Search results

This study aims to construct a sentiment series generation method for danmu comments based on deep learning, and explore the features of sentiment series after clustering.

This study consisted of two main parts: danmu comment sentiment series generation and clustering. In the first part, the authors proposed a sentiment classification model based on BERT fine-tuning to quantify danmu comment sentiment polarity. To smooth the sentiment series, they used methods, such as comprehensive weights. In the second part, the shaped-based distance (SBD)-K-shape method was used to cluster the actual collected data.

The filtered sentiment series or curves of the microfilms on the Bilibili website could be divided into four major categories. There is an apparently stable time interval for the first three types of sentiment curves, while the fourth type of sentiment curve shows a clear trend of fluctuation in general. In addition, it was found that “disputed points” or “highlights” are likely to appear at the beginning and the climax of films, resulting in significant changes in the sentiment curves. The clustering results show a significant difference in user participation, with the second type prevailing over others.

Their sentiment classification model based on BERT fine-tuning outperformed the traditional sentiment lexicon method, which provides a reference for using deep learning as well as transfer learning for danmu comment sentiment analysis. The BERT fine-tuning–SBD-K-shape algorithm can weaken the effect of non-regular noise and temporal phase shift of danmu text.

Sarcasm is a linguistic expression that usually carries the opposite meaning of what is being said by words, thus making it difficult for machines to discover the actual meaning. It is mainly distinguished by the inflection with which it is spoken, with an undercurrent of irony, and is largely dependent on context, which makes it a difficult task for computational analysis. Moreover, sarcasm expresses negative sentiments using positive words, allowing it to easily confuse sentiment analysis models. This paper aims to demonstrate the task of sarcasm detection using the approach of machine and deep learning.

For the purpose of sarcasm detection, machine and deep learning models were used on a data set consisting of 1.3 million social media comments, including both sarcastic and non-sarcastic comments. The data set was pre-processed using natural language processing methods, and additional features were extracted and analysed. Several machine learning models, including logistic regression, ridge regression, linear support vector and support vector machines, along with two deep learning models based on bidirectional long short-term memory and one bidirectional encoder representations from transformers (BERT)-based model, were implemented, evaluated and compared.

The performance of machine and deep learning models was compared in the task of sarcasm detection, and possible ways of improvement were discussed. Deep learning models showed more promise, performance-wise, for this type of task. Specifically, a state-of-the-art model in natural language processing, namely, BERT-based model, outperformed other machine and deep learning models.

This study compared the performance of the various machine and deep learning models in the task of sarcasm detection using the data set of 1.3 million comments from social media.

As global issues such as climate change, economic growth, social equality and the wealth gap are widely discussed, education for sustainable development (ESD) allows every human being to acquire the knowledge, skills, attitudes and values necessary to shape a sustainable future. It also requires participatory teaching and learning methods that motivate and empower learners to change their behavior and take action for sustainable development. Teachers have begun rating pupils based on peer assessment for open evaluation. Peer assessment enables students to transition from passive to active feedback recipients. The assessors improve critical thinking and encourage introspection, resulting in more significant recommendations. However, the quality of peer assessment is variable, resulting in reviewers not recognizing the remarks of other reviewers, therefore the benefits of peer assessment cannot be fulfilled. In the past, researchers frequently employed post-event questionnaires to examine the effects of peer assessment on learning effectiveness, which did not accurately reflect the quality of peer assessment in real time.

This study employs a multi-label model and develops a self-feedback system in order to use the AIOLPA system in the classroom to enhance students' learning efficacy and the validity of peer assessment.

The research findings indicate that the better peer assessment through the rapid feedback system, for the evaluator, encourages more self-reflection and attempts to provide more ideas, so bringing the peer rating closer to the instructor rating and assisting the evaluator. Improve self-evaluation and critical thinking for the evaluator, peers make suggestions and comments to help improve the work and support the growth of students' learning effectiveness, which can lead to more suggestions and an increase in the work’s quality.

