Search results

The purpose of this paper is to propose an attention alignment method for opinion mining of massive open online course (MOOC) comments. Opinion mining is essential for MOOC applications. In this study, the authors analyze some of bidirectional encoder representations from transformers (BERT’s) attention heads and explore how to use these attention heads to extract opinions from MOOC comments.

The approach proposed is based on an attention alignment mechanism with the following three stages: first, extracting original opinions from MOOC comments with dependency parsing. Second, constructing frequent sets and using the frequent sets to prune the opinions. Third, pruning the opinions and discovering new opinions with the attention alignment mechanism.

The experiments on the MOOC comments data sets suggest that the opinion mining approach based on an attention alignment mechanism can obtain a better F1 score. Moreover, the attention alignment mechanism can discover some of the opinions filtered incorrectly by the frequent sets, which means the attention alignment mechanism can overcome the shortcomings of dependency analysis and frequent sets.

To take full advantage of pretrained language models, the authors propose an attention alignment method for opinion mining and combine this method with dependency analysis and frequent sets to improve the effectiveness. Furthermore, the authors conduct extensive experiments on different combinations of methods. The results show that the attention alignment method can effectively overcome the shortcomings of dependency analysis and frequent sets.

The paper's aim is to mine and analyze opinion formation on the basis of consumer dialogs in online forums.

The study identifies opinions, communication relationships, and dialog acts of forum users using different text mining methods. Utilizing this data, social networks can be derived and analyzed to detect influential users and opinion tendencies. The approach is applied to sample online forums discussing the iPhone.

Combining text mining and social network analysis enables the study of opinion formation and yields encouraging results. Out of the four methods employed for text mining, support vector machines performed best.

The data set applied here is fairly small. More threads on different products will be considered in future work to improve validation.

The approach represents a valuable instrument for online market research. It enables companies to recognize opportunities and risks and to initiate appropriate marketing actions.

This work is one of the first studies that combine communication content, relationships and dialog acts for analyzing opinion formation.

The suddenness, urgency and social publicity of emergency events lead to great impacts on public life. The deep analysis of emergency events can provide detailed and comprehensive information for the public to get trends of events timely. With the development of social media, users prefer to express opinions on emergency events online. Thus, massive public opinion information of emergencies has been generated. Hence, this paper aims to conduct multidimensional mining on emergency events based on user-generated contents, so as to obtain finer-grained results.

This paper conducted public opinion analysis via fine-grained mining. Specifically, public opinion about an emergency event was collected as experimental data. Secondly, opinion mining was conducted to get users’ opinion polarities. Meanwhile, users’ information was analysed to identify impacts of users’ characteristics on public opinion.

The experimental results indicate that public opinion is mainly negative in emergencies. Meanwhile, users in developed regions are more active in expressing opinions. In addition, male users, especially male users with high influence, are more rational in public opinion expression.

To the best of the authors’ knowledge, this is the first research to identify public opinion in emergency events from multiple dimensions, which can get in-detail differences of users’ online expression.

The purpose of this paper is to mine competitive intelligence in social media to find the market insight by comparing consumer opinions and sales performance of a business and one of its competitors by analyzing the public social media data.

An exploratory test using a multiple case study approach was used to compare two competing smartphone manufacturers. Opinion mining and sentiment analysis are conducted first, followed by further validation of results using statistical analysis. A total of 229,948 tweets mentioning the iPhone6 or the GalaxyS5 have been collected for four months following the release of the iPhone6; these have been analyzed using natural language processing, lexicon-based sentiment analysis, and purchase intention classification.

The analysis showed that social media data contain competitive intelligence. The volume of tweets revealed a significant gap between the market leader and one follower; the purchase intention data also reflected this gap, but to a less pronounced extent. In addition, the authors assessed whether social opinion could explain the sales performance gap between the competitors, and found that the social opinion gap was similar to the shipment gap.

This study compared the social media opinion and the shipment gap between two rival smart phones. A business can take the consumers’ opinions toward not only its own product but also toward the product of competitors through social media analytics. Furthermore, the business can predict market sales performance and estimate the gap with competing products. As a result, decision makers can adjust the market strategy rapidly and compensate the weakness contrasting with the rivals as well.

This paper’s main contribution is to demonstrat the competitive intelligence via the consumer opinion mining of social media data. Researchers, business analysts, and practitioners can adopt this method of social media analysis to achieve their objectives and to implement practical procedures for data collection, spam elimination, machine learning classification, sentiment analysis, feature categorization, and result visualization.

Timely detection of emergency events and effective tracking of corresponding public opinions are critical in emergency management. As media are immediate sources of information on emergencies, the purpose of this paper is to propose cross-media analytics to detect and track emergency events and provide decision support for government and emergency management departments.

In this paper, a novel emergency event detection and opinion mining method is proposed for emergency management using cross-media analytics. In the proposed approach, an event detection module is constructed to discover emergency events based on cross-media analytics, and after the detected event is confirmed as an emergency event, an opinion mining module is used to analyze public sentiments and then generate public sentiment time series for early warning via a semantic expansion technique.

Empirical results indicate that a specific emergency can be detected and that public opinion can be tracked effectively and efficiently using cross-media analytics. In addition, the proposed system can be used for decision support and real-time response for government and emergency management departments.

This paper takes full advantage of cross-media information and proposes novel emergency event detection and opinion mining methods for emergency management using cross-media analytics. The empirical analysis results illustrate the efficiency of the proposed method.

The proposed method can be applied for detection of emergency events and tracking of public opinions for emergency decision support and governmental real-time response.

