本文選題：信息檢索　切入點(diǎn)：查詢理解　出處：《大連理工大學(xué)》2016年博士論文　論文類型：學(xué)位論文

【摘要】：廣泛迅捷的分享和交換信息是互聯(lián)網(wǎng)最重要的優(yōu)點(diǎn)之一,然而隨著互聯(lián)網(wǎng)中承載的數(shù)據(jù)量和信息量呈指數(shù)級(jí)爆炸式增長,導(dǎo)致人們必須面對(duì)日益嚴(yán)重的信息過載問題。在該背景下信息檢索技術(shù)應(yīng)運(yùn)而生并隨著互聯(lián)網(wǎng)一起迅猛的發(fā)展,成為當(dāng)前解決信息過載問題最直接和有效的手段。同時(shí)查詢理解和相關(guān)反饋技術(shù)是近幾年被廣泛驗(yàn)證的用于改善信息檢索性能的有效途徑,盡管現(xiàn)有研究已經(jīng)取得了一些重要進(jìn)展,但是仍然有很多關(guān)鍵的問題沒有得到很好的解決。針對(duì)現(xiàn)有研究中存在的種種不足,本文在分析當(dāng)前各查詢理解和相關(guān)反饋算法基礎(chǔ)上,重點(diǎn)對(duì)查詢理解技術(shù)以及以其為基礎(chǔ)的相關(guān)反饋技術(shù)進(jìn)行了深入的研究。本文的主要研究工作和貢獻(xiàn)如下：1.針對(duì)查詢理解技術(shù)中的詞項(xiàng)權(quán)重預(yù)測問題,本文將其轉(zhuǎn)換為序列標(biāo)注問題,提出了一種新的基于循環(huán)神經(jīng)網(wǎng)絡(luò)的查詢詞項(xiàng)權(quán)重學(xué)習(xí)模型。該模型通過綜合考慮查詢中各詞項(xiàng)的統(tǒng)計(jì)、語法、語義以及詞項(xiàng)間關(guān)系信息構(gòu)造查詢詞項(xiàng)特征向量,同時(shí)利用遺傳算法結(jié)合真實(shí)文檔相關(guān)性標(biāo)注得到最優(yōu)詞項(xiàng)權(quán)重值,最后利用雙向循環(huán)神經(jīng)網(wǎng)絡(luò)對(duì)查詢詞項(xiàng)序列與相應(yīng)最優(yōu)權(quán)重序列之間的關(guān)系進(jìn)行有監(jiān)督學(xué)習(xí)建模,實(shí)現(xiàn)了對(duì)查詢詞項(xiàng)權(quán)重自動(dòng)、合理和有效地預(yù)測。實(shí)驗(yàn)結(jié)果表明,通過該方法得到的查詢詞項(xiàng)權(quán)重能夠明顯地提升檢索效果,并且在多個(gè)數(shù)據(jù)集和檢索結(jié)果準(zhǔn)確率評(píng)價(jià)指標(biāo)上均顯著地優(yōu)于各對(duì)比方法。2.針對(duì)查詢理解技術(shù)中現(xiàn)有查詢意圖分類方法普遍存在的嚴(yán)重依賴人工標(biāo)注數(shù)據(jù)和面對(duì)類別體系變化不靈活的問題,本文將其轉(zhuǎn)換為由一個(gè)序列分類問題和一個(gè)經(jīng)典分類問題組成的兩階段分類問題,并根據(jù)該分類問題的特點(diǎn),提出了一種新的基于級(jí)聯(lián)深度學(xué)習(xí)的查詢意圖分類方法。該方法首先從提高分類靈活度和效率的角度出發(fā),提出了一種級(jí)聯(lián)的深度神經(jīng)網(wǎng)絡(luò),構(gòu)造了一個(gè)兩階段查詢意圖分類器；然后從降低對(duì)人工標(biāo)注依賴的角度出發(fā),通過隱式相關(guān)反饋技術(shù)挖掘源于真實(shí)用戶的標(biāo)注行為,實(shí)現(xiàn)了查詢分類訓(xùn)練數(shù)據(jù)的自動(dòng)構(gòu)造。實(shí)驗(yàn)結(jié)果表明,該方法能夠有效的對(duì)查詢按主題意圖進(jìn)行分類,且分類效果顯著的優(yōu)于各對(duì)比方法。3.針對(duì)現(xiàn)有基于查詢擴(kuò)展技術(shù)的相關(guān)反饋方法對(duì)檢索系統(tǒng)查詢?nèi)罩炯捌渲胁樵冊~項(xiàng)間關(guān)系挖掘不足的問題,本文從充分利用檢索系統(tǒng)查詢?nèi)罩具M(jìn)行查詢擴(kuò)展的角度出發(fā),提出了一種新的基于兩階段SimRank算法和查詢擴(kuò)展技術(shù)的相關(guān)反饋方法。該方法通過引入權(quán)重關(guān)系改進(jìn)了基于圖結(jié)構(gòu)的相似度算法SimRank,并使用改進(jìn)算法在由查詢點(diǎn)擊圖經(jīng)多次轉(zhuǎn)換得到的詞項(xiàng)關(guān)系圖上全面深入地挖掘詞項(xiàng)間相似度及語義關(guān)聯(lián),從而篩選出高質(zhì)量的查詢擴(kuò)展詞項(xiàng)。通過在公開標(biāo)準(zhǔn)數(shù)據(jù)集上的實(shí)驗(yàn)表明該方法可以有效地選擇高質(zhì)量相關(guān)擴(kuò)展詞項(xiàng),使得檢索效果有顯著的提升。4.針對(duì)現(xiàn)有基于語言模型的相關(guān)反饋方法未能同時(shí)充分利用正向和負(fù)向相關(guān)信息的問題,本文從充分利用隱式反饋和同時(shí)挖掘正負(fù)兩種相關(guān)信息的角度出發(fā),提出了一種新的基于語言模型的正負(fù)雙向相關(guān)反饋方法。該方法通過分析疑難查詢場景下隱式反饋的正負(fù)雙向相關(guān)文檔,基于語言模型檢索框架,同時(shí)構(gòu)造正向和負(fù)向相關(guān)語言模型,并利用正向模型進(jìn)一步優(yōu)化負(fù)向模型,最大化地提高相關(guān)文檔排名并盡量過濾非相關(guān)文檔,從而改善反饋檢索的效果。通過基于多個(gè)TREC標(biāo)準(zhǔn)數(shù)據(jù)集的實(shí)驗(yàn)驗(yàn)證了該相關(guān)反饋方法的有效性,且效果顯著優(yōu)于僅使用正向或負(fù)向相關(guān)信息的相關(guān)反饋方法,使得反饋檢索效果有顯著的提升。通過以上四個(gè)方面的研究,能夠得到一個(gè)利用查詢理解和相關(guān)反饋技術(shù)改善信息檢索整體過程的解決方案,幫助信息檢索系統(tǒng)提升檢索效果并改善用戶體驗(yàn)。
