(xué)報(bào);2010年01期

本文編號(hào)：2420738