Search results

With the rapid development of short videos in China, the public has become accustomed to using short videos to express their opinions. This paper aims to solve problems such as how to represent the features of different modalities and achieve effective cross-modal feature fusion when analyzing the multi-modal sentiment of Chinese short videos (CSVs).

This paper aims to propose a sentiment analysis model MSCNN-CPL-CAFF using multi-scale convolutional neural network and cross attention fusion mechanism to analyze the CSVs. The audio-visual and textual data of CSVs themed on “COVID-19, catering industry” are collected from CSV platform Douyin first, and then a comparative analysis is conducted with advanced baseline models.

The sample number of the weak negative and neutral sentiment is the largest, and the sample number of the positive and weak positive sentiment is relatively small, accounting for only about 11% of the total samples. The MSCNN-CPL-CAFF model has achieved the Acc-2, Acc-3 and F1 score of 85.01%, 74.16 and 84.84%, respectively, which outperforms the highest value of baseline methods in accuracy and achieves competitive computation speed.

This research offers some implications regarding the impact of COVID-19 on catering industry in China by focusing on multi-modal sentiment of CSVs. The methodology can be utilized to analyze the opinions of the general public on social media platform and to categorize them accordingly.

This paper presents a novel deep-learning multimodal sentiment analysis model, which provides a new perspective for public opinion research on the short video platform.

Despite the extensive academic interest in social media sentiment for financial fields, multimodal data in the stock market has been neglected. The purpose of this paper is to explore the influence of multimodal social media data on stock performance, and investigate the underlying mechanism of two forms of social media data, i.e. text and pictures.

This research employs panel vector autoregressive models to quantify the effect of the sentiment derived from two modalities in social media, i.e. text information and picture information. Through the models, the authors examine the short-term and long-term associations between social media sentiment and stock performance, measured by three metrics. Specifically, the authors design an enhanced sentiment analysis method, integrating random walk and word embeddings through Global Vectors for Word Representation (GloVe), to construct a domain-specific lexicon and apply it to textual sentiment analysis. Secondly, the authors exploit a deep learning framework based on convolutional neural networks to analyze the sentiment in picture data.

The empirical results derived from vector autoregressive models reveal that both measures of the sentiment extracted from textual information and pictorial information in social media are significant leading indicators of stock performance. Moreover, pictorial information and textual information have similar relationships with stock performance.

To the best of the authors’ knowledge, this is the first study that incorporates multimodal social media data for sentiment analysis, which is valuable in understanding pictures of social media data. The study offers significant implications for researchers and practitioners. This research informs researchers on the attention of multimodal social media data. The study’s findings provide some managerial recommendations, e.g. watching not only words but also pictures in social media.

This study aims to detect consumer sarcasm through inconsistencies in sentiment features between text and images of hotel reviews.

This paper proposes a model for sarcasm detection based on multimodal deep learning using reviews of three hotel brands collected from two travel platforms, which can identify emotional inconsistencies within a modality and across modalities. Text-image interaction information is explored using graph neural networks (GNN) to detect essential clues in sarcasm sentiment.

The research results show that the multimodal deep learning model outperforms other baseline models, which can help to understand hotel service evaluation and provide hotel managers with decision-making opinions.

This research can help hoteliers in two ways: detecting service quality and formulating strategies. By selecting reference hotel brands, hoteliers can better assess their level of service quality (optimal resource allocation ensues); therefore, sarcasm detection research is not only beneficial for hotel managers seeking to improve service quality. The multimodal deep learning method introduced in the present study can be replicated in other industries to help travel platforms optimize their products and services.