[Abstract]:Wide quick exchanging and sharing information is one of the most important advantages of the Internet, but with the Internet in the carrying amount of data and information of the exponentially explosive growth, leading people must face the increasingly serious problem of information overload. And with the development of the Internet with rapid and in the context of information retrieval technology should be. To solve the problem of information overload is the most direct and effective means. At the same time query understanding and relevance feedback technology in recent years has been widely used to validate the effective way to improve the performance of information retrieval, although the existing research has made some important progress, but there are still many key problems have been solved very well. Aiming at the shortcomings of in the existing studies, based on the analysis of the current understanding of the query and relevance feedback algorithm based on query technology and focus on the understanding of its base Relevant feedback technology foundation are studied. The main research work and contributions are as follows: 1. for the query understanding in lexical entry weight prediction problem, this paper converts it into sequence labeling problems, put forward a new query lexical entry weight learning model based on the model of recurrent neural network. According to the statistics, the considering the lexical entry in the query syntax, semantic and lexical entry information to construct the relationship between lexical entry query feature vector, using genetic algorithm to get the optimal combination of real document relevance marking lexical entry weight value, finally using double to recurrent neural network supervised learning modeling on the relationship between the query sequence and the corresponding lexical entry sequence to achieve the optimal weights. The lexical entry query weights automatically, reasonable and effective prediction. Experimental results show that the query obtained by this method can obviously raise the weight of lexical entry L search results, and in multiple data sets and retrieval accuracy evaluation indicators were significantly better than the contrast method for.2. query understanding relies heavily on common classification methods in existing query intention of labeled data and the flexibility of system changes the face of the category, the two stage is converted to a sequence the classification problem and a classical classification problem consisting of classification problems, and according to the characteristics of the classification problem, proposed a new cascade deep learning based query intent classification method. This method firstly improve the flexibility and efficiency of classification point of