User-generated Content Data Research Articles

A critical discussion on applying user-generated content to interdisciplinary research in tourism and health science

ABSTRACT The increasing popularity of user-generated content (UGC) as a type of secondary data is evident in many social science disciplines, including tourism. Although tourism research can enjoy numerous advantages from adopting UGC as a data source, relevant limitations must be acknowledged – especially for interdisciplinary studies combining tourism with other domains such as medical science. For example, some methodological caveats should be kept in mind when seeking to understand specific populations’ behaviours, experiences, and the therapeutic efficacy of tourism engagement (e.g. for vulnerable populations with health conditions). UGC data have been critiqued in the hard sciences for not adhering to the fundamental principles of reproducibility, reliability, and relevance; these data are thus typically seen as supplementary rather than a principal knowledge source. This article is meant to open a constructive academic debate on the use of UGC in interdisciplinary research between tourism and health science to ensure study rigour via suitable sampling strategies and data. A framework is also proposed to guide UGC adoption in studies between these disciplines. Research involving tourists with intellectual disabilities is described to illustrate the importance of leveraging UGC effectively.

Satisfaction and sustainability concerns in whale-watching tourism: A user-generated content model

This study examines the satisfaction and sustainability concerns of whale-watching tourists by analysing user-generated content (UGC) on social media. A satisfaction model was developed and estimated utilising an ordered probit analysis with UGC data from TripAdvisor over the last 13 years that includes a specific whale-watching lexicon. The model addresses most of the physical, human, environmental, experiential, and operational aspects of the activity, including consumers’ feelings and sustainability concerns. The significance of the variables in the model was proven with the available empirical data. The findings provide a comprehensive description of the underpinnings of whale watchers’ preferences and concerns. The evidence reveals compelling social trends towards higher sustainability concerns influencing satisfaction, providing valuable information for the industry and its decision-makers for understanding preferences for sustainability in whale-watching tourism.

Deconfounded Cross-modal Matching for Content-based Micro-video Background Music Recommendation

Object-oriented micro-video background music recommendation is a complicated task where the matching degree between videos and background music is a major issue. However, music selections in user-generated content (UGC) are prone to selection bias caused by historical preferences of uploaders. Since historical preferences are not fully reliable and may reflect obsolete behaviors, over-reliance on them should be avoided as knowledge and interests dynamically evolve. In this article, we propose a Deconfounded Cross-Modal matching model to mitigate such bias. Specifically, uploaders’ personal preferences of music genres are identified as confounders that spuriously correlate music embeddings and background music selections, causing the learned system to over-recommend music from majority groups. To resolve such confounders, backdoor adjustment is utilized to deconfound the spurious correlation between music embeddings and prediction scores. We further utilize Monte Carlo estimator with batch-level average as the approximations to avoid integrating the entire confounder space calculated by the adjustment. Furthermore, we design a teacher–student network to utilize the matching of music videos, which is professionally generated content (PGC) with specialized matching, to better recommend content-matching background music. The PGC data are modeled by a teacher network to guide the matching of uploader-selected UGC data of student network by Kullback–Leibler–based knowledge transfer. Extensive experiments on the TT-150k-genre dataset demonstrate the effectiveness of the proposed method. The code is publicly available on https://github.com/jing-1/DecCM

Application of Social Network Analysis for Comparison and Ranking of Internet Service Providers

In this digital era, the Internet has become a basic necessity in life. This has had a significant impact on the growth of internet service provider (ISP) companies in Indonesia. Comparison and ranking of ISPs is needed to make it easier for users to choose services according to their needs as well as to encourage healthy competition between ISPs in improving their services. The problem is ranking ISPs using conventional methods (surveys) to obtain primary data is expensive and takes a long time. On the other hand, Social Network Analysis (SNA) is a method that has been widely used to understand customer desires by extracting information from social media. This information is in the form of User Generated Content (UGC), namely track records left by customers on social media. This research aims to measure the ISP rankings of Indihome, Biznet and FirstMedia using UGC data. The research method used is to collect consumer tweet data rapidly, carry out preprocessing to eliminate irrelevant data and apply SNA, including network structure analysis in the form of visualization and network property analysis with the Gephi application, as well as network content analysis in the form of sentiment analysis and WordCloud analysis. The number of dominant network properties and sentiment analysis calculates ISP ranking. Apart from that, the results of this SNA are in the form of recommendations for ISPs to improve services to customers.

