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.

Content analysis, restricted within the limits of written textual documents (wtdca), is a field which is greatly in need of extensive interdisciplinary research. This would clarify certain concepts, especially those concerned with ‘text’, as a new central nucleus of semiotic research, and ‘content’, or the informative power of text. The objective reality (syntax) of the written document should be, in the cognitive process that all content analysis entails, interpreted (semantically and pragmatically) in an intersubjective manner with regard to the context, the analyst's knowledge base and the documentary objectives. The contributions of semiolinguistics (textual), logic (formal) and psychology (cognitive) are fundamental to the conduct of these activities. The criteria used to validate the results obtained complete the necessary conceptual reference panorama.

The purpose of this paper is to longitudinally analyse the evolution of multiple narrative textual characteristics in the chairman’s statements of Guinness from 1948 to 1996, with the aim of studying impression management influences. It attempts to contribute insights on impression management over time.

The paper attempts to contribute to external accounting communication literature, by building on the socio-psychological tradition within the functionalist-behavioural transmission perspective. The paper analyses multiple textual characteristics (positive, negative, tentative, future and external references, length, numeric references and first person pronouns) over 49 years and their potential relationship to profitability. Other possible disclosure drivers are also controlled.

The findings show that Guinness consistently used qualitative textual characteristics with a self-serving bias, but did not use those with a more quantitative character. Continual profits achieved by the company, and the high corporate/personal reputation of the company/chairpersons, inter alia, may well explain limited evidence of impression management associated with quantitative textual characteristics. The context appears related to the evolution of the broad communication pattern.

Impression management is likely to be present in some form in corporate disclosures of most companies, not only those companies with losses. If successful, financial reporting quality may be undermined and capital misallocations may result. Companies with a high public exposure such as those with a high reputation or profitability may use impression management in a different way.

Studies analysing multiple textual characteristics in corporate narratives tend to focus on different companies in a single year, or in two consecutive years. This study analyses multiple textual characteristics over many consecutive years. It also gives an original historical perspective, by studying how impression management relates to its context, as demonstrated by a unique data set. In addition, by using the same company, the possibility that different corporate characteristics between companies will affect results is removed. Moreover, Guinness, a well-known international company, was somewhat unique as it achieved continual profits.

The purpose of this paper is to provide insightful evidence of phenomena in organization and management theory. Textual data sets consist of two different elements, namely qualitative and quantitative aspects. Researchers often combine methods to harness both aspects. However, they frequently do this in a comparative, convergent, or sequential way.

The paper illustrates and discusses a hybrid textual data analysis approach employing the qualitative software application GABEK-WinRelan in a case study of an Austrian retail bank.

The paper argues that a hybrid analysis method, fully intertwining qualitative and quantitative analysis simultaneously on the same textual data set, can deliver new insight into more facets of a data set.

A hybrid approach is not a universally applicable solution to approaching research and management problems. Rather, this paper aims at triggering and intensifying scientific discussion about stronger integration of qualitative and quantitative data and analysis methods in management research.

The purpose of this paper is to apply the wavelet thresholding technique in order to analyze economic socio-political situations in Tunisia using textual data sets. This technique is used to remove noise from contingency table. A comparative study is done on correspondence analysis and classification results (using k-means algorithm) before and after denoising.

Textual data set is collected from an electronic newspaper that offers actual economic news about Tunisia. Both the hard and the soft-thresholding techniques are applied based on various Daubechies wavelets with different vanishing moments.

The results obtained have proved the effectiveness of wavelet denoising method in textual data analysis. On one hand, this technique allowed reducing the loss of information generated by correspondence analysis, ensured a better quality of representation of the factorial plan, neglected the interest of lemmatization in textual analysis and improved the results of classification by k-means algorithm. On the other hand, the proximities provided by the factorial visualization validate the economic situation of Tunisia during the studied period showing mainly a stable situation before the revolution and a deteriorated one after the revolution.

The results are the first to analyze economic socio-political relations using textual data. The originality of this paper comes also from the joint use of correspondence analysis and wavelet thresholding in textual data analysis.

To provide guidance for the formalised analysis of qualitative data and observations, to raise awareness about systematic analysis and illustrate promising avenues for the application of qualitative methodologies in international marketing research.

Conceptually, the nature of qualitative research, globalisation and its implications for the research landscape, text‐data as a source for international research and equivalence issues in international qualitative research are discussed. The methodology section applies these concepts and analysis challenges to a real‐world example using N*Vivo software.

A 14‐step analytic design is developed, introducing procedures of data analysis and interpretation which help to formalise qualitative research of textual data.

The use of software programs (e.g. N*Vivo) helps to substantiate the analysis and interpretation process of textual data.

Step‐by‐step guidance on performing qualitative analysis of textual data and documenting findings.

The paper is valuable for researchers and practitioners looking for guidance in analysing and interpreting textual data from interviews. Specific support is given for N*Vivo software and its application.

The increasingly booming e-commerce development has stimulated vehicle consumers to express individual reviews through online forum. The purpose of this paper is to probe into the vehicle consumer consumption behavior and make recommendations for potential consumers from textual comments viewpoint.

