EOSentiMiner: an opinion-aware system based on emotion ontology for sentiment analysis of Chinese online reviews

With the growing availability and popularity of online reviews, consumers' opinions towards certain products or services are generated and spread over the Internet; sentiment analysis thus arises in response to the requirement of opinion seekers. Most prior studies are concerned with statistics-based methods for sentiment classification. These methods, however, suffer from weak comprehension of text-based messages at semantic level, thus resulting in low accuracy. We propose an ontology-based opinion-aware framework – EOSentiMiner – to conduct sentiment analysis for Chinese online reviews from a semantic perspective. The emotion space model is employed to express emotions of reviews in the EOSentiMiner, where sentiment words are classified into two types: emotional words and evaluation words. Furthermore, the former contains eight emotional classes, and the latter is divided into two opinion evaluation classes. An emotion ontology model is then built based on HowNet to express emotion in a fuzzy way. Based on emotion ontology, we evaluate some factors possibly affecting sentiment classification including features of products (services), emotion polarity and intensity, degree words, negative words, rhetoric and punctuation. Finally, sentiment calculation based on emotion ontology is proposed from sentence level to document level. We conduct experiments by using the data from online reviews of cellphone and wedding photography. The result shows the EOSentiMiner outperforms baseline methods in term of accuracy. We also find that emotion expression forms and connection relationship vary across different domains of review corpora.     

基于权值算法的中文情感分析系统研究与实现

为了能够更加准确地对语句结构进行划分、对语句表达的内容进行判断,提出了一种全新的基于权值的计算算法,在完善中文分词的基础上对语句进行情感分析。首先利用中文分词算法对句式结构进行分割,然后依据词性对词库进行扩展,词库对句式中干扰词汇进行过滤,最后利用全新的权值计算算法对语句情感进行准确分析。经有效测试结果表明,情感分析准确率较高,并广泛适用于网络舆情分析等应用中。     

加权维吾尔语句子倾向性分析

准确可靠的文本倾向性分析是网络舆情分析与网络内容安全的前提.本文提出了利用中文极性情感词典HowNet、NTUSD以及大连理工大学发布的褒贬情感词词典进行并交运算,选择并翻译为维吾尔语词汇,借助于维吾尔语同义近义词词典,扩展构建了维吾尔语极性情感词典;然后分析总结了否定词、程度副词以及句中的转折连词等情感修饰成分对维吾尔语句子情感极性的影响,并量化为情感词权值;最后设计了基于维吾尔语极性情感词和权值相结合的加权句子情感极性判定算法.利用自建语料库进行测试,并与汉语倾向性判定实验结果比较,证明了本算法进行维吾尔语句子褒贬情感性分析基本是有效地.     

融合语音情感词局部特征的语音情感识别方法

为有效利用语音情感词局部特征,提出了一种融合情感词局部特征与语音语句全局特征的语音情感识别方法。该方法依赖于语音情感词典的声学特征库,提取出语音语句中是否包含情感词及情感词密度等局部特征,并与全局声学特征进行融合,再通过机器学习算法建模和识别语音情感。对比实验结果表明,融合语音情感词局部特征与全局特征的语音情感识别方法能取得更好的效果,局部特征的引入能有效提高语音情感识别准确率。     

一种基于组合语义的文本情绪分析模型

文本情绪分析属于细颗粒度文本情感分析范畴.传统的基于 监督学习的方法,大多注重从表面词形提取特征,对语言的结构化特征 考虑较少,无法应对特征稀疏问题,也无法挖掘文本中隐含的深层语 言信息(包括词语搭配和语义韵).上述问题的存在导致现有系统 的分类性能不高,尤其对隐性文本情绪分类问题表现出较大的局限 性.本文尝试将基于依存句法的词语搭配特征和基于组合语义的深度 特征应用于文本情绪分类,提出了一种以短语为主要线索的半马 尔科夫条件随机场文本情绪分析模型.为了验证模型的有效性,利 用实际构建的相关实验语料,开展了相关实验研究.实验结果表 明,本文方法不仅可以显著提高文本情绪分类的准确率,而且对解 决隐性情感分析问题也具有重要作用.     

