【摘要】：互聯(lián)網(wǎng)作為信息載體聯(lián)通了整個(gè)世界,使得人們可以隨時(shí)隨地獲取和分享資訊。本文基于微博文本,從一個(gè)全新的角度---基于深度學(xué)習(xí)的情感詞向量來挖掘微博的情感表達(dá)。情感詞向量是深度學(xué)習(xí)模型完成情感分析任務(wù)后的副產(chǎn)物,同時(shí)與情感分類準(zhǔn)確率相輔相成。好的情感詞向量輸入是情感分類高效的基礎(chǔ),同時(shí)有監(jiān)督分類效果好同時(shí)又反饋生成高質(zhì)量的情感詞向量。主要從三方面進(jìn)行研究:第一,提出基于橫向卷積和縱向卷積相結(jié)合的卷積神經(jīng)網(wǎng)絡(luò)(HV_CNN),它結(jié)合了動(dòng)態(tài)卷積神經(jīng)網(wǎng)絡(luò)(DCNN)根據(jù)樣本長度不同自動(dòng)提取不同長度特征的優(yōu)點(diǎn),和卷積神經(jīng)網(wǎng)絡(luò)(CNN)詞向量不同維度之間相互聯(lián)系及速度快的優(yōu)點(diǎn)。分別采用詞向量和字向量作為網(wǎng)絡(luò)輸入,實(shí)驗(yàn)不同的模型情感分類效果。第二、分別從語言學(xué)角度利用四種相似度算法計(jì)算相似詞語之間情感極性關(guān)系,以及提升現(xiàn)有模型情感分類效果角度判定不同模型生成的情感詞向量質(zhì)量。第三、提出情感詞向量的兩方面應(yīng)用,一是應(yīng)用于高效的淺層機(jī)器學(xué)習(xí)模型進(jìn)行微博情感分類,避免復(fù)雜的句法分析和特征抽取工作。二是根據(jù)詞語在情感語義空間中的分布進(jìn)行未知極性詞語情感極性判定。分別使用微博語料集和手機(jī)評論語料集做對比試驗(yàn),結(jié)果表明本文提出的HV_CNN模型生成的情感詞向量質(zhì)量高,同時(shí)模型本身情感分類效果突出。
[Abstract]:As a carrier of information, the Internet connects the whole world and enables people to access and share information anytime and anywhere. Based on the Weibo text, this paper exploits the emotional expression of Weibo from a new perspective-emotion word vector based on deep learning. The affective word vector is the by-product of the deep learning model after completing the task of affective analysis and complements with the accuracy of emotion classification. Good input of affective word vector is the basis of high efficiency of affective classification. At the same time, it has good effect of supervised classification and feedback to generate high quality affective word vector. Firstly, a convolution neural network (HV_CNN) based on the combination of transverse convolution and longitudinal convolution is proposed. It combines the advantages of dynamic convolution neural network (DCNN) to automatically extract different length features according to different sample lengths. And convolutional neural network (CNN) word vector between different dimensions and the advantages of fast speed. Word vector and word vector are used as network input, and the effects of emotion classification of different models are tested. Secondly, from the perspective of linguistics, four similarity algorithms are used to calculate the affective polarity relationship between similar words, and to improve the emotional classification effect of existing models to determine the quality of emotional word vectors generated by different models. Thirdly, two applications of affective word vector are proposed. One is to apply the efficient shallow machine learning model to Weibo affective classification to avoid complex syntactic analysis and feature extraction. The second is to judge the emotional polarity of unknown polarity words according to the distribution of words in affective semantic space. By using the Weibo corpus and the mobile phone review corpus respectively, the results show that the HV_CNN model has high quality of emotion word vector and the effect of emotion classification of the model itself is outstanding.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.1;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 宋施恩;樊興華;;基于詞共現(xiàn)和詞上下文的領(lǐng)域觀點(diǎn)詞抽取方法[J];計(jì)算機(jī)工程與設(shè)計(jì);2013年11期

2 魏晶晶;吳曉吟;;電子商務(wù)產(chǎn)品評論多級(jí)情感分析的研究與實(shí)現(xiàn)[J];軟件;2013年09期

3 楊立公;朱儉;湯世平;;文本情感分析綜述[J];計(jì)算機(jī)應(yīng)用;2013年06期

4 黃碩;周延泉;;基于知網(wǎng)和同義詞詞林的詞匯語義傾向計(jì)算[J];軟件;2013年02期

5 林政;譚松波;程學(xué)旗;;基于情感關(guān)鍵句抽取的情感分類研究[J];計(jì)算機(jī)研究與發(fā)展;2012年11期

6 孫志軍;薛磊;許陽明;王正;;深度學(xué)習(xí)研究綜述[J];計(jì)算機(jī)應(yīng)用研究;2012年08期

7 魏椺;向陽;陳千;;中文文本情感分析綜述[J];計(jì)算機(jī)應(yīng)用;2011年12期

8 喬向杰;王志良;王萬森;;基于OCC模型的E-learning系統(tǒng)情感建模[J];計(jì)算機(jī)科學(xué);2010年05期

9 王根;趙軍;;基于多重冗余標(biāo)記CRFs的句子情感分析研究[J];中文信息學(xué)報(bào);2007年05期

10 朱嫣嵐;閔錦;周雅倩;黃萱菁;吳立德;;基于HowNet的詞匯語義傾向計(jì)算[J];中文信息學(xué)報(bào);2006年01期

相關(guān)碩士學(xué)位論文 前2條

1 宋靜靜;中文短文本情感傾向性分析研究[D];重慶理工大學(xué);2013年

2 蔣超;基于用戶聚類和語義詞典的微博推薦系統(tǒng)[D];浙江大學(xué);2013年

，

本文編號(hào)：2166098