本文選題：學術(shù)文獻摘要 + 引文上下文��； 參考：《西北農(nóng)林科技大學》2017年碩士論文

【摘要】：隨著科研產(chǎn)出的不斷增加,爆發(fā)式增長的學術(shù)文獻給學術(shù)研究人員檢索和查閱帶來困難,科研工作量也日益加大。學術(shù)文獻自動摘要方法借助現(xiàn)代計算機技術(shù)自動從海量文獻中高效準確地獲取信息成為當前研究的熱點。本文以引文上下文為研究對象,針對當前基于引文的學術(shù)文獻摘要方法存在的不足,設(shè)計并實現(xiàn)了基于引文上下文的學術(shù)文獻摘要方法,改善了學術(shù)文獻自動摘要的質(zhì)量。本文的主要研究內(nèi)容和成果如下:(1)設(shè)計基于卷積神經(jīng)網(wǎng)絡(luò)的引文上下文分類算法。首先分析學術(shù)文獻的結(jié)構(gòu)特點,提出了基于學術(shù)文獻論述結(jié)構(gòu)的分類模型,解決摘要信息覆蓋不全的問題。為了定位引文上下文在被引文獻中所處的論述點,模擬卷積神經(jīng)網(wǎng)絡(luò)在圖像領(lǐng)域中的應(yīng)用,設(shè)計了簡單的神經(jīng)網(wǎng)絡(luò)模型結(jié)構(gòu),并采用基于深度學習的詞向量表示作為句子的輸入,實現(xiàn)基于卷積神經(jīng)網(wǎng)絡(luò)的引文上下文分類。選擇CNN-static和CNN-non-static兩種模式與傳統(tǒng)的基于線性SVM的分類算法進行對比實驗,實驗結(jié)果表明基于卷積神經(jīng)網(wǎng)絡(luò)的引文上下文分類在兩種模式下都取得較高的準確率,其中CNN-non-static的準確率最高達79.91%,整體平均提高了3.12%,能有效解決引文上下文論述點的分布問題。摘要評測結(jié)果也證實該方法提高了摘要的信息量和可讀性。(2)構(gòu)建基于向量空間模型的引文上下文抽取算法。分析引文與引文上下文之間的關(guān)系,對引文句子進行特征選擇與權(quán)值計算,構(gòu)建引文和被引文獻的向量空間模型,采用余弦距離度量引文和引文上下文之間的語義關(guān)系,實現(xiàn)利用引文從被引文獻中抽取引文上下文。抽取結(jié)果表明,引文和引文上下文表述一致性較低,抽取準確率較低,總體均值為17.66%。同時也說明引文并不能準確反映被引文獻,基于引文的學術(shù)文獻摘要存在與被引文獻信息不一致的缺陷。(3)提出基于圖的句子重要性排序改進算法。傳統(tǒng)的圖排序僅僅只考慮句子之間的重要性,并沒有考慮句子之間的冗余性,導(dǎo)致生成的摘要存在信息冗余的問題。借助句子之間詞的語義位置關(guān)系,并結(jié)合句子之間的語義相似關(guān)系對句子的冗余性進行評估,加權(quán)句子的重要性和冗余性對其綜合打分排序,解決了摘要句子信息冗余的問題。摘要評測結(jié)果表明,改進方法提高了摘要的Rouge評測值,改善了摘要質(zhì)量,使得摘要更貼近標準摘要。
[Abstract]:With the increasing of scientific research output, the explosive growth of academic literature makes it difficult for academic researchers to search and consult, and the workload of scientific research is increasing day by day.With the help of modern computer technology, the automatic abstracting method of academic literature has become a hot topic in current research.In this paper, the citation context is taken as the research object. Aiming at the shortcomings of the current citation based abstracting methods of academic literature, a citation context based approach is designed and implemented to improve the quality of the automatic abstracts of academic documents.The main contents and results of this paper are as follows: (1) A citation context classification algorithm based on convolutional neural network is designed.Firstly, the structural characteristics of academic literature are analyzed, and a classification model based on the structure of academic literature is proposed to solve the problem of incomplete coverage of summary information.In order to locate the argumentation of citation context in cited literature and simulate the application of convolutional neural network in image field, a simple neural network model structure is designed.Furthermore, the word vector representation based on deep learning is used as the input of sentences to realize the classification of citation context based on convolutional neural network.CNN-static and CNN-non-static are selected to compare with the traditional classification algorithm based on linear SVM. The experimental results show that the citation context classification based on convolution neural network achieves high accuracy in both modes.The accuracy of CNN-non-static is as high as 79.91, and the whole average increase is 3.12, which can effectively solve the problem of the distribution of citation contextual argumentation points.The evaluation results also confirm that the proposed method improves the information and readability of the abstract, and constructs a citation context extraction algorithm based on vector space model.The relationship between citation and citation context is analyzed, the feature selection and weight calculation of citation sentences are carried out, the vector space model of citation and citation is constructed, and the semantic relationship between citation and citation context is measured by cosine distance.The context of citation is extracted from the cited document by citation.The extraction results show that the consistency of citation and citation context is low, the extraction accuracy is low, and the total mean value is 17.66.It also shows that citation can not accurately reflect the cited literature, and the citation based abstracts of academic documents have the defect of inconsistent with the cited literature information. (3) an improved algorithm of sentence importance ranking based on graph is proposed.The traditional graph sorting only considers the importance of sentences, and does not consider the redundancy of sentences, which leads to the problem of information redundancy in the generated abstracts.The redundancy of sentences is evaluated by the semantic position relation of words between sentences and the semantic similarity between sentences. The importance and redundancy of weighted sentences are comprehensively graded to solve the problem of redundancy of summary sentence information.The evaluation results show that the improved method improves the Rouge evaluation value of the summary, improves the quality of the summary, and makes the summary more close to the standard summary.
【學位授予單位】：西北農(nóng)林科技大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP391.1

