SentiMI: Introducing point-wise mutual information with SentiWordNet to improve sentiment polarity detection

Supervised learning has attracted much attention in recent years. As a consequence, many of the state-of-the-art algorithms are domain dependent as they require a labeled training corpus to learn the domain features. This requires the availability of labeled corpora which is a cumbersome task in itself. However, for text sentiment detection SentiWordNet (SWN) may be used. It is a vocabulary where terms are arranged in synonym groups called synsets. This research makes use of SentiWordNet and treats it as the labeled corpus for training. A sentiment dictionary, SentiMI, builds upon the mutual information calculated from these terms. A complete framework is developed by using feature selection and extracting mutual information, from SentiMI, for the selected features. Training, testing and evaluation of the proposed framework are conducted on a large dataset of 50,000 movie reviews. A notable performance improvement of 7% in accuracy, 14% in specificity, and 8% in F-measure is achieved by the proposed framework as compared to the baseline SentiWordNet classifier. Comparison with the state-of-the-art classifiers is also performed on widely used Cornell Movie Review dataset which also proves the effectiveness of the proposed approach.     

基于概率图模型的文本对象情感分析

情感分析旨在从文本数据中自动识别主观情感,即文本中表达的观点、态度、感受等,在线评论通常都涉及特定的对象,通过在JST模型基础上加入对象层提出了一种无监督的对象情感联合模型(UOSU model),UOSU模型对每个词同时采样对象、情感和主题标签,最终得到各个主题的对象情感词以及文本的对象情感分布。在汽车评论数据集上进行的情感分类实验取得了74.19%的精确率和73.97%的召回率。     

Prominent feature extraction for review analysis: an empirical study

Sentiment analysis (SA) research has increased tremendously in recent times. SA aims to determine the sentiment orientation of a given text into positive or negative polarity. Motivation for SA research is the need for the industry to know the opinion of the users about their product from online portals, blogs, discussion boards and reviews and so on. Efficient features need to be extracted for machine-learning algorithm for better sentiment classification. In this paper, initially various features are extracted such as unigrams, bi-grams and dependency features from the text. In addition, new bi-tagged features are also extracted that conform to predefined part-of-speech patterns. Furthermore, various composite features are created using these features. Information gain (IG) and minimum redundancy maximum relevancy (mRMR) feature selection methods are used to eliminate the noisy and irrelevant features from the feature vector. Finally, machine-learning algorithms are used for classifying the review document into positive or negative class. Effects of different categories of features are investigated on four standard data-sets, namely, movie review and product (book, DVD and electronics) review data-sets. Experimental results show that composite features created from prominent features of unigram and bi-tagged features perform better than other features for sentiment classification. mRMR is a better feature selection method as compared with IG for sentiment classification. Boolean Multinomial Naïve Bayes) algorithm performs better than support vector machine classifier for SA in terms of accuracy and execution time.     

基于深度表示学习和高斯过程迁移学习的情感分析方法

情感分析是自然语言处理领域的重要研究问题。现有方法往往难以克服样本偏置与领域依赖问题,严重制约了情感分析的发展和应用。为此,该文提出了一种基于深度表示学习和高斯过程知识迁移学习的情感分析方法。该方法首先利用深度神经网络获得文本样本的分布式表示,而后基于深度高斯过程,从辅助数据中迁移与测试集数据分布相符的高质量样例扩充训练数据集用于分类器训练,以此提高文本情感分类系统性能。在COAE2014文本情感分类数据集上进行的实验结果显示,该文提出的方法可以有效提高文本情感分类性能,同时可以有效缓解训练数据的样本偏置以及领域依赖问题的影响。     

基于特征选择和深度信念网络的文本情感分类算法

由于人类语言的复杂性，文本情感分类算法大多都存在因为冗余而造成的词汇量过大的问题。深度信念网络（DBN）通过学习输入语料中的有用信息以及它的几个隐藏层来解决这个问题。然而对于大型应用程序来说，DBN是一个耗时且计算代价昂贵的算法。针对这个问题，提出了一种半监督的情感分类算法，即基于特征选择和深度信念网络的文本情感分类算法（FSDBN）。首先使用特征选择方法（文档频率（DF）、信息增益（IG）、卡方统计（CHI）、互信息（MI））过滤掉一些不相关的特征从而使词汇表的复杂性降低；然后将特征选择的结果输入到DBN中，使得DBN的学习阶段更加高效。将所提算法应用到中文以及维吾尔语中，实验结果表明在酒店评论数据集上，FSDBN在准确率方面比DBN提高了1.6%，在训练时间上比DBN缩短一半。     