Exploring the user-generated content data to improve quality management

PurposeThe focus of this research is to investigate if user-generated content (UGC) generated in the Booking platform can support quality management improvement within the hospitality industry by increasing customer satisfaction and eliminating defects more efficiently. Hence, it contributes to understanding how data-driven companies can rely on customer data to focus on innovation and performance improvement to meet customer requirements, eliminate defects and increase customer satisfaction.Design/methodology/approachFollowing the literature review, information was collected from the digital platform Booking, encompassing 15 hotel industry companies in Portugal Porto and Braga regions, selected due to their high number of customer reviews. This data was organized and categorized, eliminating all unnecessary information for the research and building an Excel database. The database was subsequently analysed with SPSS and Voyant software, performing statistical analysis, hypothesis testing and text-mining techniques to analyse the comments. After these analyses, applying quality tools allowed for more in-depth conclusions.FindingsThe research results highlight that customers' most relevant requirements in the Portuguese hospitality industry are breakfast, parking and a swimming pool. It was also possible to realize that the location is an attractive requirement, the bathroom is a must-be requirement and breakfast is a performance requirement. The results also allowed us to answer the most critical research question: “Is user-generated content a valuable aid to quality?” the answer is yes since it was possible to use the data to find improvements and faults/failures in the services.Originality/valueThe results of this study represent an essential step towards a complete understanding of how to take advantage of UGC within the hospitality industry by establishing a solid base of techniques, methods and quality tools for UGC analysis that can be applied in future research on different industry sectors.

Research on the perception of the terrain image of the tourism destination based on multimodal user-generated content data.

Destination image is a powerful means by which destinations compete in the tourism industry, and the accurate identification of a destination image better serves destination marketing and management. This study uses multimodal data, such as text, images, and videos uploaded by tourists, to construct a comprehensive and systematic destination image process. The "cognitive-emotional-overall image" model, latent Dirichlet allocation (LDA) model, and deep residual neural networks are implemented to build a framework to examine the perception of a destination image, travelogues, and short videos from the sources called Ctrip, Qunar, and TikTok. The results show that tourists' overall perception of Sanya is based mainly on the cognitive image of natural scenery, human resources, and food. In addition, there are differences between textual and visual cognitive images among the perceptual images when multimodal data is under consideration. Furthermore, tourists have an overall positive affective image of Sanya as a destination.

ANALISIS JARINGAN SOSIAL PENGGUNA TWITTER DALAM SKEMA PENYEDIAAN LAPTOP BAGI TOKO KOMPUTER

Twitter is a social media that provides information for its users. Interactions on Twitter may increase the amount of data from its users. Business development cannot be separated from product competition between business actors. One way for companies to win the competition is to know the purchasing power of customers. A product's selling power can be known one way or another through user interaction on Twitter. User-generated content (UGC) results from data on Twitter. UGC data can be used to determine the product supply scheme taken by laptop products in this study. The method used to analyze the interaction is Social Network Analysis (SNA) by comparing the properties of the network. Social network modeling on UGC data was carried out on three laptops: Lenovo v14, Asus ZenBook S, and Acer Swift 3. From the graph visualization, it was found that Acer excels in two properties, namely "Size" and "Average Degree." Meanwhile, the Asus ZenBook S dominates the "Modularity" property.

Use of social media data, online reviews and wikipedia page views to measure visitation patterns of outdoor attractions

This study explores the suitability of activity counts extracted from social media platforms (Twitter, Flickr), review portals (TripAdvisor, Google Maps) and Wikipedia article views to model official visitor counts at selected outdoor attractions in Florida (U.S.) and Carinthia (Austria). It applies correlation analysis, multiple regression, and time series analysis to identify which of these user-generated content (UGC) sources and their combinations best match official monthly visitor count patterns for an analysis period of three years (2019–2021). With travel activities being severely hampered during 2020 due to the COVID-19 pandemic, the analysis also aims to analyze to which extent reduced visitor counts are reflected in the respective UGC sources. Results show that the number of Google Maps reviews combined with Wikipedia pageviews best explain the variability of monthly official visitor counts in Ordinary Least Squares (OLS) regression in both study areas. While the comparison analysis was conducted for monthly counts, data from some UGC sources can reflect shorter term activity fluctuations. Time series analysis detected a seasonality of 12 months for Wikipedia pageviews, Google Maps reviews, and official visitor counts for Austria due a clearly distinct summer season. As opposed to this, for Florida, due to a climate that facilitates all-year round park visitations, periodograms yielded different seasonality frequencies for UGC sources and official visitor counts. A short-term drop in visitor counts due to COVID-19 was evident in Florida both in UGC sources and official visitor counts in spring 2020, whereas for Austria attractions reduced activity is only somewhat reflected in UGC sources but not in actual visitor counts due to attractions still being closed in April. Management implicationsThis research explores multiple UGC sources for modeling official visitor counts at outdoor attractions. A combination of several sources has the advantage that it can help mitigate known limitations of UGC from individual sources, such as sparsity of geodata, data retrieval restrictions, sociodemographic bias, and varying popularity across regions. This research also revealed a better model fit of combined UGC data with official visitor counts than using any of the UGC sources alone. Data-rich UGC sources offer daily or weekly activity counts that provide a more refined temporal resolution than typical official visitor counts which are often limited to monthly data aggregations.

