基于质心迁移的领域适应性情感分类

针对监督学习方法在文本的跨领域情感分析效果较差的问题,提出基于质心迁移的领域间适应性情感分类方法。该方法利用源领域的标注文本对目标领域的大量未标注文本进行分类,选择一部分可信度高的文本加入到训练集,同时去除源领域中距离目标领域测试集质心较远的文本,通过迭代逐渐缩小两个领域间的质心距离,减小领域间差异。实验结果表明,该方法能提高跨领域倾向性分析的精度。     

Sentimental causal rule discovery from Twitter

Social media, especially Twitter is now one of the most popular platforms where people can freely express their opinion. However, it is difficult to extract important summary information from many millions of tweets sent every hour. In this work we propose a new concept, sentimental causal rules, and techniques for extracting sentimental causal rules from textual data sources such as Twitter which combine sentiment analysis and causal rule discovery. Sentiment analysis refers to the task of extracting public sentiment from textual data. The value in sentiment analysis lies in its ability to reflect popularly voiced perceptions that are stated in natural language. Causal rules on the other hand indicate associations between different concepts in a context where one (or several concepts) cause(s) the other(s). We believe that sentimental causal rules are an effective summarization mechanism that combine causal relations among different aspects extracted from textual data as well as the sentiment embedded in these causal relationships. In order to show the effectiveness of sentimental causal rules, we have conducted experiments on Twitter data collected on the Kurdish political issue in Turkey which has been an ongoing heated public debate for many years. Our experiments on Twitter data show that sentimental causal rule discovery is an effective method to summarize information about important aspects of an issue in Twitter which may further be used by politicians for better policy making.     

融合多种情感资源的微博情感分类研究

为了通过融合多种情感资源库中的词汇情感特征来提高微博情感分类精度,提出了一种词汇情感确定性度量的计算方法,并以此为基础将在多种情感词汇上获取的情感特征融合为词汇的综合情感特征,然后采用机器学习的分类方法实现微博观点句识别和观点句情感倾向性判定.实验表明,本方法利用词汇的情感确定性度量,统一了词汇情感倾向性的强度度量,在观点句识别和观点句情感倾向性判定两个情感分类任务中都取得了较好的性能.     

Twitter brand sentiment analysis: A hybrid system using n-gram analysis and dynamic artificial neural network

Twitter messages are increasingly used to determine consumer sentiment towards a brand. The existing literature on Twitter sentiment analysis uses various feature sets and methods, many of which are adapted from more traditional text classification problems. In this research, we introduce an approach to supervised feature reduction using n-grams and statistical analysis to develop a Twitter-specific lexicon for sentiment analysis. We augment this reduced Twitter-specific lexicon with brand-specific terms for brand-related tweets. We show that the reduced lexicon set, while significantly smaller (only 187 features), reduces modeling complexity, maintains a high degree of coverage over our Twitter corpus, and yields improved sentiment classification accuracy. To demonstrate the effectiveness of the devised Twitter-specific lexicon compared to a traditional sentiment lexicon, we develop comparable sentiment classification models using SVM. We show that the Twitter-specific lexicon is significantly more effective in terms of classification recall and accuracy metrics. We then develop sentiment classification models using the Twitter-specific lexicon and the DAN2 machine learning approach, which has demonstrated success in other text classification problems. We show that DAN2 produces more accurate sentiment classification results than SVM while using the same Twitter-specific lexicon.     

网络意见挖掘、摘要与检索研究综述

网络上带有人的主观感情色彩的评论性文本反映了人们的意见、态度和立场,因而具有很大的利用价值.信息挖掘技术针对这些主观文本进行处理,获得有用的意见、结论和知识.首先介绍了意见挖掘出现的背景和应用意义,然后从词汇情感极性识别、粗粒度的情感分类、细粒度的意见挖掘与摘要、意见检索和相关语言资源与系统5个方面综述了研究历程和现状,最后总结了研究难点与研究趋势.     

基于树核函数的英文句子情感分类研究

最近几年,由于在线客户评论信息飞快地增长。如何把这些信息分类为正向和负向情感是一个迫切需要解决的问题。提出了一种细粒度级别(句子级别)的情感分类方法,该方法在SVM分类器中使用了树核和复合核函数来进行句子级别情感的分类。实验结果表明在句子级别的情感分类中树核和复合核的方法比线性核具有更佳的性能。     

Damaged corporate reputation: Can celebrity Tweets repair it?