A big data analytic-based approach is designed to discover vehicle consumer consumption behavior from online perspective. To reduce subjectivity of expert-based approaches, a parallel Naïve Bayes approach is designed to analyze the sentiment analysis, and the Saaty scale-based (SSC) scoring rule is employed to obtain specific sentimental value of attribute class, contributing to the multi-grade sentiment classification. To achieve the intelligent recommendation for potential vehicle customers, a novel SSC-VIKOR approach is developed to prioritize vehicle brand candidates from a big data analytical viewpoint.

The big data analytics argue that “cost-effectiveness” characteristic is the most important factor that vehicle consumers care, and the data mining results enable automakers to better understand consumer consumption behavior.

The case study illustrates the effectiveness of the integrated method, contributing to much more precise operations management on marketing strategy, quality improvement and intelligent recommendation.

Researches of consumer consumption behavior are usually based on survey-based methods, and mostly previous studies about comments analysis focus on binary analysis. The hybrid SSC-VIKOR approach is developed to fill the gap from the big data perspective.

The purpose of this study is to estimate the effect of information disclosure on firm value for firms in the Integrated Latin American Market (MILA) over the period 2011–2017.

The study uses structural equation modeling (SEM), where the latent variable “Disclosure Quality” is measured using five textual analysis variables as indicators. The final sample is composed of 1,412 observations representing 198 firms from which we were able to collect annual reports and financial information required.

The authors find a positive and statistically significant effect of “Disclosure Quality” on firm value. The indirect effect of language on firm value is also captured. Text similarity, negative tone, readability and text length in corporate disclosure are negatively related to firm value while using positive tone is positively related. In the exploratory analysis, the authors have significant effects of textual measures on disclosure quality.

The research is original and unique as it approaches the relation between disclosure quality and market valuation of the firm using SEM for firms participating in the MILA.

El propósito es estimar el efecto de la calidad en la revelación de información sobre el valor de la firma para empresas que hacen parte del mercado integrado Latinoamericano (MILA) durante el periodo 2011–2017.

El estudio utiliza un modelo de ecuaciones estructurales (SEM), donde la variable latente “Calidad de la información” es medida usando cinco variables de análisis textual como indicadores. La muestra final está conpuesta de 1,412 observaciones que representan 198 empresas para las cuales se pudo recolectar los reportes anuales y la información financiera requerida

Nosotros encontramos una relación positiva y estadísticamente significativa entre la variable “Calidad de la información” y valor de la firma. El efecto indirecto del lenguaje en el valor de la firma es igualmente observado. La similaridad en el texto, el tono negativo, la legibilidad y el largo del texto en la revelación corporativa están relacionados de manera negativa con el valor de la firma mientras que el tono positivo está relacionado de manera positiva. En el análisis exploratorio, nosotros encontramos un efecto estadísticamente significativo entre las medidas de texto y la calidad de la revelación.

La investigación es original y única en cuanto aproxima la relación entre calidad de la revelación y desempeño de la firma usando un modelo de ecuaciones estructurales para las firmas participantes del mercado integrado Latinoamericano (MILA).

The purpose of this paper is to investigate the sales impact of different types of online word-of-mouth based on their source (user vs critic) and form (structured vs unstructured).

The paper proposed a model by adopting the heuristic-systematic perspective of information processing and tested it using online movie reviews collected from Rotten Tomatoes. A unique dataset was constructed, which matched critic reviews and user reviews with metadata such as box-office sales and advertisement spending for 90 movies. Sentiment information from the textual contents of both user and critic reviews were text-mined and extracted. Data analyses were used to compare the box-office responsiveness of four types of reviews: user numeric ratings, user text reviews, critic numeric ratings and critic text reviews.

Critic reviews and user reviews influence sales through different forms: while user reviews impact sales through their aggregate numeric ratings, critic reviews exert their impact through textual narratives.

This study provides managerial implications to businesses on how to allocate their resources on different social media-related marketing strategies to maximize the economic value of online user-generated information.

The major contribution of this study is to extend the current understanding of the sales impact of online reviews to their textual aspect, as well as investigate how these textual narratives play different roles when offered by critics and users.

This paper aims to elaborate on current research on fourth party logistics “4PL” by offering a taxonomy that provides a deeper understanding of 4PL service offerings, thus drawing clear frontiers between existing 4PL business models.

The authors collected data using semi-structured interviews conducted with 60 logistics executives working in 44 “4PL” providers located in France. Using automatic analysis of textual data, the authors combined spatial visualisation, clustering analysis and hierarchical descending classification to generate the taxonomy.

Two key dimensions emerged, allowing the authors to clearly identify and distinguish four 4PL business models: the level of reliance on interpersonal relationships and the level of involvement in 4PL service offering. As a result, 4PL providers fall under one of the following business models in the taxonomy: (1) The Metronome, (2) The Architect, (3) The Nostalgic and (4) The Minimalist.

The study focuses on investigating 4PL providers located in France; thus, future studies should explore the classification of 4PL business models across different cultural contexts and social structures.

The findings offer valuable managerial insights for logistics executives and clients of 4PL to better orient their needs, the negotiations and the contracting process with 4PLs.

Using a Lexicometric analysis, the authors develop taxonomy of 4PL service providers based on empirical evidence from logistics executives; the work addresses the existing confusion regarding the conceptualisation of 4PL firms with other types of logistical providers and the role of in/formal interpersonal relationships in the logistical intermediation.