本文關(guān)鍵詞：基于多元特征融合和LSTM神經(jīng)網(wǎng)絡(luò)的中文評(píng)論情感分析　出處：《太原理工大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 情感分析 情感特征 權(quán)重信息 多元特征 LSTM神經(jīng)網(wǎng)絡(luò)

【摘要】：隨著移動(dòng)互聯(lián)網(wǎng)的飛速發(fā)展,網(wǎng)購(gòu)成了人們?nèi)粘Ｉ畹囊徊糠�。電商網(wǎng)站上存在大量的產(chǎn)品評(píng)論信息。挖掘這些評(píng)論的情感傾向不僅可以為商家提供商品的各種信息,方便商家做出銷售決策,也有利于買家對(duì)商品做出客觀判斷,從而做出購(gòu)買決策。面對(duì)數(shù)量龐大的評(píng)論文本信息,僅靠人工瀏覽去獲取評(píng)論的情感傾向費(fèi)時(shí)且費(fèi)力,如何利用人工智能領(lǐng)域中的相關(guān)技術(shù)對(duì)產(chǎn)品評(píng)論自動(dòng)化地進(jìn)行情感分析成為了一個(gè)重要且有意義的課題�，F(xiàn)有的情感分析方法主要有基于規(guī)則的方法、基于機(jī)器學(xué)習(xí)的方法和基于深度神經(jīng)網(wǎng)絡(luò)的方法,隨著大數(shù)據(jù)技術(shù)的發(fā)展以及語(yǔ)言的形式越來(lái)越多元化,深度神經(jīng)網(wǎng)絡(luò)技術(shù)成為了自然語(yǔ)言處理領(lǐng)域的主流技術(shù),在情感分析領(lǐng)域也取得了很大的突破,本文主要研究基于深度神經(jīng)網(wǎng)絡(luò)的情感分析方法。本文的主要研究工作如下:(1)針對(duì)文本情感分析中對(duì)文本表示時(shí)遇到的維度過(guò)高和語(yǔ)義不相關(guān)的問(wèn)題,本文采用word embedding機(jī)制,通過(guò)神經(jīng)網(wǎng)絡(luò)語(yǔ)言模型對(duì)大量評(píng)論文本進(jìn)行訓(xùn)練,并在此基礎(chǔ)上通過(guò)distributed representation的方式表示文本,從而將文本映射為一個(gè)低維實(shí)數(shù)向量。這種文本表示方法同時(shí)也可以表達(dá)文本的語(yǔ)義信息,有助于神經(jīng)網(wǎng)絡(luò)模型對(duì)文本更好地理解。(2)針對(duì)情感分析任務(wù)的特殊性,本文提出了一種新的文本表示方法-——多元特征詞向量。這種表示方法是對(duì)distributed representation表示方法的優(yōu)化�？紤]到情感分析中含有情感要素的詞對(duì)文本整體情感極性的影響,通過(guò)構(gòu)建情感要素詞典捕捉文本中含有情感要素的詞,并通過(guò)構(gòu)造詞的情感特征向量來(lái)表達(dá)詞的情感要素,接著與用distributed representation方式表示的詞向量進(jìn)行特征融合構(gòu)成多元特征詞向量。用多元特征詞向量表示的文本不僅含有文本的語(yǔ)義信息,而且可以捕捉文本的情感特征,更適合情感分析任務(wù)。(3)情感分析的本質(zhì)是一個(gè)分類問(wèn)題,計(jì)算特征權(quán)重是分類問(wèn)題的重要步驟,基于此理論,本文在提出的多元特征詞向量的基礎(chǔ)上,進(jìn)一步對(duì)其優(yōu)化,借鑒特征權(quán)重算法為多元特征詞向量分配權(quán)重,從而突出對(duì)分類更重要的詞。本文提出的基于權(quán)重分配的多元特征詞向量的文本表示方法對(duì)傳統(tǒng)的文本表示方法從兩方面進(jìn)行了改進(jìn),豐富了對(duì)文本語(yǔ)義的表達(dá),將其作為神經(jīng)網(wǎng)絡(luò)分類模型的輸入,更適合神經(jīng)網(wǎng)絡(luò)模型對(duì)文本進(jìn)行深層次特征捕捉與情感分類。(4)本文使用LSTM神經(jīng)網(wǎng)絡(luò)模型挖掘文本的深層特征。用基于權(quán)重分配的多元特征詞向量表示文本,并作為L(zhǎng)STM神經(jīng)網(wǎng)絡(luò)模型的輸入,然后利用LSTM神經(jīng)網(wǎng)絡(luò)能夠?qū)W習(xí)文本中遠(yuǎn)距離依賴的特性捕捉文本的序列特征及上下文的依賴關(guān)系。最后本文通過(guò)和傳統(tǒng)的基于LSTM神經(jīng)網(wǎng)絡(luò)的情感分析方法做對(duì)比實(shí)驗(yàn),驗(yàn)證本文提出的改進(jìn)方案的有效性。在上述四個(gè)工作中,本文充分考慮情感分析任務(wù)的特性,將情感詞典資源以及特征權(quán)重信息等先驗(yàn)知識(shí)引入神經(jīng)網(wǎng)絡(luò)模型,在此基礎(chǔ)上提出的基于權(quán)重分配的多元特征詞向量可以捕捉更適用于情感分析任務(wù)的特征,利用LSTM神經(jīng)網(wǎng)絡(luò)模型的特性可以捕捉更豐富的特征組合,從而有效提高情感分類模型對(duì)文本的理解以及情感分類的準(zhǔn)確率。
[Abstract]:With the rapid development of mobile Internet, online shopping has become a part of people's daily life. There is a lot of product comment information on the e-commerce website. Mining these reviews' emotional tendencies can not only provide businesses with various kinds of information, facilitate businesses to make sales decisions, but also help buyers make objective judgments on goods, so as to make purchase decisions. Faced with the huge amount of comment text information, it is time-consuming and laborious to get emotional sentiment of reviews only by manual browsing. How to make use of the related technology in artificial intelligence to automatically analyze the product reviews has become an important and meaningful topic. The emotion of the existing analysis methods are mainly based on rule based methods, machine learning method and method based on the depth of the neural network, with the development of big data technology and language in the form of more and more diversified, the depth of the neural network technology has become a mainstream technology in the field of Natural Language Processing, in the field of sentiment analysis has made great