These days, many corporations engage in Twitter activities as a part of their communication strategy. Corporations can use this medium to share information with stakeholders, to answer customer questions, or to build on their image. In this study we examined the extent to which celebrity Tweet messages can be used to repair a damaged corporate reputation, and how this message should be designed and what celebrity should be ‘used’.In two experiments, a 2 × 2 (attractive celebrity versus intelligent celebrity) × (personal message versus general message) design was used. In total, 163 respondents first expressed their feelings regarding the two organisations in a baseline reputation measurement (M = 4.72 on 7 point Likert scale). After that a news items was presented communicating a big fraud and mismanagement, resulting in a decreased reputation score (M = 4.10). In the final stage one of the four experimental Tweets was presented, aimed at repairing the damaged reputation, which succeeded (M = 4.43). For both organisations, the crisis prime significantly decreased reputation scores, and the Tweet significantly increased reputation score again. The analysis of variance shows a main effect for type of celebrity. In our experiment the intelligent celebrity’s Tweet was best to use.The study reveals that celebrities’ Tweets can restore a positive public opinion about corporations. This study shows that when it comes to serious matters, an intelligent celebrity, who has the best fit with the topic, is of best impact. Consequences for corporate communication and future research are discussed.     

Sense-level subjectivity in a multilingual setting

Recent research on English word sense subjectivity has shown that the subjective aspect of an entity is a characteristic that is better delineated at the sense level, instead of the traditional word level. In this paper, we seek to explore whether senses aligned across languages exhibit this trait consistently, and if this is the case, we investigate how this property can be leveraged in an automatic fashion. We first conduct a manual annotation study to gauge whether the subjectivity trait of a sense can be robustly transferred across language boundaries. An automatic framework is then introduced that is able to predict subjectivity labeling for unseen senses using either cross-lingual or multilingual training enhanced with bootstrapping. We show that the multilingual model consistently outperforms the cross-lingual one, with an accuracy of over 73% across all iterations.     

使用最大熵模型进行中文文本分类

随着WWW的迅猛发展,文本分类成为处理和组织大量文档数据的关键技术．由于最大熵模型可以综合观察到各种相关或不相关的概率知识,对许多问题的处理都可以达到较好的结果．但是,将最大熵模型应用在文本分类中的研究却非常少,而使用最大熵模型进行中文文本分类的研究尚未见到．使用最大熵模型进行了中文文本分类．通过实验比较和分析了不同的中文文本特征生成方法、不同的特征数目,以及在使用平滑技术的情况下,基于最大熵模型的分类器的分类性能．并且将其和Baves,KNN,SVM三种典型的文本分类器进行了比较,结果显示它的分类性能胜于Bayes方法,与KNN和SVM方法相当,表明这是一种非常有前途的文本分类方法．     

用于情感分类的双向深度LSTM

对商品、电影等的评论的体现人们对商品的喜好程度,从而为意向购买该商品的人提供参考,也有助于商家调整橱窗货品以取得最大利润。近年来,深度学习在文本上强大的表示和学习能力为理解文本语义、抓取文本所蕴含的情感倾向提供了极好的支持,特别是深度学习中的长短记忆模型(Long Short-Term Memory,LSTM)。评论是一种时序数据形式,通过单词前向排列来表达语义信息。而LSTM恰好是时序模型,可以前向读取评论,并把它编码到一个实数向量中,该向量隐含了评论的潜在语义,可以被计算机存储和处理。利用两个LSTM模型分别从前、后两个方向读取评论,从而获取评论的双向语义信息;再通过层叠多层双向LSTM来达到获取评论深层特征的目的;最后把这个模型放到一个情感分类模型中,以实现情感分类任务。实验证明,该模型相对基准LSTM取得了更好的实验效果,这表示双向深度LSTM能抓取更准确的文本信息。将双向深度LSTM模型和卷积神经网络(Convolutional Neural Network,CNN)进行实验对比,结果表明双向深度LSTM模型同样取得了更好的效果。     