本研究通过分析酒店评论文本和图像之间情感特征的不一致性来检测消费者的讽刺。

本文提出了一种基于多模态深度学习的讽刺检测模型, 使用从两个旅行平台收集的三个酒店品牌的评论, 该模型能够识别模态内部和模态之间的情感不一致性。利用图神经网络（GNN）探索文本-图像交互信息, 以检测讽刺情感中的关键线索。

研究结果显示, 多模态深度学习模型优于其他基线模型, 这有助于理解酒店服务评估, 并为酒店经理提供决策建议。

该研究可以在两方面帮助酒店业者：检测服务质量和制定策略。通过选择参考酒店品牌, 酒店业者可以更好地评估其服务质量水平（随之而来的是最佳资源分配）, 因此, 讽刺检测研究不仅有助于寻求提高服务质量的酒店经理。本研究介绍的多模态深度学习方法可以在其他行业复制, 帮助旅行平台优化其产品和服务。

Although numerous signal modalities are available for emotion recognition, audio and visual modalities are the most common and predominant forms for human beings to express their emotional states in daily communication. Therefore, how to achieve automatic and accurate audiovisual emotion recognition is significantly important for developing engaging and empathetic human–computer interaction environment. However, two major challenges exist in the field of audiovisual emotion recognition: (1) how to effectively capture representations of each single modality and eliminate redundant features and (2) how to efficiently integrate information from these two modalities to generate discriminative representations.

A novel key-frame extraction-based attention fusion network (KE-AFN) is proposed for audiovisual emotion recognition. KE-AFN attempts to integrate key-frame extraction with multimodal interaction and fusion to enhance audiovisual representations and reduce redundant computation, filling the research gaps of existing approaches. Specifically, the local maximum–based content analysis is designed to extract key-frames from videos for the purpose of eliminating data redundancy. Two modules, including “Multi-head Attention-based Intra-modality Interaction Module” and “Multi-head Attention-based Cross-modality Interaction Module”, are proposed to mine and capture intra- and cross-modality interactions for further reducing data redundancy and producing more powerful multimodal representations.

Extensive experiments on two benchmark datasets (i.e. RAVDESS and CMU-MOSEI) demonstrate the effectiveness and rationality of KE-AFN. Specifically, (1) KE-AFN is superior to state-of-the-art baselines for audiovisual emotion recognition. (2) Exploring the supplementary and complementary information of different modalities can provide more emotional clues for better emotion recognition. (3) The proposed key-frame extraction strategy can enhance the performance by more than 2.79 per cent on accuracy. (4) Both exploring intra- and cross-modality interactions and employing attention-based audiovisual fusion can lead to better prediction performance.

The proposed KE-AFN can support the development of engaging and empathetic human–computer interaction environment.

Sentiment analysis is a text analysis method that is developed for systematically detecting, identifying, or extracting the emotional intent of words to infer if the text expresses a positive or negative tone. Although this novel method has opened an exciting new avenue for organizational research – mainly due to the abundantly available text data in organizations and the well-developed sentiment analysis techniques, it has also posed a serious challenge to many organizational researchers. This chapter aims to introduce the sentiment analysis method in the text mining area to the organizational research community. In this chapter, the authors first briefly discuss the central role of sentiment in organizational research and then introduce the traditional and modern approaches to sentiment analysis. The authors further delineate research paradigms for text analysis research, advocating the iterative research paradigm (cf., inductive and deductive research paradigms) that is more suitable for text mining research, and also introduce the analytical procedures for sentiment analysis with three stages – discovery, measurement, and inference. More importantly, the authors highlight both the dictionary-based and machine learning (ML) approaches in the measurement stage, with special coverage on deep learning and word embedding techniques as the latest breakthroughs in sentiment and text analyses. Lastly, the authors provide two illustrative examples to demonstrate the applications of sentiment analysis in organizational research. It is the authors’ hope that this chapter – by providing these practical guidelines – will help facilitate more applications of this novel method in organizational research in the future.

Social Media is one of the largest platforms to voluntarily communicate thoughts. With increase in multimedia data on social networking websites, information about human behaviour is increasing. This user-generated data are present on the internet in different modalities including text, images, audio, video, gesture, etc. The purpose of this paper is to consider multiple variables for event detection and analysis including weather data, temporal data, geo-location data, traffic data, weekday’s data, etc.

In this paper, evolution of different approaches have been studied and explored for multivariate event analysis of uncertain social media data.

Based on burst of outbreak information from social media including natural disasters, contagious disease spread, etc. can be controlled. This can be path breaking input for instant emergency management resources. This has received much attention from academic researchers and practitioners to study the latent patterns for event detection from social media signals.

