Search results

This paper aims to propose a novel way of using textual clustering as a feature selection method. It is applied to identify the most important keywords in the profile classification. The method is demonstrated through the problem of sick-leave promoters on Twitter.

Four machine learning classifiers were used on a total of 35,578 tweets posted on Twitter. The data were manually labeled into two categories: promoter and nonpromoter. Classification performance was compared when the proposed clustering feature selection approach and the standard feature selection were applied.

Radom forest achieved the highest accuracy of 95.91% higher than similar work compared. Furthermore, using clustering as a feature selection method improved the Sensitivity of the model from 73.83% to 98.79%. Sensitivity (recall) is the most important measure of classifier performance when detecting promoters’ accounts that have spam-like behavior.

The method applied is novel, more testing is needed in other datasets before generalizing its results.

The model applied can be used by Saudi authorities to report on the accounts that sell sick-leaves online.

The research is proposing a new way textual clustering can be used in feature selection.

Conventional time series modeling may not satisfy the model validity for short-period time series data. In this study, we apply the Kernel Variant Multi-Way Principal Component Analysis (KMPCA) to cluster multivariate time series data which havemultiple dimensions with auto- and cross-correlations. We then check whether this method works well in clustering those data by employing simulation for generalization. Two simulation studies with two different mean structures with nine combinations of auto- and cross-correlations were conducted. The results showed that KMPCA cluster two different mean structure groups over 90% success rates with an appropriate kernel function. We also found that when the mean structures are the same, auto-correlation, the number of temporal points, and the kernel function parameter have the statistically significant effects on clustering performance. The second and third order interaction effects with each of those factors also have effects on clustering success rates. Among the effects of the main factors, the kernel function parameter is the most critical factor to consider for obtaining better performance. A similar error structure may obstruct the clustering performance: strong cross-correlation, weak auto-correlation, and a larger number of temporal points. The paper also discussed some limitations of the KMPCA model and suggested directions for future research that could improve the model.

The purposes of this research are to study the theory and method of multi-attribute index system design and establish a set of systematic, standardized, scientific index systems for the design optimization and inspection process. The research may form the basis for a rational, comprehensive evaluation and provide the most effective way of improving the quality of management decision-making. It is of practical significance to improve the rationality and reliability of the index system and provide standardized, scientific reference standards and theoretical guidance for the design and construction of the index system.

Using modern methods such as complex networks and machine learning, a system for the quality diagnosis of index data and the classification and stratification of index systems is designed. This guarantees the quality of the index data, realizes the scientific classification and stratification of the index system, reduces the subjectivity and randomness of the design of the index system, enhances its objectivity and rationality and lays a solid foundation for the optimal design of the index system.

Based on the ideas of statistics, system theory, machine learning and data mining, the focus in the present research is on “data quality diagnosis” and “index classification and stratification” and clarifying the classification standards and data quality characteristics of index data; a data-quality diagnosis system of “data review – data cleaning – data conversion – data inspection” is established. Using a decision tree, explanatory structural model, cluster analysis, K-means clustering and other methods, classification and hierarchical method system of indicators is designed to reduce the redundancy of indicator data and improve the quality of the data used. Finally, the scientific and standardized classification and hierarchical design of the index system can be realized.

The innovative contributions and research value of the paper are reflected in three aspects. First, a method system for index data quality diagnosis is designed, and multi-source data fusion technology is adopted to ensure the quality of multi-source, heterogeneous and mixed-frequency data of the index system. The second is to design a systematic quality-inspection process for missing data based on the systematic thinking of the whole and the individual. Aiming at the accuracy, reliability, and feasibility of the patched data, a quality-inspection method of patched data based on inversion thought and a unified representation method of data fusion based on a tensor model are proposed. The third is to use the modern method of unsupervised learning to classify and stratify the index system, which reduces the subjectivity and randomness of the design of the index system and enhances its objectivity and rationality.

The purpose of this work is twofold. First, looks to identify the main homogenous groups of companies after environmental, social, economic and governance (ESEG) disclosures, non-financial statement and earnings per share (EPS), and second investigates the connection between variables.

Using financial and non-financial information from annual reports of private listed companies, the authors performed two-step cluster analysis (TSCA) in the first stage of the research, followed by parametric, nonparametric correlation analysis, as well as regression analysis based on panel data, in the second stage.

Results of TSCA revealed a cluster of companies with good financial and non-financial outcomes and a cluster of companies with poor performance. The performance dynamics showed a slight improvement during the period for few companies and composition analysis of clusters by industries through Kruskal–Wallis test highlighted differences between clusters, only for 2017. The main findings confirm a direct, although weak in intensity but statistically significant correlation between ESEG disclosure index, its sustainability component and financial performance (FP), valid for the entire period. Also, the results showed a direct link of low intensity to average, but statistically significant between the non-financial statement and EPS, valid only for 2017 and 2018.