breakthrough in this paper study on the depth of the neural network analysis method based on emotion. The main research work of this paper is as follows: (1) according to the related encountered problems of high dimension and semantic representation for text sentiment analysis, this paper uses the word embedding mechanism, the neural network model of language training on a large number of comments and text, on the basis of the distributed representation representation of the text, which maps text as a low dimensional real vector. This method of text representation can also express the semantic information of the text, which helps the neural network model to understand the text better. (2) in view of the particularity of the emotional analysis task, a new method of text representation, multi feature word vector, is proposed in this paper. This representation is an optimization of the distributed representation representation method. Considering the effect of sentiment analysis contains the emotional factors of the word polarity on the whole text by emotion, emotional factors to construct the emotional elements containing the words in the dictionary to capture the text, and through the emotional feature vector to construct the word to express emotion words, then the features are fused to form multiple feature vectors and vector expressed by distributed representation the way. The text expressed with multiple feature words not only contains the semantic information of the text, but also can capture the emotional features of the text, which is more suitable for the emotional analysis task. (3) the nature of sentiment analysis is a classification problem, feature weight calculation is an important step in the classification problem, based on this theory, based on the characteristics of multi word vector proposed on the further optimization, using feature weighting algorithm for multi feature vector weights, which are more important to the classification of words. This text multi term vector representation method based on the weight distribution on the traditional text representation methods are improved from two aspects, enrich the expression of the text, as the neural network classification model is more suitable for the input of the neural network model of the deep features capture and sentiment classification. (4) this paper uses the LSTM neural network model to excavate the deep features of the text. We use weight based multi feature vector to represent text and use it as input of LSTM neural network model. Then we use LSTM neural network to learn the characteristics of long distance dependency in text, and capture the sequence characteristics and Contextual Dependency of text. Finally, this paper compares the traditional LSTM neural network based affective analysis method to verify the effectiveness of the proposed scheme. In the four work, considering the characteristics of sentiment analysis tasks, emotional dictionary resources and feature weight information such as prior knowledge into the neural network model is proposed based on multiple feature vector based on weight distribution can capture more suitable features in sentiment analysis tasks, using the characteristics of LSTM neural network model you can capture the feature combination more abundant, so as to effectively improve the accuracy of text classification model of understanding emotion and sentiment classification.
【學(xué)位授予單位】：太原理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.1;TP18