ESD consequently promotes competencies like critical thinking, imagining future scenarios and making decisions in a collaborative way. This study builds an online peer assessment system with a self-feedback mechanism capable of classifying peer comments, comparing them with scores in a consistent manner and providing prompt feedback to critics.

Research on artificial intelligence (AI) and its potential effects on the workplace is increasing. How AI and the futures of work are framed in traditional media has been examined in prior studies, but current research has not gone far enough in examining how AI is framed on social media. This paper aims to fill this gap by examining how people frame the futures of work and intelligent machines when they post on social media.

We investigate public interpretations, assumptions and expectations, referring to framing expressed in social media conversations. We also coded the emotions and attitudes expressed in the text data. A corpus consisting of 998 unique Reddit post titles and their corresponding 16,611 comments was analyzed using computer-aided textual analysis comprising a BERTopic model and two BERT text classification models, one for emotion and the other for sentiment analysis, supported by human judgment.

Different interpretations, assumptions and expectations were found in the conversations. Three subframes were analyzed in detail under the overarching frame of the New World of Work: (1) general impacts of intelligent machines on society, (2) undertaking of tasks (augmentation and substitution) and (3) loss of jobs. The general attitude observed in conversations was slightly positive, and the most common emotion category was curiosity.

Findings from this research can uncover public needs and expectations regarding the future of work with intelligent machines. The findings may also help shape research directions about futures of work. Furthermore, firms, organizations or industries may employ framing methods to analyze customers’ or workers’ responses or even influence the responses. Another contribution of this work is the application of framing theory to interpreting how people conceptualize the future of work with intelligent machines.

This study aims to use big data analysis and sheds light on key hotel features that play a role in the revisit intention of customers. In addition, this study endeavors to highlight hotel features for different customer segments.

This study uses a machine learning method and analyzes around 100,000 reviews of customers of 100 selected hotels around the world where they had indicated on Trip Advisor their intention to return to a particular hotel. The important features of the hotels are then extracted in terms of the 7Ps of the marketing mix. This study has then segmented customers intending to revisit hotels, based on the similarities in their reviews.

In total, 71 important hotel features are extracted using text analysis of comments. The most important features are the room, staff, food and accessibility. Also, customers are segmented into 15 groups, and key hotel features important for each segment are highlighted.

In this research, the number of repetitions of words was used to identify key hotel features, whereas sentence-based analysis or group analysis of adjacent words can be used.

This study highlights key hotel features that are crucial for customers’ revisit intention and identifies related market segments that can support managers in better designing their strategies and allocating their resources.

By using text mining analysis, this study identifies and classifies important hotel features that are crucial for the revisit intention of customers based on the 7Ps. Methodologically, the authors suggest a comprehensive method to describe the revisit intention of hotel customers based on customer reviews.

Drawing on Suchman’s conception of cognitive legitimacy and Boswell’s account of the political functions of expert knowledge, this paper aims to study the due process followed by the International Integrated Reporting Council (IIRC) prior to the publication of the first version of the International Integrated Reporting Framework (IIRF). Specifically, the author analyses the lobbying strategies used in the comment letters sent by a subset of lobbyists, “the experts”, represented by accounting bodies and firms, regulators and academics.

From both a form- and meaning-oriented analysis, this paper focuses on how the experts resorted to the functions of knowledge when they took part in the IIRF’s public consultation. The author first carries out a quantitative content analysis of the responses to the 2013 Consultation Draft submitted by those constituents considered as accounting expert lobbyists. Then, the author analyse how these actors framed their comments under expert knowledge to legitimise the IIRC, the IIRF and the accounting profession itself.

The findings suggest that the expert groups welcomed the opportunity, not simply to legitimise the IIRC through their democratic support, but to provide a technocratic settlement that ensures the due process is based on the mobilisation of expert knowledge as a legitimate source. By drawing on the cognitive legitimacy of expert lobbyists, the IIRC drew on the political functions of expert knowledge to reduce uncertainty and gain stability.

Analysis of the lobbying strategies used by the accounting experts whose position could make a difference and receive more attention from the IIRC makes this contribution of particular interest, especially since the first version of the IIRF sought to guide disclosure and sustainable business practices around the world.