Sentiment analysis of tweets using a unified convolutional neural network‐long short‐term memory network model

Sentiment analysis focuses on identifying and classifying the sentiments expressed in text messages and reviews. Social networks like Twitter, Facebook, and Instagram generate heaps of data filled with sentiments, and the analysis of such data is very fruitful when trying to improve the quality of both products and services alike. Classic machine learning techniques have a limited capability to efficiently analyze such large amounts of data and produce precise results; they are thus supported by deep learning models to achieve higher accuracy. This study proposes a combination of convolutional neural network and long short‐term memory (CNN‐LSTM) deep network for performing sentiment analysis on Twitter datasets. The performance of the proposed model is analyzed with machine learning classifiers, including the support vector classifier, random forest (RF), stochastic gradient descent (SGD), logistic regression, a voting classifier (VC) of RF and SGD, and state‐of‐the‐art classifier models. Furthermore, two feature extraction methods (term frequency‐inverse document frequency and word2vec) are also investigated to determine their impact on prediction accuracy. Three datasets (US airline sentiments, women's e‐commerce clothing reviews, and hate speech) are utilized to evaluate the performance of the proposed model. Experiment results demonstrate that the CNN‐LSTM achieves higher accuracy than those of other classifiers.     

基于注意力机制的特征融合-双向门控循环单元多模态情感分析

针对视频多模态情感分析中,未考虑跨模态的交互作用以及各模态贡献程度对最后情感分类结果的影响的问题,提出一种基于注意力机制的特征融合-双向门控循环单元多模态情感分析模型（AMF-BiGRU）。首先,利用双向门控循环单元（BiGRU）来考虑各模态中话语间的相互依赖关系,并得到各模态的内部信息;其次,通过跨模态注意力交互网络层将模态内部信息与模态之间的交互作用相结合;然后,引入注意力机制来确定各模态的注意力权重,并将各模态特征进行有效融合;最后,通过全连接层和softmax层获取情感分类结果。在公开的CMU-MOSI和CMU-MOSEI数据集上进行实验。实验结果表明,与传统的多模态情感分析方法（如多注意力循环网络（MARN））相比,AMF-BiGRU模型在CMU-MOSI数据集上的准确率和F1值分别提升了6.01%和6.52%,在CMU-MOSEI数据集上的准确率和F1值分别提升了2.72%和2.30%。可见,AMF-BiGRU模型能够有效提高多模态的情感分类性能。     

Creating sentiment lexicon for sentiment analysis in Urdu: The case of a resource‐poor language

The sentiment analysis (SA) applications are becoming popular among the individuals and organizations for gathering and analysing user's sentiments about products, services, policies, and current affairs. Due to the availability of a wide range of English lexical resources, such as part‐of‐speech taggers, parsers, and polarity lexicons, development of sophisticated SA applications for the English language has attracted many researchers. Although there have been efforts for creating polarity lexicons in non‐English languages such as Urdu, they suffer from many deficiencies, such as lack of publically available sentiment lexicons with a proper scoring mechanism of opinion words and modifiers. In this work, we present a word‐level translation scheme for creating a first comprehensive Urdu polarity resource: “Urdu Lexicon” using a merger of existing resources: list of English opinion words, SentiWordNet, English–Urdu bilingual dictionary, and a collection of Urdu modifiers. We assign two polarity scores, positive and negative, to each Urdu opinion word. Moreover, modifiers are collected, classified, and tagged with proper polarity scores. We also perform an extrinsic evaluation in terms of subjectivity detection and sentiment classification, and the evaluation results show that the polarity scores assigned by this technique are more accurate than the baseline methods.     