Study on the Perceived Brand Image of Nongfu Spring based on UGC Data

Brand image is a key factor in attracting consumers, and User Generated Content (UGC) on e-commerce platforms accurately reflects consumers' perception and preferences for brand image to a certain extent.This paper takes the reviews of products sold on the JD platform as samples, and takes the reviews of Nongfu Spring products before 2023 as the research object. Through collecting UGC and using word frequency analysis, semantic network analysis, and emotion analysis in the ROST CM6.0 analysis software, this paper conducts a systematic study on the brand image perception of Nongfu Spring products. Research has found that consumers' perception of the image of Nongfu Spring products can be divided into three dimensions: product, service, and price. In addition, consumers show a positive emotional tone towards the product quality and logistics of Nongfu Spring products, and the factors that affect the negative emotions of Nongfu Spring include individual product issues and slow delivery. Finally, propose discussions and suggestions.

Quality Assessment of UGC Videos Based on Decomposition and Recomposition

The prevalence of short-video applications imposes more requirements for video quality assessment (VQA). User-generated content (UGC) videos are captured under an unprofessional environment, thus suffering from various dynamic degradations, such as camera shaking. To cover the dynamic degradations, existing recurrent neural network-based UGC-VQA methods can only provide implicit modeling, which is unclear and difficult to analyze. In this work, we consider explicit motion representation for dynamic degradations, and propose a motion-enhanced UGC-VQA method based on decomposition and recomposition. In the decomposition stage, a dual-stream decomposition module is built, and VQA task is decomposed into single frame-based quality assessment problem and cross frames-based motion understanding. The dual streams are well grounded on the two-pathway visual system during perception, and require no extra UGC data due to knowledge transfer. Hierarchical features from shallow to deep layers are gathered to narrow the gaps from tasks and domains. In the recomposition stage, a progressively residual aggregation module is built to recompose features from the dual streams. Representations with different layers and pathways are interacted and aggregated in a progressive and residual manner, which keeps a good trade-off between representation deficiency and redundancy. Extensive experiments on UGC-VQA databases verify that our method achieves the state-of-the-art performance and keeps a good capability of generalization. The source code will be available in <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Sissuire/DSD-PRO</uri> .

Deep learning-based social media mining for user experience analysis: A case study of smart home products

Understanding and enhancing user experience (UX) is crucial for new product innovation. Abundant user-generated content (UGC) from social media contains information about customers' product experience and provides an alternative channel for firms to understand UX and improve their products. However, only a few studies have focused on this issue. This research develops a deep learning-based methodology to identify the major UX elements from UGC and analyze their relationships for improving customers' product experiences. The state-of-the-art deep learning approach BERT (Bidirectional Encoder Representations from Transformers) is used to identify the major UX elements from UGC. The Plutchik's wheel of emotions model is used to elaborate users' complex emotional experiences. Association rule mining (ARM) is employed to extract significant patterns of association between the major UX elements. The UGC data from an online discussion group for smart home products is used as an example. The results demonstrate that the methodology can effectively identify relevant UX content and the important relationships between major UX elements for improving products and services. Further, the methodology can help companies better understand UX based on multiple emotional states and develop actions that respond more effectively to user behaviors triggered by their emotional states.

Protecting the anonymity of online users through Bayesian data synthesis

Privacy concerns emerge when online users of popular user-generated content (UGC) platforms are identified through a combination of their structured data (e.g., location and name) and textual content (e.g., word choices and writing style). To overcome this problem, we introduce a Bayesian sequential synthesis methodology for organizations to share structured data adjoined to textual content. Our proposed approach enables platforms to use a single shrinkage parameter to control the privacy level of their released UGC data. Our results show that our synthesis strategy decreases the probability of identification of a user to an acceptable threshold while maintaining much of the textual content present in the structured data. Additionally, we find that the value of sharing our protected data exceeds that of sharing the unprotected structured data and textual content separately. These findings encourage UGC platforms that wish to be known for consumer privacy to protect anonymity of their online users with synthetic data.