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 朱艷輝;張永平;杜銳;徐葉強(qiáng);;基于LDA與評(píng)價(jià)對(duì)象的微博觀點(diǎn)摘要[J];鄭州大學(xué)學(xué)報(bào)(理學(xué)版);2017年01期

2 楊小平;張中夏;王良;張永俊;馬奇鳳;吳佳楠;張悅;;基于Word2Vec的情感詞典自動(dòng)構(gòu)建與優(yōu)化[J];計(jì)算機(jī)科學(xué);2017年01期

3 劉艷梅;;深度學(xué)習(xí)技術(shù)下的中文微博情感的分析與研究[J];軟件;2016年05期

4 劉勇華;李愛(ài)萍;段利國(guó);邸鵬;王鴻翔;;以主觀線索為特征的主觀性文本識(shí)別[J];計(jì)算機(jī)工程與設(shè)計(jì);2015年09期

5 梁軍;柴玉梅;原慧斌;高明磊;昝紅英;;基于極性轉(zhuǎn)移和LSTM遞歸網(wǎng)絡(luò)的情感分析[J];中文信息學(xué)報(bào);2015年05期

6 張佳明;席耀一;王波;唐浩浩;李天彩;;基于詞向量的微博事件追蹤方法[J];計(jì)算機(jī)工程與應(yīng)用;2016年17期

7 汝承森;饒嵐;王挺;;網(wǎng)絡(luò)信息中評(píng)價(jià)搭配識(shí)別及傾向性判斷[J];計(jì)算機(jī)科學(xué);2013年07期

8 劉歡歡;李壽山;周國(guó)棟;李逸薇;;中文情緒識(shí)別方法研究[J];江西師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年02期

9 陸文星;王燕飛;;中文文本情感分析研究綜述[J];計(jì)算機(jī)應(yīng)用研究;2012年06期

10 夏火松;劉建;朱慧毅;;中文情感分類挖掘預(yù)處理關(guān)鍵技術(shù)比較研究[J];情報(bào)雜志;2011年09期

相關(guān)博士學(xué)位論文 前1條

1 萬(wàn)源;基于語(yǔ)義統(tǒng)計(jì)分析的網(wǎng)絡(luò)輿情挖掘技術(shù)研究[D];武漢理工大學(xué);2012年

相關(guān)碩士學(xué)位論文 前10條

1 王飛飛;基于改進(jìn)卷積神經(jīng)網(wǎng)絡(luò)算法的研究與應(yīng)用[D];南京郵電大學(xué);2016年

2 隋浩;基于Word2Vec的微博情感新詞識(shí)別與傾向判斷研究[D];廣西大學(xué);2016年

3 廖純;基于句法和語(yǔ)義的話題細(xì)粒度情感分析的研究[D];北京理工大學(xué);2016年

4 楊云;基于句法結(jié)構(gòu)的評(píng)價(jià)對(duì)象抽取方法研究[D];東北師范大學(xué);2015年

5 于永波;Web信息抽取中的若干關(guān)鍵問(wèn)題研究[D];中國(guó)科學(xué)技術(shù)大學(xué);2015年

6 趙亞紅;面向多類標(biāo)分類的隨機(jī)森林算法研究[D];哈爾濱工業(yè)大學(xué);2014年

7 朱少杰;基于深度學(xué)習(xí)的文本情感分類研究[D];哈爾濱工業(yè)大學(xué);2014年

8 李惠;組合型中文分詞方法的研究[D];廣東工業(yè)大學(xué);2014年

9 雷龍艷;中文微博細(xì)粒度情緒識(shí)別研究[D];南華大學(xué);2014年

10 陸洋;基于語(yǔ)義分析的文本挖掘研究[D];浙江工業(yè)大學(xué);2012年

，

本文編號(hào)：1339147