基于情绪词的非监督中文情感分类方法研究

情感分类任务旨在识别文本所表达的情感色彩信息(例如,褒或者贬,支持或者反对)。该文提出一种基于情绪词的中文情感分类方法,使用大规模未标记数据和少量情绪词实现情感分类。具体来讲,首先使用情绪词从未标注数据中抽取高正确率的自动标注数据作为训练样本,然后采用半监督学习方法训练分类器进行情感分类。实验表明,该文提出的方法在产品评论与酒店评论两个领域的情感分类任务中取得了较好地分类效果。     

SentiTurkNet: a Turkish polarity lexicon for sentiment analysis

Sentiment analysis aims to extract the sentiment polarity of given segment of text. Polarity resources that indicate the sentiment polarity of words are commonly used in different approaches. While English is the richest language in regard to having such resources, the majority of other languages, including Turkish, lack polarity resources. In this work we present the first comprehensive Turkish polarity resource, SentiTurkNet, where three polarity scores are assigned to each synset in the Turkish WordNet, indicating its positivity, negativity, and objectivity (neutrality) levels. Our method is general and applicable to other languages. Evaluation results for Turkish show that the polarity scores obtained through this method are more accurate compared to those obtained through direct translation (mapping) from SentiWordNet.     

Multiaspect‐based opinion classification model for tourist reviews

Tourist reviews on social media websites reflect the tourist's opinions concerning various aspects of a tourist place or service (e.g., “comfortable room” and “terrible service” in hotel reviews). Extracting these aspects from reviews is a challenging task in opinion mining. Therefore, aspect‐based opinion mining has emerged as a new area of social review mining. Existing approaches in this area focus on extracting explicit aspects and classification of opinions around these aspects. However, the implicit and coreferential aspects during aspect extraction are often neglected, and the classification of multiaspect opinions is relatively less emphasized in prior art. In this paper, we propose a model, namely, “enhanced multiaspect‐based opinion classification” that addresses existing challenges by automatically extracting both explicit and implicit aspects and classifying the multiaspect opinions. In this model, first, a probabilistic co‐occurrence‐based method is proposed that utilizes the co‐occurrence between aspects and sentiment words to identify the coreferential aspects and merge them into groups. Second, an implicit aspect extraction method is proposed that associates the sentiment words with suitable aspects to build an aspect‐sentiment hierarchy. Third, a multiaspect opinion classification approach is proposed that employs multilabel classification algorithms to classify opinions into different polarity classes. The effectiveness of the proposed model is evaluated by conducting experiments on benchmark and real‐world datasets. The experimental results revealed the supremacy of multilabel classifiers by achieving 90% accuracy per label on classification when extracting 87% domain‐relevant aspects. A state‐of‐the‐art performance comparison is conducted that also verifies the advantages of the proposed model.     

基于时序和上下文特征的中文隐式情感分类模型

对社交网络上的海量文本信息进行情感分析可以更好地挖掘网民行为规律,从而帮助决策机构了解舆情倾向以及帮助商家改善服务质量。由于不存在关键情感特征、表达载体形式和文化习俗等因素的影响,中文隐式情感分类任务比其他语言更加困难。已有的中文隐式情感分类方法以卷积神经网络（CNN）为主,这些方法存在着无法获取词语的时序信息和在隐式情感判别中未合理利用上下文情感特征的缺陷。为了解决以上问题,采用门控卷积神经网络（GCNN）提取隐式情感句的局部重要信息,采用门控循环单元（GRU）网络增强特征的时序信息;而在隐式情感句的上下文特征处理上,采用双向门控循环单元（BiGRU）+注意力机制（Attention）的组合提取重要情感特征;在获得两种特征后,通过融合层将上下文重要特征融入到隐式情感判别中;最后得到的融合时序和上下文特征的中文隐式情感分类模型被命名为GGBA。在隐式情感分析评测数据集上进行实验,结果表明所提出的GGBA模型在宏平均准确率上比普通的文本CNN即TextCNN提高了3.72%、比GRU提高了2.57%、比中断循环神经网络（DRNN）提高了1.90%,由此可见, GGBA模型在隐式情感分析任务中比基础模型获得了更好的分类性能。     