view, put forward the depth of the neural network in a cascaded structure, a a two stage query intent classifier; then from the perspective of reducing manual annotation angle of annotation by implicit feedback from real users of mining technology For the realization of the automatic classification, structure of the training data. The experimental results show that this method can classify the query according to the theme of intention, and the classification results were significantly better than the existing.3. comparison method based on query expansion technology problems of relevance feedback retrieval system and the query log query between lexical entry in this paper, mining, query expansion from the perspective of making full use of the retrieval system query log, proposes a relevance feedback method extended two phase SimRank algorithm and query based on new technology. The method by introducing weights of SimRank similarity algorithm based on the graph structure is improved, and the improved algorithm in the lexical entry diagram by a query click through many transformations on the comprehensive and in-depth mining word similarity and semantic association between, in order to find out the query expansion of high quality Lexical entry through open standard data set on the experiment shows that this method can effectively select high quality related extended lexical entry, the retrieval results have significantly improved the existing.4. language model based on relevance feedback method also failed to make full use of positive and negative related information, the paper will make full use of implicit feedback and at the same time, two kinds of mining positive and negative information point of view, put forward a new language model and bidirectional feedback method based on this method. Through the analysis of positive and negative related documents difficult query scenarios implicit feedback retrieval model based on language, framework, and construct the positive and negative correlation model to the language, and the use of positive the model to further optimize the negative model, to maximize the relevant document ranking and try to filter the non relevant documents, so as to improve the effect of feedback retrieval based on multi pass. TREC standard data sets the experimental results verify the effectiveness of the relevant feedback method, and the effect was better than using only the feedback method of positive or negative information, the feedback effect is improved. Through the study of the above four aspects, can get a query using relevance feedback technology to improve understanding and solving for the whole process of information retrieval and information retrieval system to help improve the retrieval effectiveness and improve the user experience.