Study on Tourism Consumer Behavior and Countermeasures Based on Big Data.

In our study, through consulting, summarizing, and analyzing a large number of related literature studies on tourism consumer behavior, tourism big data, text data analysis, and so on, a framework of research ideas on tourism consumption was constructed. The train browser, NLPIR, and other software packages are used to crawl, preprocess, and mine the travel sample data, and the word frequency analysis, co-occurrence analysis, content analysis, sentiment analysis, network analysis, and other methods are used to analyze the characteristics and decision-making behavior of tourists. Based on the results of behavioral analysis, we proposed tourism development strategies from three aspects: reforming and promoting tourism marketing strategies, improving tourism product and service quality, and improving tourism destination management methods. The results show that (1) for the tourist characteristics, taking into account the factors of climate and geographical location, the domestic market is divided into four grades of markets, and different marketing strategies are adopted according to different market characteristics; (2) for the tourism decision-making behavior, a “push-pull resistance” tourism decision-making model was established through word frequency analysis, co-occurrence analysis, and content analysis; (3) for the tourism consumption preferences, through network analysis of scenic spots, it is found that there are three tourist routes preferred by tourists; and (4) for the tourism perception evaluation behavior, based on the “cognitive-emotional” model, this study describes the tourism image from the two dimensions of the cognitive image and emotional image. Generally speaking, tourists show a positive perception state. The research on tourism consumer behavior based on UGC (user-generated content) data can help scenic spots and other tourism companies to understand the characteristics and rules of tourists' behavior, understand the consumption preferences of different tourism groups, develop diversified tourism products, improve the quality of tourism services, and further cater to market segments. This research provides a new idea for tourist attractions and tourism management departments to monitor tourist behavior through big data analysis.

Mapping a historic neighbourhood through user-generated content: the case of Alfama, Lisbon (Portugal)

Abstract. Participant-based methods aimed at extracting neighbourhood definitions are labour and time intensive. On the other hand, user-generated content (UGC) can provide locations to assess the extent of neighbourhoods. We investigated the definitions of Alfama - a historic neighbourhood in Lisbon (Portugal) - using six sources of UGC and applied a modification of the DBSCAN algorithm developed in the literature. By generating shapes from each source, we were able to visually and quantitatively evaluate their agreement as well as their differences.We demonstrate how different profiles of user activity from each source yielded varied geographies of Alfama. Although discrete representations are not the optimal choice, practical applications such as urban planning usually demand sharp definitions. Lastly, our approach can be extended and improved by adding more sources of UGC data and by picking other case studies.

What people TikTok (Douyin) about influencer-endorsed short videos on wine? An exploration of gender and generational differences

顾客在 TikTok（抖音）对网红葡萄酒营销短视频有何见解？探索性别和代际差异研究目的本研究旨在了解社交媒体用户对最受欢迎的平台 - 抖音- 的网红认可的葡萄酒短视频的反应, 依靠消息解释过程 (MIP) 框架, 并识别性别和世代群组的细分差异（Z世代和 Y 世代）。研究设计/方法论利用主题建模、内容分析和定量分析来破译从30 名社交媒体网红发布的 81 个葡萄酒视频中关注葡萄酒产品和消费体验信息检索到的 UGC 数据。研究发现结果表明, 社交媒体用户最倾向于回应产品相关话题, 其次是影响者、情绪、饮酒意图、行为和怀疑态度。同时, 调查结果还显示, 女性群体更可能参与有社交媒体网红主题的话题, 而男性群体对产品主题的贡献更大。 Z 世代队列涉及更多关于饮酒意图的主题, 而 Y 世代则表现更多怀疑。研究实践意义研究结果为酒店和葡萄酒从业者、社交媒体影响者和品牌赞助商提供了宝贵的信息, 以在各种社交网络服务中使用这种先进的策略。研究原创性/价值这项研究首次通过 UGC 数据掌握了抖音用户对这种新兴的网红营销策略的看法。通过 MIP 体现性别和代际群体的差异从而体现本研究的原创价值。

HeBERT and HebEMO: A Hebrew BERT Model and a Tool for Polarity Analysis and Emotion Recognition

