Search results

Along with the ever growingWeb is the proliferation of objectionable content, such as sex, violence, racism, etc. We need efficient tools for classifying and filtering undesirable web content. In this paper, we investigate this problem through WebGuard, our automatic machine learning based pornographic website classification and filtering system. Facing the Internet more and more visual and multimedia as exemplified by pornographic websites, we focus here our attention on the use of skin color related visual content based analysis along with textual and structural content based analysis for improving pornographic website filtering. While the most commercial filtering products on the marketplace are mainly based on textual content‐based analysis such as indicative keywords detection or manually collected black list checking, the originality of our work resides on the addition of structural and visual content‐based analysis to the classical textual content‐based analysis along with several major‐data mining techniques for learning and classifying. Experimented on a testbed of 400 websites including 200 adult sites and 200 non pornographic ones, WebGuard, our Web filtering engine scored a 96.1% classification accuracy rate when only textual and structural content based analysis are used, and 97.4% classification accuracy rate when skin color related visual content based analysis is driven in addition. Further experiments on a black list of 12 311 adult websites manually collected and classified by the French Ministry of Education showed that WebGuard scored 87.82% classification accuracy rate when using only textual and structural content‐based analysis, and 95.62% classification accuracy rate when the visual content‐based analysis is driven in addition. The basic framework of WebGuard can apply to other categorization problems of websites which combine, as most of them do today, textual and visual content.

In this article, we develop and advance an understanding of institutions as multimodal accomplishments. We draw on social semiotics and the linguistic concept of metafunctions to establish the visual as a specific mode of meaning construction. In addition, we make semiotic modes conducive to institutional inquiry by introducing the notion of distinct “modal registers” – specialized configurations of linguistic signs within a particular mode that are adapted and applied in the reproduction of institutions or institutional domains. At the core of our article, we operationalize metafunctions to develop methodology for the analysis of visual registers. We illustrate our approach with data from Corporate Social Responsibility (CSR) reporting in Austria.

The purpose of this paper is to provide an analysis of intellectual capital (IC) disclosures in annual reports (mandatory and voluntary) and draw attention to the specific issues related to the methodology used i.e. content analysis. The focus is to incorporate all forms of IC disclosure – narratives, numbers, and visual images – into the analysis as well as highlight the need to study both quantity (extent) and quality of disclosure.

Using content analysis, this paper analyzes 30 of Malaysia's largest public‐listed companies from the IC disclosure of 2008 annual reports. The results are used to discuss specific methodological issues such as the usage of an IC index, choice of unit of analysis, quantity versus quality, presence/absence versus multiple disclosures, and the usage of narratives, numbers, and visual images.

This paper proposes that themes are the most appropriate recording and counting unit to analyze IC information combining narratives, numbers, and visual images. The discussion finds, among others, that while quantity and quality are highly related, quality of disclosure provides the most insights into the disclosure behavior adopted by companies.

This paper provides methodological guidelines to future IC researchers interested in analyzing the quantity and quality of IC disclosure.

To the best of the authors' knowledge, so far there are no studies published that provide a detailed discussion on ways to capture the quantity and quality of IC information disclosed in annual reports using all three forms of disclosure – narratives, numbers, and visual images.

Iran, home to 22 World Heritage Sites in 2017, is blessed with an abundance of breathtaking scenery and a multitude of historic environments for tourists to enjoy. This exploratory chapter investigates tourists’ perceptions about the nature of heritage tourism in Iran, using the five most important cultural attractions as a starting point. A total of 25 photos and 60 reviews constitute the sample. The data were content-analyzed using a deductive and inductive procedure. The results show the relevance of sensory perceptions as a way of creating a bond with cultural heritage that should be considered in the future when defining a brand for Iran as a destination.

Can sustainability disclosures be a tool for executing image restoration strategies after corporate manslaughter? This is the question explored in this study of Costa Crociere's sustainability reports after the Concordia disaster.