以关键词抽取为核心的文摘句选择策略

针对面向查询的多文档自动文摘,该文提出了一种以关键词抽取为核心的文摘句选择策略。通过查询扩展的相关技术得到相关多文档集中词语的查询相关性特征,利用最大似然估计法得到语料中词语的话题相关性特征,并将这两个特征值进行特征融合得到词语的重要度以确定关键词。然后通过关键词的重要度来给候选句打分,进一步利用改进的MMR(Maximal Marginal Relevance)技术来调整候选句的得分,最后生成文摘。该文将特征融合引入到词语层面,在DUC2005的语料中测试取得了较好的效果。     

Segment-based emotion recognition from continuous Mandarin Chinese speech

Recognition of emotion in speech has recently matured to one of the key disciplines in speech analysis serving next generation human-machine interaction and communication. However, compared to automatic speech recognition, that emotion recognition from an isolated word or a phrase is inappropriate for conversation. Because a complete emotional expression may stride across several sentences, and may fetch-up on any word in dialogue. In this paper, we present a segment-based emotion recognition approach to continuous Mandarin Chinese speech. In this proposed approach, the unit for recognition is not a phrase or a sentence but an emotional expression in dialogue. To that end, the following procedures are presented: First, we evaluate the performance of several classifiers in short sentence speech emotion recognition architectures. The results of the experiments show that the WD-KNN classifier achieves the best accuracy for the 5-class emotion recognition what among the five classification techniques. We then implemented a continuous Mandarin Chinese speech emotion recognition system with an emotion radar chart which is based on WD-KNN; this system can represent the intensity of each emotion component in speech. This proposed approach shows how emotions can be recognized by speech signals, and in turn how emotional states can be visualized.     

一种基于情绪表达与情绪认知分离的新型情绪词典

目前的情绪词典通常对情绪词语进行情绪类别和强度的标注,但缺乏对词语的情绪表达和情绪认知结果进行区分的能力。同时,直接在词语条目上进行标注经常由于词语的语义歧义导致情绪标注结果存在歧义。该文在对个体情绪产生和迁移机制进行分析的基础上,建立了基于“刺激认知—反射表达”的文本情绪计算框架。并在此框架下对情绪相关词语的功能和特性进行分析,探索了一种新型情绪词典建设方法。首先,引入HowNet提供的词语语义信息,将同一词语转变为不同语义的多个词条进行标注减少情绪标注歧义。其次,将词语的情绪表达方式和情绪认知结果加以区分,分别标注从不同角度观测到的词条情绪类别和强度,同时对词语的情绪表达和情绪认知类型进行了细化分类。最终初步构建出一个具有清晰框架、丰富情绪信息和较低歧义的新型情绪词典。     

Improving neural sentence alignment with word translation

Sentence alignment is a basic task in natural lan-guage processing which aims to extract high-quality paral-lel sentences automatically.Motivated by the observation that aligned sentence pairs contain a larger number of aligned words than unaligned ones,we treat word translation as one of the most useful external knowledge.In this paper,we show how to explicitly integrate word translation into neural sentence alignment.Specifically,this paper proposes three cross-lingual encoders to incorporate word translation:1)Mixed Encoder that learns words and their translation annotation vectors over sequences where words and their translations are mixed alterma-tively;2)Factored Encoder that views word translations as fea-tures and encodes words and their translations by concatenating their embeddings;and 3)Gated Encoder that uses gate mechanism to selectively control the amount of word translations moving forward.Experimentation on NIST MT and Opensub-titles Chinese-English datasets on both non-monotonicity and monotonicity scenarios demonstrates that all the proposed encoders significantly improve sentence alignment performance.     

基于声学特征的语言情感识别

语音情感识别是语音处理领域中一个具有挑战性和广泛应用前景的研究课题。探索了语音情感识别中的关键问题之一:生成情感识别的有效的特征表示。从4个角度生成了语音信号中的情感特征表示:(1)低层次的声学特征,包括能量、基频、声音质量、频谱等相关的特征,以及基于这些低层次特征的统计特征;(2)倒谱声学特征根据情感相关的高斯混合模型进行距离转化而得出的特征;(3)声学特征依据声学词典进行转化而得出的特征;(4)声学特征转化为高斯超向量的特征。通过实验比较了各类特征在情感识别上的独立性能,并且尝试了将不同的特征进行融合,最后比较了不同的声学特征在几个不同语言的情感数据集上的效果(包括IEMOCAP英语情感语料库、CASIA汉语情感语料库和Berlin德语情感语料库)。在IEMOCAP数据集上,系统的正确识别率达到了71.9%,超越了之前在此数据集上报告的最好结果。     