Experts as political actors play a legitimising role since they are capable of producing relevant knowledge that, due to its nature and scope, certainly affects policymaking and sustainable development.

This research provides a sociopolitical perspective to comprehend how some lobbying strategies, in this case, of expert actors, contribute to legitimising a standard-setter body and its endeavours in the context of voluntary standards.

This paper aims to examine the types of user misbehaviours in the sharing economy (SE) context. SE offers a fruitful study setting due to the scope of potential misbehaviour and the expanded role of consumers.

The study drew on online archival data from the AirbnbHell.com website, where people share their stories about their Airbnb-related negative experiences. The authors reviewed 405 hosts’, guests’ and neighbours’ stories and coded the identified forms of misbehaviours into categories. The typology thus developed was validated in the context of the Uber Rides service.

User misbehaviours in the SE context can be distinguished based on the domain in which the user role is violated and the nature of violated norms. These two conceptual distinctions delineate a four-fold typology of user misbehaviours: illegal, unprofessional, unbefitting and uncivil behaviours.

The trustworthiness of the stories could not be assessed.

The presented typology can be used as a mapping tool that facilitates detection of the full scope of misbehaviours and as a managerial tool that provides ideas for effective management of misbehaviours that correspond to each category.

The paper presents the first empirically derived comprehensive typology of user misbehaviours in SE settings. This typology enables classification of a broad set of misbehaviours, including previously overlooked unprofessional behaviours carried out by peer-service providers. The study also puts forward a revised definition of consumer misbehaviours that encompasses the impact of misbehaviours on parties not directly involved in the SE-mediated exchange.

Contextual cues have become a hot research topic in the field of mobile consumer behavior, owing to the continuous rise of digital marketing. However, the complex online shopping scene makes it challenging to directly identify the association between the characteristics of contextual cues and consumer behavior. Presently, few studies have only systematically extracted and refined the types and characteristics of contextual cues. The purpose of this study is to explore the types and mechanisms of contextual cues in online shopping scenarios.

This study uses the word2vec algorithm, grounded theory and co-occurrence cluster method, along with online shopping word-of-mouth (WOM) text and consumer behavior theory, in order to explore different types of contextual cues and its efficiency from 5,619 comment corpus.

This study puts forward the following conclusions. (1) From the perspective of online shopping, contextual cues comprise aesthetic perception cues, value perception cues, trust-dependent cues, time perception cues, memory attention cues, spatial perception cues, attribute cues and relationship cues. (2) Based on the online shopping scenarios, contextual cues and their interaction effects exert an effect on consumer satisfaction, recommendation, purchase and return behavior.

The study conclusions are helpful to further reveal the deep association between contextual cues and consumer behavior in the process of online shopping, thus providing practical and theoretical enlightenment for enterprises to not only effectively reshape the scene but also promote the consumers' active purchase behavior.

This paper aims to bridge the gap by understanding the context of ageing consumer behaviour in the online community. Specifically, this research seeks to identify which content typologies are critical to generating high engagement levels and, consequently, online brand advocacy and to understand the underlying motivation behind consumer online engagement.

A netnographic approach was used to comprehensively analyse older consumers’ online communities on Facebook, namely, “Silversurfers”. A total of 3,991 posts were included in the study and analysed using a content analysis approach over two years, from 2020 to 2022.

Results revealed that photography is the most active media type among older consumers. This study extends the literature on content marketing, identifying 17 new content types that reflect the four motivation states of older consumers to engage with the online community: cognitive/informative oriented, affective/emotional oriented, co-creation/interactive oriented and nostalgic oriented. Moreover, this investigation stressed affective/emotional oriented and nostalgic oriented as the primary motivations for higher engagement levels.

The older population is growing, which makes the ageing market potentially huge. However, more literature needs to address it, especially in online communities. Finally, to the best of the authors’ knowledge, this study develops an original content typology framework in which firms can consider implementing effective content typology strategies for the older consumer segment.