【參考文獻】

相關(guān)期刊論文 前10條

1 陳海華;黃永;張炯;陸偉;;基于引文上下文的學術(shù)文本自動摘要技術(shù)研究[J];數(shù)字圖書館論壇;2016年08期

2 劉天yN;步一;趙丹群;黃文彬;;自動引文摘要研究述評[J];現(xiàn)代圖書情報技術(shù);2016年05期

3 莫鵬;胡珀;黃湘冀;何婷婷;;基于超圖的文本摘要與關(guān)鍵詞協(xié)同抽取研究[J];中文信息學報;2015年06期

4 李峰;黃金柱;李舟軍;楊偉銘;;使用關(guān)鍵詞擴展的新聞文本自動摘要方法[J];計算機科學與探索;2016年03期

5 郭麗麗;丁世飛;;深度學習研究進展[J];計算機科學;2015年05期

6 程園;吾守爾·斯拉木;買買提依明·哈斯木;;基于綜合的句子特征的文本自動摘要[J];計算機科學;2015年04期

7 李強;鄭宇杰;;基于多視角特征融合與隨機森林的蛋白質(zhì)結(jié)晶預(yù)測[J];現(xiàn)代電子技術(shù);2015年08期

8 肖曉;張敏;;支持向量機多分類問題研究[J];淮海工學院學報(自然科學版);2014年03期

9 肖升;何炎祥;;改進的潛在語義分析中文摘錄方法[J];計算機應(yīng)用研究;2012年12期

10 衛(wèi)佳君;宋繼華;;自動文摘的方法研究[J];計算機技術(shù)與發(fā)展;2011年08期

相關(guān)會議論文 前1條

1 劉盛博;丁X;;基于引用內(nèi)容的引文評價分析[A];第九屆中國科技政策與管理學術(shù)年會論文集[C];2013年

相關(guān)博士學位論文 前3條

1 張琳;基于引用聚類的多文檔自動文摘技術(shù)研究[D];大連海事大學;2013年

2 胡珀;融合上下文信息的自動文摘研究[D];武漢大學;2013年

3 劉娜;文本自動摘要和信息抽取方法及其應(yīng)用研究[D];大連海事大學;2012年

相關(guān)碩士學位論文 前2條

1 任瀟雨;基于引文的英文文檔文摘自動生成方法研究[D];西安電子科技大學;2014年

2 石子言;用戶驅(qū)動的特定領(lǐng)域自動文摘系統(tǒng)設(shè)計與實現(xiàn)[D];東北師范大學;2009年

，

本文編號：1773015