Particle swarm optimization-based feature selection in sentiment classification

Sentiment classification is one of the important tasks in text mining, which is to classify documents according to their opinion or sentiment. Documents in sentiment classification can be represented in the form of feature vectors, which are employed by machine learning algorithms to perform classification. For the feature vectors, the feature selection process is necessary. In this paper, we will propose a feature selection method called fitness proportionate selection binary particle swarm optimization (F-BPSO). Binary particle swarm optimization (BPSO) is the binary version of particle swam optimization and can be applied to feature selection domain. F-BPSO is a modification of BPSO and can overcome the problems of traditional BPSO including unreasonable update formula of velocity and lack of evaluation on every single feature. Then, some detailed changes are made on the original F-BPSO including using fitness sum instead of average fitness in the fitness proportionate selection step. The modified method is, thus, called fitness sum proportionate selection binary particle swarm optimization (FS-BPSO). Moreover, further modifications are made on the FS-BPSO method to make it more suitable for sentiment classification-oriented feature selection domain. The modified method is named as SCO-FS-BPSO where SCO stands for “sentiment classification-oriented”. Experimental results show that in benchmark datasets original F-BPSO is superior to traditional BPSO in feature selection performance and FS-BPSO outperforms original F-BPSO. Besides, in sentiment classification domain, SCO-FS-BPSO which is modified specially for sentiment classification is superior to traditional feature selection methods on subjective consumer review datasets.     

Enriched Latent Dirichlet Allocation for Sentiment Analysis

One of the main benefits of unsupervised learning is that there is no need for labelled data. As a method of this category, latent Dirichlet allocation (LDA) estimates the semantic relations between the words of the text effectively and can play an important role in solving various issues, including emotional analysis in combination with other parameters. In this study, three novel topic models called date sentiment LDA (DSLDA), author–date sentiment LDA (ADSLDA), and pack–author–date sentiment LDA (PADSLDA) are proposed. The proposed models extend LDA through some extra parameters such as date, author, helpfulness, sentiment, and subtopic. The proposed models use helpfulness in the Gibbs sampling algorithm. Helpfulness is a part of readers who found the review helpful. The proposed models divide the words into two categories: the words more affected by the distribution of subtopic and the words more affected by the main topic. In this study, a new concept called pack is introduced, and a new model called PADSLDA is proposed for sentiment analysis at pack level. The proposed models outperformed the baseline models because according to evaluations results, the extra parameters can appropriately affect the generating process of words in a review. Sentiment analysis at the document level, perplexity, and topic coherence are the main parameters used in the evaluations.     

融合词性和注意力的卷积神经网络对象级情感分类方法

多个对象同时讨论时,对文本的情感分析结果与针对特定对象的情感倾向可能不一致,对象级情感分类任务需在文本整体语义的场景下,重点关注与给定对象相关的内容.文中提出融合词性和注意力的卷积神经网络对象级情感分类方法.引入词性信息,通过长短时记忆神经网络建模输入序列,构建对象注意力,将注意力融入到卷积神经网络结构中分析关于给定对象的情感倾向.词性信息有助于捕获与对象具有修饰关系的内容和弱化内容或距离相近但无搭配关系的句子成分的影响.结合长短时记忆神经网络和卷积神经网络结构建模文本,更有利于同时建模文本整体语义与对象相关语义.在SemEval2014数据集上的实验表明,文中方法取得优于基于长短时记忆神经网络的注意力机制方法的分类效果.     

Comparative experiments using supervised learning and machine translation for multilingual sentiment analysis

Sentiment analysis is the natural language processing task dealing with sentiment detection and classification from texts. In recent years, due to the growth in the quantity and fast spreading of user-generated contents online and the impact such information has on events, people and companies worldwide, this task has been approached in an important body of research in the field. Despite different methods having been proposed for distinct types of text, the research community has concentrated less on developing methods for languages other than English. In the above-mentioned context, the present work studies the possibility to employ machine translation systems and supervised methods to build models able to detect and classify sentiment in languages for which less/no resources are available for this task when compared to English, stressing upon the impact of translation quality on the sentiment classification performance. Our extensive evaluation scenarios show that machine translation systems are approaching a good level of maturity and that they can, in combination to appropriate machine learning algorithms and carefully chosen features, be used to build sentiment analysis systems that can obtain comparable performances to the one obtained for English.     