The results indicate mixed findings which invites further in-depth research. Limits of the study can be found in selected indicators and the short period of time analyzed. However, the practical implications are worth considering from the perspective of finding new managerial tools that can better shape the relationship between ESEG disclosures and FP.

ESEG Dindx can be an instrument for managers that can optimize the link between the FP of companies and its sustainable development.

ESEG Dindx measures the disclosure degree of ESEG information by the companies listed on Bucharest Stock Exchange (BSE). The main findings of the work confirm a direct, although weak in intensity but statistically significant correlation between ESEG disclosure index, its sustainability component and FP, valid for the entire period.

This study adds value to the existing literature by the proposed research framework, design of ESEG Dindx and the way correlations between variables were investigated.

In developing countries like Tanzania, gems and jewellery industry mainly consists of disintegrated and unstable micro and small workshops which operate in a way that misalign value addition processes. This study is aimed to bridge gap by focussing on exploitation of industrial clusters in social normalisation and economic resilience to developing countries. The world economic shocks has been not only individually experienced but also globally shared while disrupted lives across all countries and communities and negatively affected global socio-economic growth.

Furthermore, the explorative design was adopted in this study in order to explore needs of respondents, and with the aim to direct the study towards a descriptive design. The sample frame consists of participants in gems and jewellery activities in Tanzania whereby sample was drawn from Dar es Salaam and Arusha. Semi-structured interview was used to collect quantitative data to establish evidence of Tanzanians’ SSJs linked to global value chains (GVCs).

Results revealed the benefits of exploitation of artisanal industrial clusters to Tanzanians’ SSJs when linked to global value chains (GVCs). Findings of the study demonstrate the importance of artisanal industrial clusters in facilitating Tanzanians’ SSJs to access GVCs. Further, insufficient education, trust and social protection directly affects inclusive GVCs, inferring that the impact of artisanal industrial clusters on inclusive GVCs in social normalisation and economic resilience.

Study findings reveals shortcomings in existing regulatory framework of linking Tanzanians’ SSJs to artisanal industrial clusters, for improvements to better support the inclusiveness in GVCs. Findings of this research invite interventions on institutional capabilities and entrepreneurial competencies to enhance the capabilities of small-scale jewellers (SSJs). Like other studies, this study involved cross-sectional data, limit targeted study population as representative of SSJs in industrial clusters and GVCs in economic crises at limited time.

The study findings makes important practical contributions to the Tanzania’s SSJs by examining mediating role of artisanal industrial clusters hence informing policymakers of mining sector how to improve accessibility on GVCs by focus on offering great institutional capabilities and entrepreneurial competencies. These findings will help SSJs and policy makers to get better understanding of the relationships in exploitation of artisanal industrial clusters when accessing GVCs. Therefore, they can make better decisions on implementing artisanal industrial clusters as well as management accessing GVCs, so that SSJs will attain the best possible performance.

This emphasises the importance of community empowerment in the GVCs process through artisanal industrial clusters. Study findings indicate the influence of industrial relations to social dynamics which are previously inadequately addressed and scantly researched. In actual fact study propose initiatives that ensure local communities benefit socially from the integration of SSJs into GVCs through artisanal industrial clusters. Findings suggest local communities that take into account inter-sectionality of artisanal industrial clusters and inclusive GVCs, by considering how factors like education, trust and social protection status intersect to influence the social inclusiveness of SSJs.

There is limited evidence of linking Tanzanians’ SSJs to GVCs in social normalisation and economic resilience and few researchers have explored this topic. This article leverages exploitation of industrial clusters in normalisation and economic resilience to developing countries such as Tanzania as way of improving shared prosperity, sustainability, inclusive growth, cohesion, value chain upgrading and financial inclusion to SSJs.

The contribution of the banking industry to the financial crisis of 2007/8 has raised public concerns about the financial soundness of banks around the world with many countries still suffering the backlogs of this crisis. The continuous emergence of such crises at both national and international levels increases governments', bank regulators' and financial market participants' need for reliable tools to assess the financial soundness of banks. In this context, this study investigates the financial soundness of the Kazakh banking sector, which is ranked by the World Bank as the first in the world in terms of the percentage of nonperforming loans (NPL) to total gross loans in 2012.

Using data about all Kazakh banks over the period January 01, 2008 to January 01, 2014, the study identifies a number of accounting indicators that influence the financial soundness of banks using principal component analysis (PCA). Then, it uses the outcomes of the PCA in a cluster analysis and groups the Kazakh banks into sound, risky and unsound banks at two points in time: January 01, 2008 and January 01, 2014. This methodology was further tested against a ranking system of banks and proved to be more reliable in detecting risky banks.

Fifteen financial ratios were initially selected as accounting indicators for the assessment of bank financial soundness. Using PCA, twelve indicators were isolated, which explain five principal components of capital adequacy, return on assets, profitability, asset quality, liquidity and leverage. Then using the “k-means” method, the results suggest a structure of the Kazakh banking sector on January 01, 2008 that includes two groups of banks: sound and risky banks. On January 01, 2014, this structure of the banking system has changed to include three groups of banks: sound, risky and unsound banks. Thus, in 2014 a new group of banks has emerged, i.e. financially unsound banks.