Merging traditional textual content analysis with visual analysis and supported by machine learning tools, this is a predominantly qualitative study framed by legitimacy theory, image restoration theory and impression management.

Costa Crociere's voluntary sustainability reporting is strongly influenced by a mix of text and visual signals that distract readers' attention from the disaster. A “nothing really happened” communication strategy pervades the disclosures, with the only rational motivation being to change perceptions and erase memories of this tragic and avoidable event.

Although the analysis covered multiple sources of corporate information, media coverage was not one of them. A more in-depth exploration of sustainability reporting in the cruise industry, including evidence of similar cases, to test impression management theory would be a worthwhile avenue for future research.

While Costa Crociere technically followed the customary guidelines of disclosing human resource impacts, there was almost no acknowledgement of the people involved in the accident. Costa Concierevastly understated their responsibility for the accident, did not apologize, and conveyed very little remorse. The majority of disclosures centred on disaster recovery management.

The authors discuss why and how a company can overcome a legitimacy threat by completely freezing its voluntary sustainability reporting, and the authors show how a company can restore its image by minimizing specific aspects of an accident and shifting attention from the human victims to corporate operations. Incorporating image recognition driven by AI models and combining the results with narrative disclosures contributes an innovative and original analysis technique to the field of impression management. In addition, this research also contributes to our knowledge on the cruise industry – a sector currently under scrutiny for its ethical, social and environmental practices.

Interactive media strategies and digital tools have enabled advertisers to target children with promotional offers and creative appeals.

Based on theories related to metaphors in advertisements, cognitive comprehension by children, promotional appeals, and presentation techniques, the research for this study comprised a content analysis of 1,980 online banner advertisements with reference to use of metaphors, promotional appeals, creative content, and selling techniques.

The research study concludes that online advertising to children, in contrast to traditional advertising vehicles, is characterized by (a) a vibrant visual metaphor, (b) surfeit of animated content, (c) interactive features, (d) myriad product types, and (e) creative content for a mixed audience of adults and children.

This study argues that the impact and content of the Internet as a new advertising medium are distinctly different from traditional characteristics of television and print.

The purpose of this study is to examine cross-cultural differences in the use of visual tangible cues in local corporate Web sites of six nations.

A quantitative content analysis was used to obtain a numerically based summary of visual tangible cues utilized in 207 Web sites of global service corporations.

A clear pattern of differences was observed in the major visual functions (literal vs symbolic), the use of photographs vs illustrations and the utilization of interactive elements between two groups of nations. Eastern visuals tend to rely more on symbolic visuals performing association function, mixed use of photographs and illustrations and customer endorsement, whereas Western visuals are more likely to perform literal functions, use photographs and feature customer–employee interactions.

International services marketers who are planning a global campaign on the Web can benefit by using differentiated visual strategies, which reflect unique cultural characteristics of a target market.

This study adds a new contribution to an international account of Web services advertising in maintaining a comprehensive understanding of contemporary use of visual tangibles cues. It could benefit global services advertisers with both practical and theoretical implications, for no systematic studies have ever touched the visualization strategies on the Web.

The purpose of this study is to gather insights into digital consumer behaviour related to Chinese restaurents by examining visual contents on Tripadvisor platform.

Using the deep learning approach, this research assessed consumer-posted online content of dining experiences by implementing image analysis and clustering. Text mining using word cloud analysis revealed the most frequently repeated keywords.

First, 4,000 photos of nine Chinese restaurants posted on Tripadvisor’s website were analyzed using image recognition via Inception V3 and Google’s deep learning network; this revealed 12 hierarchical image clusters. Then, an open-questionnaire survey of 125 Chinese respondents investigated consumers’ information needs before visiting a restaurant and after purchasing behavior (motives to share).

This study contributes to culinary marketing development by introducing a new analysis methodology and demonstrating its application by exploring a wide range of keywords and visual images published on the internet.

This research extends and contributes to the literature regarding visual user-generated content in culinary tourism.