【學(xué)位授予單位】：大連理工大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP183

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 殷亞玲;張蕾;李海軍;;基于概念圖的相關(guān)反饋技術(shù)研究[J];計(jì)算機(jī)工程與應(yīng)用;2006年03期

2 黎琳;趙英;;基于內(nèi)容的圖像檢索反饋技術(shù)概述[J];圖書情報(bào)工作;2006年11期

3 徐建軍;吳玲達(dá);;基于內(nèi)容的圖像檢索中的相關(guān)反饋技術(shù)發(fā)展[J];計(jì)算機(jī)科學(xué);2004年07期

4 李蕾;;應(yīng)用互動(dòng)反饋技術(shù)開展有效教學(xué)的探索[J];中國電化教育;2008年05期

5 李云,劉嘉敏,強(qiáng)保華,葉春曉,吳中福;圖像檢索中相關(guān)反饋技術(shù)的特性研究[J];計(jì)算機(jī)工程;2004年07期

6 郭江濤;梁新濤;裴志強(qiáng);;論泄露反饋技術(shù)在煤礦井下管理中的應(yīng)用[J];現(xiàn)代商貿(mào)工業(yè);2011年17期

7 汪紅秋;反饋在修改文獻(xiàn)表達(dá)中的應(yīng)用[J];情報(bào)學(xué)刊;1989年03期

8 梁波;楊樾;;互動(dòng)反饋技術(shù)究竟給我們帶來了什么[J];中小學(xué)信息技術(shù)教育;2010年01期

9 王穎;高新波;;基于支持向量機(jī)和相關(guān)反饋技術(shù)的腫塊檢測算法[J];西安電子科技大學(xué)學(xué)報(bào);2007年02期

10 ;觸覺反饋技術(shù):逼真震動(dòng)體驗(yàn)的改革者[J];金卡工程;2012年04期

相關(guān)會(huì)議論文 前3條

1 林家洋;陶紅;黃建熒;陳學(xué)仁;;新型槳葉反饋技術(shù)在軸流轉(zhuǎn)槳式水輪機(jī)上的應(yīng)用[A];中國水力發(fā)電工程學(xué)會(huì)信息化專委會(huì)、水電控制設(shè)備專委會(huì)2013年學(xué)術(shù)交流會(huì)論文集[C];2013年

2 董們;唐萬斌;;基于Grassmann流型理論的有限反饋技術(shù)在MIMO系統(tǒng)中的研究[A];2008年中國西部青年通信學(xué)術(shù)會(huì)議論文集[C];2008年

3 溫馨;王鵬;李少亮;郭文琪;陳群;婁穎;;基于關(guān)鍵字的XML信息檢索反饋技術(shù)的研究[A];NDBC2010第27屆中國數(shù)據(jù)庫學(xué)術(shù)會(huì)議論文集(B輯)[C];2010年

相關(guān)重要報(bào)紙文章 前2條

1 夏慶 代果;應(yīng)用互動(dòng)反饋技術(shù) 提高課堂教學(xué)有效性[N];中國教育報(bào);2014年

2 本報(bào)記者 李旭陽;讓虛擬應(yīng)用“觸手可及”[N];計(jì)算機(jī)世界;2012年

相關(guān)博士學(xué)位論文 前2條

1 高歡芹;無線信道有限反饋技術(shù)研究[D];南京郵電大學(xué);2015年

2 馬云龍;查詢理解與正負(fù)雙向相關(guān)反饋技術(shù)研究[D];大連理工大學(xué);2016年

相關(guān)碩士學(xué)位論文 前10條

1 王士欣;面向5G的大規(guī)模MIMO言道反饋技術(shù)研究與設(shè)計(jì)[D];南京郵電大學(xué);2016年

2 程先捷;基于相關(guān)反饋技術(shù)的檢索結(jié)果排序模型研究[D];哈爾濱工程大學(xué);2012年

3 孫華彬;基于內(nèi)容的圖像檢索中相關(guān)反饋技術(shù)的研究及系統(tǒng)實(shí)現(xiàn)[D];西北大學(xué);2009年

4 徐欽;基于內(nèi)容的圖像檢索及相關(guān)反饋技術(shù)研究[D];南京信息工程大學(xué);2012年

5 唐海龍;協(xié)作多點(diǎn)傳輸與有限反饋技術(shù)研究[D];上海師范大學(xué);2014年

6 鄧劍勛;信息檢索中的相關(guān)反饋技術(shù)研究[D];重慶大學(xué);2007年

7 郭云;MIMO無線通信系統(tǒng)中有限反饋技術(shù)的研究[D];鄭州大學(xué);2014年

8 石磊;協(xié)作多點(diǎn)傳輸系統(tǒng)中的有限反饋技術(shù)研究[D];北京郵電大學(xué);2015年

9 廖國銳;MIMO系統(tǒng)的有限反饋技術(shù)研究[D];華南理工大學(xué);2010年

10 蔣曉紅;基于查詢向量轉(zhuǎn)移和聚類相結(jié)合的相關(guān)反饋技術(shù)研究[D];中國石油大學(xué);2008年

，

本文編號(hào)：1584261