Showing papers by "Hong Kong Polytechnic University published in 2011"

Sparse representation or collaborative representation: Which helps face recognition?

Abstract: As a recently proposed technique, sparse representation based classification (SRC) has been widely used for face recognition (FR). SRC first codes a testing sample as a sparse linear combination of all the training samples, and then classifies the testing sample by evaluating which class leads to the minimum representation error. While the importance of sparsity is much emphasized in SRC and many related works, the use of collaborative representation (CR) in SRC is ignored by most literature. However, is it really the l 1 -norm sparsity that improves the FR accuracy? This paper devotes to analyze the working mechanism of SRC, and indicates that it is the CR but not the l1-norm sparsity that makes SRC powerful for face classification. Consequently, we propose a very simple yet much more efficient face classification scheme, namely CR based classification with regularized least square (CRC_RLS). The extensive experiments clearly show that CRC_RLS has very competitive classification results, while it has significantly less complexity than SRC.

Effect of valence electron concentration on stability of fcc or bcc phase in high entropy alloys

Abstract: Phase stability is an important topic for high entropy alloys (HEAs), but the understanding to it is very limited. The capability to predict phase stability from fundamental properties of constituent elements would benefit the alloy design greatly. The relationship between phase stability and physicochemical/thermodynamic properties of alloying components in HEAs was studied systematically. The mixing enthalpy is found to be the key factor controlling the formation of solid solutions or compounds. The stability of fcc and bcc solid solutions is well delineated by the valance electron concentration (VEC). The revealing of the effect of the VEC on the phase stability is vitally important for alloy design and for controlling the mechanical behavior of HEAs.

Nonfinancial Disclosure and Analyst Forecast Accuracy: International Evidence on Corporate Social Responsibility Disclosure

Abstract: We examine the relationship between disclosure of nonfinancial information and analyst forecast accuracy using firm-level data from 31 countries. We use the issuance of standalone corporate social responsibility (CSR) reports to proxy for disclosure of nonfinancial information. We find that the issuance of standalone CSR reports is associated with lower analyst forecast error. This relationship is stronger in countries that are more stakeholder-oriented — i.e., in countries where CSR performance is more likely to affect firm financial performance. The relationship is also stronger for firms and countries with more opaque financial disclosure, suggesting that issuance of standalone CSR reports plays a role complementary to financial disclosure. These results hold after we control for various factors related to firm financial transparency and other potentially confounding institutional factors. Collectively, our findings have important implications for academics and practitioners in understanding the function of CSR disclosure in financial markets.

Fisher Discrimination Dictionary Learning for sparse representation

Abstract: Sparse representation based classification has led to interesting image recognition results, while the dictionary used for sparse coding plays a key role in it. This paper presents a novel dictionary learning (DL) method to improve the pattern classification performance. Based on the Fisher discrimination criterion, a structured dictionary, whose dictionary atoms have correspondence to the class labels, is learned so that the reconstruction error after sparse coding can be used for pattern classification. Meanwhile, the Fisher discrimination criterion is imposed on the coding coefficients so that they have small within-class scatter but big between-class scatter. A new classification scheme associated with the proposed Fisher discrimination DL (FDDL) method is then presented by using both the discriminative information in the reconstruction error and sparse coding coefficients. The proposed FDDL is extensively evaluated on benchmark image databases in comparison with existing sparse representation and DL based classification methods.

Pathological Video Game Use Among Youths: A Two-Year Longitudinal Study

Abstract: OBJECTIVES: We aimed to measure the prevalence and length of the problem of pathological video gaming or Internet use, to identify risk and protective factors, to determine whether pathological gaming is a primary or secondary problem, and to identify outcomes for individuals who become or stop being pathological gamers. METHODS: A 2-year, longitudinal, panel study was performed with a general elementary and secondary school population in Singapore, including 3034 children in grades 3 ( N = 743), 4 ( N = 711), 7 ( N = 916), and 8 ( N = 664). Several hypothesized risk and protective factors for developing or overcoming pathological gaming were measured, including weekly amount of game play, impulsivity, social competence, depression, social phobia, anxiety, and school performance. RESULTS: The prevalence of pathological gaming was similar to that in other countries (∼9%). Greater amounts of gaming, lower social competence, and greater impulsivity seemed to act as risk factors for becoming pathological gamers, whereas depression, anxiety, social phobias, and lower school performance seemed to act as outcomes of pathological gaming. CONCLUSION: This study adds important information to the discussion about whether video game “addiction” is similar to other addictive behaviors, demonstrating that it can last for years and is not solely a symptom of comorbid disorders.

Efficient synthesis of polymeric g-C3N4 layered materials as novel efficient visible light driven photocatalysts

Abstract: In order to develop efficient visible light driven photocatalysts for environmental applications, novel polymeric g-C3N4 layered materials with high surface areas are synthesized efficiently from an oxygen-containing precursor by directly treating urea in air between 450 and 600 °C, without the assistance of a template for the first time. The as-prepared g-C3N4 materials with strong visible light absorption have a band gap around 2.7 eV. The crystallinity and specific surface areas of g-C3N4 increases simultaneously when the heating temperatures increases. The g-C3N4 materials are demonstrated to exhibit much higher visible light photocatalytic activity than that of C-doped TiO2 and g-C3N4 prepared from dicyanamide for the degradation of aqueous RhB. The large surface areas, layered structure and band structure in all contributed to the efficient visible light photocatalytic activity. The efficient synthesis method for g-C3N4 combined with efficient photocatalytic activity is of significant interest for environmental pollutants degradation and solar energy conversion in large scale applications.

