發(fā)布時(shí)間：2018-03-03 00:36

  本文選題：社區(qū)問(wèn)答　切入點(diǎn)：問(wèn)題檢索　出處：《華東師范大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：在社區(qū)問(wèn)答(Community Question Answering,CQA)中,用戶(hù)提出查詢(xún)問(wèn)題,CQA系統(tǒng)在大量已有的問(wèn)題-答案對(duì)的知識(shí)庫(kù)中搜索相似的問(wèn)題,然后把該問(wèn)題的最佳答案當(dāng)作查詢(xún)問(wèn)題的答案返回給用戶(hù)。CQA系統(tǒng)包括兩個(gè)關(guān)鍵的子任務(wù):(1)問(wèn)題檢索(QuestionRetrieval),通過(guò)估計(jì)問(wèn)題對(duì)的語(yǔ)義相似性來(lái)找到和查詢(xún)問(wèn)題最相似的已有問(wèn)題;(2)答案排序(AnswerRanking),按照答案回答問(wèn)題的相關(guān)程度對(duì)多個(gè)答案進(jìn)行語(yǔ)義相關(guān)性排序,選出最佳的答案。構(gòu)建問(wèn)答知識(shí)庫(kù)是一項(xiàng)龐大而復(fù)雜的工程,一種可行的替代方案是利用互聯(lián)網(wǎng)的龐大資源檢索獲得問(wèn)題的答案。因此,本文的第一個(gè)工作是借助搜索引擎來(lái)構(gòu)建一個(gè)網(wǎng)絡(luò)資源輔助的社區(qū)問(wèn)答系統(tǒng),該系統(tǒng)在2015年TREC的實(shí)時(shí)問(wèn)答競(jìng)賽中獲得了第二名。以往關(guān)于CQA的研究多將CQA中的問(wèn)題檢索和答案排序兩個(gè)任務(wù)分開(kāi)獨(dú)立解決,沒(méi)有考慮它們之間的信息交互。本文的第二個(gè)工作考慮這兩個(gè)任務(wù)的相互促進(jìn),并設(shè)計(jì)新的有效特征來(lái)進(jìn)一步提高CQA的性能,相關(guān)工作發(fā)表在2016年IJCNN會(huì)議。傳統(tǒng)CQA系統(tǒng)采用專(zhuān)家精心設(shè)計(jì)的特征,泛化性差,而深度學(xué)習(xí)的優(yōu)勢(shì)是能夠自動(dòng)學(xué)習(xí)特征。因此,本文的第三個(gè)工作探索了深度學(xué)習(xí)模型在問(wèn)題檢索和答案排序任務(wù)上特征自動(dòng)學(xué)習(xí)的性能,相關(guān)工作發(fā)表在2016年的SemEval會(huì)議。在本文第二個(gè)和第三個(gè)工作的啟發(fā)下,本文的第四個(gè)工作深入研究了深度學(xué)習(xí)框架下的CQA系統(tǒng)。本文提出一個(gè)基于門(mén)機(jī)制的深度神經(jīng)網(wǎng)絡(luò)模型,該門(mén)機(jī)制能夠自動(dòng)學(xué)習(xí)問(wèn)題檢索和答案排序任務(wù)間的交互信息,從而幫助進(jìn)一步提高CQA性能。本文廣泛而深入地研究了采用傳統(tǒng)自然語(yǔ)言處理技術(shù)與深度學(xué)習(xí)方法的問(wèn)題檢索和答案排序相互促進(jìn)的CQA系統(tǒng),大量的實(shí)驗(yàn)結(jié)果表明,本文提出的兩個(gè)任務(wù)相互促進(jìn)的策略在傳統(tǒng)方法和深度學(xué)習(xí)方法中都能夠有效地提高CQA系統(tǒng)的性能。
[Abstract]:In Community Question answering and answering (CQA), users ask queries and CQA systems search for similar questions in a large number of existing questions-answer pairs of knowledge bases. Then the best answer to the question is returned to the user. CQA system including two key sub-tasks: 1) QuestionRetrieval is retrieved by estimating the semantic similarity of the question pairs to find the most similar to the query question. The answer is sorted by AnswerRanking.According to the degree of relevance of the answer to the question, the multiple answers are sorted in terms of semantic correlation. Choose the best answer. Building a question-and-answer knowledge base is a huge and complex project, and a viable alternative is to use the vast resources of the Internet to retrieve the answer to the question. The first work of this paper is to build a community Q & A system assisted by network resources with the help of search engine. In 2015, the system won the second place in the real-time quiz of TREC. In the past studies on CQA, the two tasks of question retrieval and answer sorting in CQA were solved separately and independently. The second work of this article considers the mutual promotion of the two tasks and designs new valid features to further improve the performance of CQA. The related work was published at the IJCNN Conference in 2016. The traditional CQA system adopts the characteristics carefully designed by experts and has poor generalization, while the advantage of deep learning is the ability to learn automatically. The third work of this paper explores the performance of feature automatic learning of deep learning model in question retrieval and answer sorting tasks. The related work was published at the SemEval Conference on 2016. Inspired by the second and third work of this paper, In the fourth work of this paper, we deeply study the CQA system under the framework of deep learning. In this paper, we propose a deep neural network model based on gate mechanism, which can automatically learn the interactive information between question retrieval and answer sorting tasks. In order to further improve the performance of CQA, this paper extensively and deeply studies the CQA system which adopts the traditional natural language processing technology and the deep learning method, the question retrieval and the answer ranking promote each other. A large number of experimental results show that, The strategies proposed in this paper can effectively improve the performance of CQA systems in both traditional methods and depth learning methods.
【學(xué)位授予單位】：華東師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 熊大平;王健;林鴻飛;;一種基于LDA的社區(qū)問(wèn)答問(wèn)句相似度計(jì)算方法[J];中文信息學(xué)報(bào);2012年05期

相關(guān)博士學(xué)位論文 前1條

1 王君澤;基于大規(guī)模問(wèn)答語(yǔ)料的問(wèn)題檢索系統(tǒng)[D];華中科技大學(xué);2010年

相關(guān)碩士學(xué)位論文 前1條

1 文勖;中文問(wèn)答系統(tǒng)中問(wèn)題分類(lèi)及答案候選句抽取的研究[D];哈爾濱工業(yè)大學(xué);2006年

，

本文編號(hào)：1558809