基于深度学习的短文本情感倾向分析综述

短文本情感倾向分析是自然语言处理领域的关键研究问题之一。情感倾向分析是用于检测语言所蕴含主观倾向语义的一系列方法、技术和工具,是对文本深层语义理解的关键。短文本数据的随意性、高歧义性以及简短性使得传统基于特征工程和机器学习分类技术的情感倾向分析任务性能有限。随着深度学习技术在自然语言处理中的广泛应用,基于深度学习的短文本情感倾向分析模型取得了新的突破。通过对相关文献的梳理,首先概述和对比了传统方法和深度学习方法,介绍和剖析了近年基于深度学习的短文本情感倾向分析模型,并阐述了模型的联系、区别与优势;其次归纳了深度学习在短文本情感倾向分析中的研究热点和进展思路,介绍了情感倾向分析常用的公开数据集以及评价指标;最后结合深度学习技术特点和任务难点,对深度学习在短文本情感倾向分析方向的应用前景进行预测。     

基于深度学习的社交网络平台细粒度情感分析

文本语言的情感分析历来是自然语言处理领域的热点研究课题,尤其是在当下互联网迈入web2.0时代,多样的社交网络平台呈现出巨量而丰富的文本情感信息,因此挖掘网络数据文本信息并作情感倾向判断对人机交互与人工智能具有重大的现实意义。传统的解决文本情感分析问题的方法主要是浅层学习算法,利用回归、分类等方案实现特征的提取及分类。以这类方法为起点,本文探索采用深度学习的方法对网络文本进行细粒度的情感分析,以期达到即时获取依附于网络世界的社会人的情感,甚至是让机器达到对人类情感表达的深度理解。对于深度学习的具体实现,本文采用的是降噪自编码器来对文本进行无标记特征学习并进行情感分类,后文中利用实验训练获得最佳的参数设置,并通过对实验结果的分析和评估论证深度学习对于情感信息的强大解析能力。     

基于极性转移和LSTM的树结构网络与句子分类

长短期记忆网络(long short term memory,LSTM)是一种能长久储存序列信息的循环神经网络,在语言模型、语音识别、机器翻译等领域都得到了广泛的应用。先研究了前人如何将LSTM中的记忆模块拓展到语法树得到LSTM树结构网络模型,以获取和储存句子深层次的语义结构信息;然后针对句子词语间的极性转移在LSTM树结构网络模型中添加了极性转移信息提出了极性转移LSTM树结构网络模型,更好获取情感信息来进行句子分类。实验表明在Stanford sentiment tree-bank数据集上,提出的极性转移LSTM树结构网络模型的句子分类效果优于LSTM、递归神经网络等模型。     

Sentiment Analysis based Multi-Person Multi-criteria Decision Making methodology using natural language processing and deep learning for smarter decision aid. Case study of restaurant choice using TripAdvisor reviews

Decision making models are constrained by taking the expert evaluations with pre-defined numerical or linguistic terms. We claim that the use of sentiment analysis will allow decision making models to consider expert evaluations in natural language. Accordingly, we propose the Sentiment Analysis based Multi-person Multi-criteria Decision Making (SA-MpMcDM) methodology for smarter decision aid, which builds the expert evaluations from their natural language reviews, and even from their numerical ratings if they are available. The SA-MpMcDM methodology incorporates an end-to-end multi-task deep learning model for aspect based sentiment analysis, named DOC-ABSADeepL model, able to identify the aspect categories mentioned in an expert review, and to distill their opinions and criteria. The individual evaluations are aggregated via the procedure named criteria weighting through the attention of the experts. We evaluate the methodology in a case study of restaurant choice using TripAdvisor reviews, hence we build, manually annotate, and release the TripR-2020 dataset of restaurant reviews. We analyze the SA-MpMcDM methodology in different scenarios using and not using natural language and numerical evaluations. The analysis shows that the combination of both sources of information results in a higher quality preference vector.