This paper provides useful insights into existing methodologies and recommendations for future attempts in this area of research. An overview of architecture of event analysis and statistical approaches are used to determine the events in social media which need attention.

The aim of the research is to evaluate different modality for product reviews presentation and its impact on users’ performance, purchase intention and enjoyment.

The study utilized an experimental approach with 48 opportunistic participants in three groups (16 users per group). Participants were randomly assigned to experimental conditions to ensure unbiased treatment. Data were collected through controlled interventions or manipulations, with pre-defined measures to assess specific outcomes. Statistical techniques such as ANOVA were employed to analyse the data, allowing for comparisons between experimental variables.

The findings revealed that integrating facial expression avatars and emojis into an e-commerce platform effectively communicates product reviews and ratings. Moreover, the use of animation significantly enhanced user enjoyment. This suggests that visual representations not only convey information effectively but also contribute to a more engaging and enjoyable user experience.

While this experiment offers valuable insights into the impact of different e-commerce presentation layouts on user behaviour, further research could delve deeper into specific aspects such as the influence of individual user characteristics and the long-term effects of layout preferences.

This study contributes original insights by demonstrating the efficacy of facial expressive avatars and emojis in conveying product reviews and ratings within e-commerce platforms. Moreover, it adds value by highlighting the positive impact of animation on user enjoyment. By combining these elements, the research offers a novel approach to enhancing user engagement and understanding of customer feedback in online shopping environments. The findings provide valuable guidance for e-commerce platforms seeking innovative ways to communicate product information effectively and enhance the overall user experience, ultimately benefiting both businesses and consumers.

Most studies on Sentiment Analysis are performed in English. However, as the third most spoken language on the Internet, Sentiment Analysis for Spanish presents its challenges from a semantic and syntactic point of view. This review presents a scope of the recent advances in this area.

A systematic literature review on Sentiment Analysis for the Spanish language was conducted on recognized databases by the research community.

Results show classification systems through three different approaches: Lexicon based, Machine Learning based and hybrid approaches. Additionally, different linguistic resources as Lexicon or corpus explicitly developed for the Spanish language were found.

This study provides academics and professionals, a review of advances in Sentiment Analysis for the Spanish language. Most reviews on Sentiment Analysis are for English, and other languages such as Chinese or Arabic, but no updated reviews were found for Spanish.

After the ex-National Security Agency contractor Edward Snowden¹ disclosures of the National Security Agency surveillance of Americans’ online and phone communications, the Pew Research Center² administrated a panel survey to collect data concerning Americans’ opinions about privacy and security. This survey has mixed types of qualitative questions (closed and open-ended). In this context, to our knowledge, until today, no research has been applied on the open-ended part of these data. In this chapter, first the authors present their findings from applying sentiment analysis and topic extraction methods; second, the authors demonstrate their analysis to sentiments polarities; and finally, the authors interpret the semantic relationships between topics and their associated negativity, positivity, and neutral sentiments.

Market evaluation of products is the basis for product innovation, yet traditional expert-based evaluation methods are highly dependent on the specialization of experts. There exist a lot of weak expert-generated texts on the Internet of their own subjective evaluations of products. Analyzing these texts can indirectly extract the opinions of weak experts and transform them into decision-support information that assists product designers in understanding the market.

In social networks, a subset of users, termed “weak experts”, possess specialized knowledge and frequently share their product experiences online. This study introduces a comparative opinion mining framework that leverages the insights of “weak experts” to analyze user opinions.

An automotive product case study demonstrates that evaluations based on weak expert insights offer managerial insights with a 99.4% improvement in timeliness over traditional expert analyses. Furthermore, in the few-shot sentiment analysis module, with only 10% of the sample, the precision loss is just 1.59%. In addition, the quantitative module of specialization weighting balances low-specialization expert opinions and boosts the weight of high-specialization weak expert views. This new framework offers a valuable tool for companies in product innovation and market strategy development.

This study introduces a novel approach to opinion mining by focusing on the underutilized insights of weak experts. It combines few-shot sentiment analysis with specialization weighting and AHP, offering a comprehensive and efficient tool for product evaluation and market analysis.