英语情态句的情感倾向性分析

该文研究了英语情态句的情感倾向性分析问题。情态句是英语中的常用句型,在用户评论文本中占有很大的比例。由于其独有的语言学特点,情态句中的情感倾向很难被已有的方法有效地分析。在该文中,我们借助词性标签进行了情态句的识别,并提出了一种情态特征用于帮助情态句情感倾向性的分析。为了进一步提高分析效果,我们还给出了通过合并同义情态特征来缓解情态特征稀疏性问题的方法。实验结果表明,在二元及三元情感倾向性分类问题上,该文提出的方法在F值上较经典分类方法分别有4%及7%的提高。     

使用最大熵模型进行文本分类

最大熵模型是一种在广泛应用于自然语言处理中的概率估计方法。文中使用最大熵模型进行了文本分类的研究。通过实验,将其和Bayes、KNN、SVM三种典型的文本分类器进行了比较,并且考虑了不同特征数目和平滑技术对基于最大熵模型的文本分类器的影响。结果显示它的分类性能胜于Bayes方法,与KNN和SVM方法相当,表明这是一种非常有前途的文本分类方法。     

基于语义理解和机器学习的混合的中文文本情感分类算法框架

快速、准确和全面地从大量互联网文本信息中定位情感倾向是当前大数据技术领域面临的一大挑战.文本情感分类方法大致分为基于语义理解和基于有监督的机器学习两类.语义理解处理情感分类的优势在于其对不同领域的文本都可以进行情感分类,但容易受到中文存在的不同句式及搭配的影响,分类精度不高.有监督的机器学习虽然能够达到比较高的情感分类精度,但在一个领域方面得到较高分类能力的分类器不适应新领域的情感分类.在使用信息增益对高维文本做特征降维的基础上,将优化的语义理解和机器学习相结合,设计了一种新的混合语义理解的机器学习中文情感分类算法框架.基于该框架的多组对比实验验证了文本信息在不同领域中高且稳定的分类精度.     

基于Word2Vec的情感词典自动构建与优化

情感词典的构建是文本挖掘领域中重要的基础性工作。近几年,情感词典的极性标注从二元褒贬标注向多元情绪标注发展,词典的领域特性也日趋明显。但是情感类别的手工标注不但费时费力,而且情感强度难以得到准确量化,同时对领域性的过分关注也大大限制了情感词典的适用性[1]。通过神经网络语言模型对大规模中文语料进行统计训练,并在此基础上提出了基于转换约束集的多维情感词典自动构建方法;然后研究了基于词分布密度的感情色彩消歧方法,对兼具褒贬意味词语的感情极性进行区分和识别,并分别计算两种感情色彩下的情感类别与强度;最后提出基于多个语义资源的全局优化方案,得到包含10种情绪标注的多维汉语情感词典SentiRuc。实验证实该词典¹⁾在类别标注检验、强度标注检验、情感消歧效果及情感分类任务中均具有良好的效果,其中的情感强度检验证实该词典具有极强的情感语义描述力。     

基于双向路径约束模型的XML文档结构相似性度量

为了弥补树编辑距离方法时间复杂度高和频繁路径方法丢失过多语义信息的不足,建立XML文档的双向路径约束模型,从而更全面地提取XML文档的结构信息,为更精确的XML相似度计算打下基础.引入自然语言领域中成熟的N-Gram思想,将基于N-Gram的划分方式应用在路径约束相似度计算中,加快了计算效率和精确度.运用正整数和各种权值简化N-Gram信息的提取和运算.实验结果表明,方法提高了聚类的准确率和召回率.     

基于情感分析的酒店顾客满意度评估

随着信息技术发展和社交平台多方面渗入,在线评论凭借真实客观的优点已成为商家和消费者的主要信息来源。结合TF-IDF、K-means算法获取酒店顾客满意度影响因素;采用基于监督学习的分类算法与百度自然语言处理API得出情感极性值;利用多元线性回归建立满意度评估模型,并将该模型应用于\"2019哈尔滨冰上冰雪嘉年华\"期间754家酒店的满意度分析中。研究结果表明:与现有酒店预订平台相比,该评价体系更加客观全面,顾客对此阶段的酒店总体满意度中等偏上,单维度分析中,总体感受是顾客最为关注的问题,但是酒店设施和卫生环境等因素也会影响酒店顾客满意度的提升。