This research work contributes to the design of a decision support system for emergency event detection and opinion mining. In the proposed approaches, emergency events are detected by leveraging cross-media analytics, and public sentiments are measured using an auto-expansion of the domain dictionary in the field of emergency management to eliminate the misclassification of the general dictionary and to make the quantization more accurate.

Comparative opinions widely exist in online reviews as a common way of expressing consumers’ ideas or preferences toward certain products. Such opinion-rich texts are key proxies for detecting product competitiveness. The purpose of this paper is to set up a model for competitiveness analysis by identifying comparative relations from online reviews for restaurants based on both pattern matching and machine learning.

The authors define the sub-category of comparative sentences according to Chinese linguistics. Classification rules are set up for each type of comparative relations through class sequence rule. To improve the accuracy of classification, a comparative entity dictionary is then introduced for further identifying comparative sentences. Finally, the authors collect reviews for restaurants from Dianping.com to conduct experiments for testing the proposed model.

The experiments show that the proposed method outperforms the baseline methods in terms of precision in identifying comparative sentences. On the basis of such comparison-rich sentences, product features and comparative relations are extracted for sentiment analysis, and sentimental score is assigned to each comparative relation to facilitate competitiveness analysis.

Only the explicit comparative relations are discussed, neglecting the implicit ones. Besides that, the study is grounded in the assumption that all features are homogeneous. In some cases, however, the weights to different aspects are not of the same importance to market.

On the basis of comparative relation mining, product features and comparative opinions are extracted for competitiveness analysis, which is of interest to businesses for finding weakness or strength of products, as well as to consumers for making better purchase decisions.

Comparative relation mining could be possibly applied in social media for identifying relations among users or products, and ranking users or products, as well as helping companies target and track competitors to enhance competitiveness.

The authors propose a research framework for restaurant competitiveness analysis by mining comparative relations from online consumer reviews. The results would be able to differentiate one restaurant from another in some aspects of interest to consumers, and reveal the changes in these differences over time.

The prediction of pre-election polls is an issue of concern for both politicians and voters. The Taiwan nine-in-one election held in 2014 ended with jaw-dropping results; apparently, traditional polls did not work well. As a remedy to this problem, the purpose of this paper is to utilize the comments posted on social media to analyze civilians’ views on the two candidates for mayor of Taichung City, Chih-chiang Hu, and Chia-Lung Lin.

After conducting word segmentation and part-of-speech tagging for the collected reviews, this study constructs the opinion phrase extraction rules for identifying the opinion words associated with the attribute words. Next, this study classifies the attribute words into six municipal governance-related topics and calculates the opinion scores for each candidate. Finally, this study uses correspondence analysis to transform opinion information on the candidates into a graphical display to facilitate the interpretation of voters’ views.

The results show that the topics of candidates’ backgrounds and transport infrastructure were the two most critical factors for the election prediction. Based on the predication, Lin outscores Hu by 17.74 percent which is close to the real election results.

This study proposes new rules for the extraction of Chinese opinion words associated with attribute words.

This study applies Chinese semantic analysis to assist in predicting election results and investigating the topics of concern to voters.

The proposed opinion phrase extraction rules for Chinese social media, as well as the election forecast process, can provide valuable references for political parties and candidates to plan better nomination and election strategies.

The purpose of this paper is to develop an effective approach to support and guide production improvement processes utilising online product reviews.

This paper combines two methods: (1) natural language processing (NLP) to support advanced text mining to increase the accuracy of information extracted from product reviews and (2) quality function deployment (QFD) to utilise the extracted information to guide the product improvement process.

The paper proposes an approach to automate the process of obtaining voice of the customer (VOC) by performing text mining on available online product reviews while considering key factors such as the time of review and review usefulness. The paper enhances quality management processes in organisations and advances the literature on customer-oriented product improvement processes.

Online product reviews are a valuable source of information for companies to capture the true VOC. VOC is then commonly used by companies as the main input for QFD to enhance quality management and product improvement. However, this process requires considerable time, during which VOC may change, which may negatively impact the output of QFD. This paper addresses this challenge by providing an improved approach.

Opinion mining (OM), also known as “sentiment classification”, which aims to discover common patterns of user opinions from their textual statements automatically or semi-automatically, is not only useful for customers, but also for manufacturers. However, because of the complexity of natural language, there are still some problems, such as domain dependence of sentiment words, extraction of implicit features and others. The purpose of this paper is to propose an OM method based on topic maps to solve these problems.

Domain-specific knowledge is key to solve problems in feature-based OM. On the one hand, topic maps, as an ontology framework, are composed of topics, associations, occurrences and scopes, and can represent a class of knowledge representation schemes. On the other hand, compared with ontology, topic maps have many advantages. Thus, it is better to integrate domain-specific knowledge into OM based on topic maps. This method can make full use of the semantic relationships among feature words and sentiment words.

In feature-level OM, most of the existing research associate product features and opinions by their explicit co-occurrence, or use syntax parsing to judge the modification relationship between opinion words and product features within a review unit. They are mostly based on the structure of language units without considering domain knowledge. Only few methods based on ontology incorporate domain knowledge into feature-based OM, but they only use the “is-a” relation between concepts. Therefore, this paper proposes feature-based OM using topic maps. The experimental results revealed that this method can improve the accuracy of the OM. The findings of this study not only advance the state of OM research but also shed light on future research directions.

To demonstrate the “feature-based OM using topic maps” applications, this work implements a prototype that helps users to find their new washing machines.

This paper presents a new method of feature-based OM using topic maps, which can integrate domain-specific knowledge into feature-based OM effectively. This method can improve the accuracy of the OM greatly. The proposed method can be applied across various application domains, such as e-commerce and e-government.