The proposed cluster-based methodology has proven to be a reliable tool to detect the financial soundness of Kazakh banks, which makes us advocate its employability for bank monitoring and supervision purposes.

This study is the first to employ a cluster-based methodology to assess the financial soundness of a banking sector. This methodology can be used at a micro-level to determine the structure of a banking sector. Also, it can be used to monitor any changes in the structure of a banking sector and provide early warning signals about the financial health of banks.

There are two aims of this research: one is to prove the interactive effect of supply chain integration (SCI) on performance and the other is to ascertain gaps in performance among levels of SCI. The population of this research is international freight forwarders and the collected data is used in testing hypotheses through various analytical methods such as factor analysis, Cronbach’s alpha, cluster analysis, ANOVA, MANOVA, ANCOVA, post hoc analysis and regression analysis. First, the interaction between internal integration and external integration improves customer performance and financial performance. The forwarders improve internal processes following information acquired from customers and the information is shared with logistics service providers, followed by high performance. Second, gaps in performance among levels of SCI are verified. Managers of the forwarders make strategic decision making on the basis of their recognition of environment and, as a result, the forwarders enjoy different performance.

European countries are likely to increasingly adopt integrated reporting (IR) voluntarily, after the 2014/95/EU Directive is revised and other initiatives are implemented. Therefore, the present study provides insights on the relevance of IR in voluntary contexts by exploring analysts' reactions to the release of integrated reports in diverse institutional settings.

Drawing on voluntary disclosure theory, a quantitative empirical research method is used to explore the moderating role of country-level institutional characteristics on the associations between voluntary IR release and analyst forecast accuracy and dispersion.

IR informativeness is not uniform in the voluntary context and institutional settings play a moderating role. IR release is associated with increased consensus among analyst forecasts. However, in countries with weak institutional enforcement, a reverse association is detected, indicating that analysts rely largely on IR where the institutional setting strongly protects investors. Although a strong institutional setting boosts the IR release usefulness in terms of accuracy, it creates noise in analyst consensus.

Academics can appreciate the usefulness of voluntary IR across the institutional enforcement contexts.

Managers can use these findings to understand opportunities offered by IR voluntary release. The study recommends that policymakers, standard setters and regulators strengthen the institutional enforcement of sustainability disclosure.

This study is a unique contribution to recent calls for research on the effects of nonfinancial disclosure regulation and on IR “impacts”. It shows on the international scale that IR usefulness for analysts is moderated by institutional patterns, not country-level institutional characteristics.

The General Society Survey(GSS) is a kind of government-funded survey which aims at examining the Socio-economic status, quality of life, and structure of contemporary society. GSS data set is regarded as one of the authoritative source for the government and organization practitioners to make data-driven policies. The previous analytic approaches for GSS data set are designed by combining expert knowledges and simple statistics. By utilizing the emerging data mining algorithms, we proposed a comprehensive data management and data mining approach for GSS data sets.

The approach are designed to be operated in a two-phase manner: a data management phase which can improve the quality of GSS data by performing attribute pre-processing and filter-based attribute selection; a data mining phase which can extract hidden knowledge from the data set by performing data mining analysis including prediction analysis, classification analysis, association analysis and clustering analysis.

According to experimental evaluation results, the paper have the following findings: Performing attribute selection on GSS data set can increase the performance of both classification analysis and clustering analysis; all the data mining analysis can effectively extract hidden knowledge from the GSS data set; the knowledge generated by different data mining analysis can somehow cross-validate each other.

By leveraging the power of data mining techniques, the proposed approach can explore knowledge in a fine-grained manner with minimum human interference. Experiments on Chinese General Social Survey data set are conducted at the end to evaluate the performance of our approach.

The service marketing literature has traditionally argued that the marketing practices of service firms that operate in diverse cultures should also differ. This paper aims to investigate this argument by examining the marketing practices of service firms in two highly diverse countries “Canada and Qatar” in the context of a contemporary conceptual framework.

Survey data were collected in both countries using a self-administered questionnaire that was used in previous contemporary marketing practice (CMP) studies. The data analysis was conducted in two stages. First, descriptive statistics were used to determine cross-national differences in the intensity of use of various CMP activities in Qatar compared to Canada. Second, cross-national differences in various combinations of marketing practices were identified using a cluster analysis.

The results indicate that service firms in both countries have more similarities than differences and that the overall patterns of marketing practices are similar. In addition, the firms’ marketing practices reflect aspects of all four marketing approaches rather than just one.

The study was conducted in only two countries, thus generalisability of its findings and conclusions may not be possible.

The results of this study can help marketers to better understand the changing marketing environment and identify new marketing solutions when operating in different environments.

This study enhances the literature on service marketing and expands the application of the CMP framework to a new context that has not been addressed in previous studies.