Sentiment analysis of user-generated content (UGC) can provide valuable information across numerous domains, including marketing, psychology, and public health. Currently, there are very few Hebrew models for natural language processing in general, and for sentiment analysis in particular; indeed, it is not straightforward to develop such models because Hebrew is a morphologically rich language (MRL) with challenging characteristics. Moreover, the only available Hebrew sentiment analysis model, based on a recurrent neural network, was developed for polarity analysis (classifying text as positive, negative, or neutral) and was not used for detection of finer-grained emotions (e.g., anger, fear, or joy). To address these gaps, this paper introduces HeBERT and HebEMO. HeBERT is a transformer-based model for modern Hebrew text, which relies on a BERT (bidirectional encoder representations from transformers) architecture. BERT has been shown to outperform alternative architectures in sentiment analysis and is suggested to be particularly appropriate for MRLs. Analyzing multiple BERT specifications, we find that whereas model complexity correlates with high performance on language tasks that aim to understand terms in a sentence, a more parsimonious model better captures the sentiment of an entire sentence. Notably, regardless of the complexity of the BERT specification, our BERT-based language model outperforms all existing Hebrew alternatives on all language tasks examined. HebEMO is a tool that uses HeBERT to detect polarity and extract emotions from Hebrew UGC. HebEMO is trained on a unique COVID-19-related UGC data set that we collected and annotated for this study. Data collection and annotation followed an active learning procedure that aimed to maximize predictability. We show that HebEMO yields a better performance accuracy for polarity classification. Emotion detection reaches high performance for various target emotions, with the exception of surprise, which the model failed to capture. These results are better than the best reported performance, even among English-language models of emotion detection.

Big Data in Forecasting Research: A Literature Review

With the boom in Internet techniques and computer science, a variety of big data have been introduced into forecasting research, bringing new knowledge and improving prediction models. This paper is the first attempt to conduct a literature review on full-scale big data in forecasting research. By source, big data in forecasting research fell into user-generated content data (from the users on social media in texts, photos, etc.), device-monitored data (by meteorological monitors, smart meters, GPS, etc.) and activity log data (for web searching/visiting, online/offline marketing, clinical treatments, laboratory experiments, etc.). Different data types, bearing distinctive information and characteristics, dominated different forecasting tasks, required different analysis technologies and improved different forecasting models. This survey provides an overall review of big data-based forecasting research, details what (regarding data types and sources), where (forecasting hotspots) and how (analysis and forecasting methods used) big data improved prediction, and offers insights into future prospects. • A review on full-scale big data in forecasting research is presented. • Big data in forecasting research fell into UGC data, device data and log data. • Three steps were taken: data collection, data process and prediction improvement. • For each data type, what, where and how big data improved prediction are detailed.

User-generated content data analysis using machine learning methods: a case study in Bangkok, Thailand

User-generated content data analysis using machine learning methods: a case study in Bangkok, Thailand

Evaluating green resource branding using user-generated content data: The case study of a greenway in eastern Guangzhou, China

Examining the functional conditions of completed infrastructure will contribute new knowledge of practical performance to the original place branding. This paper evaluates greenway branding as a type of green resource branding with particular attention to media promotion, users’ outdoor activity trajectories, and the image content of geo-tagged photos. We found that greenway infrastructure in Zengcheng, Guangzhou increasingly generated concerns from more visitors and residents by the media. However, according to geographical activity information, only 35% of outdoor activities in Zengcheng take place within the greenway. Meanwhile, the average staying time of outdoor users at the greenway is shorter than that on ordinary roads. Our analysis of the number of geo-tagged photos indicates that outdoor users are more interested in the scenery outside the greenway than that in and along it. Comparing outdoor users’ staying time and the content of photos, our research shows that the public finds Zengcheng’s greenway less attractive for spending leisure time and taking landscape photos. Additionally, our image content analysis indicates that the greenway images taken by outdoor users include mainly trees, sky, and roads, while those of mountains and water are less common due to the limited variety of scenery. The implication of our research is that greenway usage can be improved by involving outdoor users in greenway planning, enhancing the variety of scenery along greenways, and establishing connections to rural resorts.

Understanding Customers’ Transport Services with Topic Clustering and Sentiment Analysis

The recent increase in user interaction with social media has completely changed the way customers communicate their opinions, questions, and concerns to brands. For this reason, many companies have established on the top of their agendas the necessity of analyzing the high amounts of user-generated content data in social networks. These analyses are helping brands to understand their customers’ experiences as well as for maintaining a competitive advantage in the sector. Due to this fact, this study aims to analyze and characterize the public opinions from the messages posted by Twitter users while addressing customer services. For this purpose, this study carried out a content analysis of a customer service platform. We extracted the general users’ viewpoints and sentiments of each of the discussed topics by using a wide range of techniques, such as topic modeling, document clustering, and opinion mining algorithms. For training these systems and drawing conclusions, a dataset containing tweets from the English-speaking customers addressing the @Uber_Support platform during the year 2020 has been used.