中文微博情感分析研究与实现

中文微博的大数据、指数传播和跨媒体等特性,决定了依托人工方式监控和处理中文微博是不现实的,迫切需要依托计算机开展中文微博情感自动分析研究.该项研究可分为3个任务：中文微博观点句识别、情感倾向性分类和情感要素抽取.为完成上述任务,我们研制了一个评测系统：通过构建多级词库、制定成词规则、开展串频统计等给出一种基于规则和统计的新词识别方法,在情感词和评价对象的依存模式的基础上给出基于词语特征的观点句识别算法;以词序流表示文本的LDA-Collocation模型,采用吉布斯抽样法推导了算法,实现中文微博情感倾向性自动分类;针对中文微博情感要素抽取的召回率较低问题,利用依存关系分析理论,按主语类和宾语类把依存模式分为2类,建立了6个优先级的评价对象和情感词汇的依存模式,通过评价对象归并算法实现计算机自动抽取情感要素.实验包括2个部分,一是参加NLPCC2012的公开评测,本文方法在微博观点句识别任务中的准确率为第2,在中文微博情感要素抽取任务中的准确率和F值均为第2,验证了本文算法的实用性.二是在分析公开评测结果的基础上,分别比较了参加公开评测的各类算法在处理中文微博情感分析时的效率,给出本文的结论.     

基于多通道深度学习网络的混合语言短文本情感分类方法

相比于单一语言的短文本情感分类而言,混合语言由于其表达情感的单词语言不唯一,语法结构复杂,仅使用传统词嵌入的方法无法使分类器学到足够有用的特征,导致分类效果不佳。针对这些问题,提出一种融合字词特征的双通道复合模型。首先,针对数据集不平衡问题,提出一种基于Bert语义相似度的数据集欠采样算法;其次,构建双通道深度学习网络,分别将以字、词方式嵌入的原始数据通过两个通道送入CNN和带有注意力机制的LSTM组成的模块中进行多粒度特征提取;最后融合多通道的特征进行分类。在NLPCC2018任务1公布的混合语言五分类数据集上的实验表明,该模型的整体性能较目前有代表性的深度学习模型有进一步提高。     

Discovering and understanding word level user intent in Web search queries

Identifying and interpreting user intent are fundamental to semantic search. In this paper, we investigate the association of intent with individual words of a search query. We propose that words in queries can be classified as either content or intent, where content words represent the central topic of the query, while users add intent words to make their requirements more explicit. We argue that intelligent processing of intent words can be vital to improving the result quality, and in this work we focus on intent word discovery and understanding. Our approach towards intent word detection is motivated by the hypotheses that query intent words satisfy certain distributional properties in large query logs similar to function words in natural language corpora. Following this idea, we first prove the effectiveness of our corpus distributional features, namely, word co-occurrence counts and entropies, towards function word detection for five natural languages. Next, we show that reliable detection of intent words in queries is possible using these same features computed from query logs. To make the distinction between content and intent words more tangible, we additionally provide operational definitions of content and intent words as those words that should match, and those that need not match, respectively, in the text of relevant documents. In addition to a standard evaluation against human annotations, we also provide an alternative validation of our ideas using clickthrough data. Concordance of the two orthogonal evaluation approaches provide further support to our original hypothesis of the existence of two distinct word classes in search queries. Finally, we provide a taxonomy of intent words derived through rigorous manual analysis of large query logs.     

融合显性和隐性特征的中文微博情感分析

微博情感分析是研究社交网络舆情的一项关键技术。微博表情符号和情感词汇等是一类直观显性的情感特征,而微博的内容语义则可视为隐性特征,且对情感判定往往具有决定性作用,因此本文提出将两类特征因素融合的微博情感分析方法。首先构建情感分析词典、网络用语词典以及表情符号库,定义微博频繁特征词集,再根据频繁特征词集,利用最大频繁项集获得微博初始情感簇;针对初始簇间存在文本重叠情况,提出基于短文本扩展语义隶属度的簇间重叠消减算法,获得完全分离的初始簇;最后根据簇语义相似度矩阵,给出一种凝聚式情感聚类方法。利用NLPCC2013 评测所提供的训练语料进行情感分类实验,说明了分析该文方法的性能优势,并以2014年3月8日马航事件微博数据为例,给出了利用微博情感分析公众随事态发展的情感变化,说明了该文方法的实用效果。     

Incorporating language syntax in visual text recognition with astatistical model

The use of a statistical language model to improve the performance of an algorithm for recognizing digital images of handwritten or machine-printed text is discussed. A word recognition algorithm first determines a set of words (called a neighborhood) from a lexicon that are visually similar to each input word image. Syntactic classifications for the words and the transition probabilities between those classifications are input to the Viterbi algorithm. The Viterbi algorithm determines the sequence of syntactic classes (the states of an underlying Markov process) for each sentence that have the maximum a posteriori probability, given the observed neighborhoods. The performance of the word recognition algorithm is improved by removing words from neighborhoods with classes that are not included on the estimated state sequence. An experimental application is demonstrated with a neighborhood generation algorithm that produces a number of guesses about the identity of each word in a running text. The use of zero, first and second order transition probabilities and different levels of noise in estimating the neighborhood are explored     

基于情绪相关事件上下文的隐含情绪分类方法研究

情绪分类是情绪分析研究中的一个基本任务,旨在对文本表达的情绪进行分类。目前,该任务是自然语言处理研究中的一个热点问题。已有的研究一般借助于情绪关键词(例如,“高兴”,“伤心”)来进行情绪分类。然而,在实际中,存在大量的没有情绪关键词但表达情绪的文本,我们称这类情绪表达为隐含情绪表达。该文关注隐含情绪分类方法研究,提出了基于情绪关联事件的隐含情绪分类方法,我们认为情绪的关联事件可以用于对情绪类别进行分类。具体实现中,我们首先采用情绪关键词获得句子群;然后,去除情绪关键词,将上下文作为关联事件表达文本;最后,利用上下文进行情绪分类。实验结果表明,以上下文进行的情绪分类结果达到了一定的性能,远远好于随机分类结果。这一结果为进一步隐含情绪分类提供了良好的基础。     

维吾尔语褒贬情感词典构建研究

为实现维吾尔语网络内容的倾向性分析,进行维吾尔语情感词典的构建研究。首先对现有成果中的情感基准词进行汇总分析,筛选使用频率高、情感倾向强烈的词汇作为维文情感种子词,并利用维文同义词电子词典建立种子扩展词集;其次对HowNet、NTUSD以及大连理工大学开发的情感词典进行并运算,翻译为维吾尔语词汇构成候选词集合;最后利用语料库,计算候选词与种子词以及同义扩展词之间的点互信息值,判别候选词的极性并将其加入到相关的褒贬情感词库中。与汉语句子情感倾向评测实验结果比较,基于该词典的维吾尔语句子倾向性判断准确率和召回率基本相同。     

一种基于情绪激励度的情绪词加权方法

在不同的上下文中,情绪词对情绪的激励程度会发生变化。现有情绪词典中大多数只标注了情绪词的情绪类别而未涉及情绪词的激励度。在极少数标注情绪强度的词典中,所标注的强度未考虑上下文的影响。提出一种根据上下文形成的情境评估情绪词对情绪的激励程度并据此对情绪词加权的方法。通过比较情绪词的共现模式与自身情绪类的分布模式计算情绪词的激励程度。然后根据激励程度计算情绪词的情绪权重并将其用于微博情绪识别。实验结果表明,与现有词典中的情绪强度相比,本文方法计算的情绪权重更准确地描述了情绪词在语料中表达的情绪,有效地提高了情绪分析的精度。并且本文方法还能够有效综合多个词典的优势,进一步提高微博情绪分析的准确率。     

Ontological reasoning for improving the treatment of emotions in text

With the advent of affective computing, the task of adequately identifying, representing and processing the emotional connotations of text has acquired importance. Two problems facing this task are addressed in this paper: the composition of sentence emotion from word emotion, and a representation of emotion that allows easy conversion between existing computational representations. The emotion of a sentence of text should be derived by composition of the emotions of the words in the sentence, but no method has been proposed so far to model this compositionality. Of the various existing approaches for representing emotions, some are better suited for some problems and some for others, but there is no easy way of converting from one to another. This paper presents a system that addresses these two problems by reasoning with two ontologies implemented with Semantic Web technologies: one designed to represent word dependency relations within a sentence, and one designed to represent emotions. The ontology of word dependency relies on roles to represent the way emotional contributions project over word dependencies. By applying automated classification of mark-up results in terms of the emotion ontology the system can interpret unrestricted input in terms of a restricted set of concepts for which particular rules are provided. The rules applied at the end of the process provide configuration parameters for a system for emotional voice synthesis.     

Emotion recognition from speech using global and local prosodic features

In this paper, global and local prosodic features extracted from sentence, word and syllables are proposed for speech emotion or affect recognition. In this work, duration, pitch, and energy values are used to represent the prosodic information, for recognizing the emotions from speech. Global prosodic features represent the gross statistics such as mean, minimum, maximum, standard deviation, and slope of the prosodic contours. Local prosodic features represent the temporal dynamics in the prosody. In this work, global and local prosodic features are analyzed separately and in combination at different levels for the recognition of emotions. In this study, we have also explored the words and syllables at different positions (initial, middle, and final) separately, to analyze their contribution towards the recognition of emotions. In this paper, all the studies are carried out using simulated Telugu emotion speech corpus (IITKGP-SESC). These results are compared with the results of internationally known Berlin emotion speech corpus (Emo-DB). Support vector machines are used to develop the emotion recognition models. The results indicate that, the recognition performance using local prosodic features is better compared to the performance of global prosodic features. Words in the final position of the sentences, syllables in the final position of the words exhibit more emotion discriminative information compared to the words and syllables present in the other positions.