The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature

Abstract: This paper presents a review of - and classification scheme for - the literature on the application of data mining techniques for the detection of financial fraud. Although financial fraud detection (FFD) is an emerging topic of great importance, a comprehensive literature review of the subject has yet to be carried out. This paper thus represents the first systematic, identifiable and comprehensive academic literature review of the data mining techniques that have been applied to FFD. 49 journal articles on the subject published between 1997 and 2008 was analyzed and classified into four categories of financial fraud (bank fraud, insurance fraud, securities and commodities fraud, and other related financial fraud) and six classes of data mining techniques (classification, regression, clustering, prediction, outlier detection, and visualization). The findings of this review clearly show that data mining techniques have been applied most extensively to the detection of insurance fraud, although corporate fraud and credit card fraud have also attracted a great deal of attention in recent years. In contrast, we find a distinct lack of research on mortgage fraud, money laundering, and securities and commodities fraud. The main data mining techniques used for FFD are logistic models, neural networks, the Bayesian belief network, and decision trees, all of which provide primary solutions to the problems inherent in the detection and classification of fraudulent data. This paper also addresses the gaps between FFD and the needs of the industry to encourage additional research on neglected topics, and concludes with several suggestions for further FFD research.

The influence of greening the suppliers and green innovation on environmental performance and competitive advantage in Taiwan

Abstract: Recently, many companies have recognized the concepts of green supply chain management or supply chain environmental management. However, relatively little research attention has been devoted to the consideration of relations between greening the supply chain, green innovation, environmental performance and competitive advantage. Hence, this paper aims to bridge this gap by providing empirical evidence to encourage companies to implement green supply chain and green innovation in order to improve their environmental performance, and to enhance their competitive advantage in the global market. A model is constructed to link the aforementioned constructs. Data were collected through a questionnaire-based survey across 124 companies from eight industry sectors in Taiwan. The data are analyzed using Structural Equation Modeling and the results from the final measurement model are used to evaluate the structural model that verifies the significance of the proposed relationships. A prominent result of this study is that greening the supplier through green innovation contributes significant benefits to the environmental performance and competitive advantage of the firm.

A survey of combinatorial testing

Abstract: Combinatorial Testing (CT) can detect failures triggered by interactions of parameters in the Software Under Test (SUT) with a covering array test suite generated by some sampling mechanisms. It has been an active field of research in the last twenty years. This article aims to review previous work on CT, highlights the evolution of CT, and identifies important issues, methods, and applications of CT, with the goal of supporting and directing future practice and research in this area. First, we present the basic concepts and notations of CT. Second, we classify the research on CT into the following categories: modeling for CT, test suite generation, constraints, failure diagnosis, prioritization, metric, evaluation, testing procedure and the application of CT. For each of the categories, we survey the motivation, key issues, solutions, and the current state of research. Then, we review the contribution from different research groups, and present the growing trend of CT research. Finally, we recommend directions for future CT research, including: (1) modeling for CT, (2) improving the existing test suite generation algorithm, (3) improving analysis of testing result, (4) exploring the application of CT to different levels of testing and additional types of systems, (5) conducting more empirical studies to fully understand limitations and strengths of CT, and (6) combining CT with other testing techniques.

Robust sparse coding for face recognition

Abstract: Recently the sparse representation (or coding) based classification (SRC) has been successfully used in face recognition. In SRC, the testing image is represented as a sparse linear combination of the training samples, and the representation fidelity is measured by the l 2 -norm or l 1 -norm of coding residual. Such a sparse coding model actually assumes that the coding residual follows Gaussian or Laplacian distribution, which may not be accurate enough to describe the coding errors in practice. In this paper, we propose a new scheme, namely the robust sparse coding (RSC), by modeling the sparse coding as a sparsity-constrained robust regression problem. The RSC seeks for the MLE (maximum likelihood estimation) solution of the sparse coding problem, and it is much more robust to outliers (e.g., occlusions, corruptions, etc.) than SRC. An efficient iteratively reweighted sparse coding algorithm is proposed to solve the RSC model. Extensive experiments on representative face databases demonstrate that the RSC scheme is much more effective than state-of-the-art methods in dealing with face occlusion, corruption, lighting and expression changes, etc.

A multi-objective optimization for green supply chain network design

Abstract: In this paper, we study a supply chain network design problem with environmental concerns. We are interested in the environmental investments decisions in the design phase and propose a multi-objective optimization model that captures the trade-off between the total cost and the environment influence. We conduct a comprehensive set of numerical experiments. The results show that our model can be applied as an effective tool in the strategic planning for green supply chain. Meanwhile, the sensitivity analysis provides some interesting managerial insights for firms.

Measuring the quality of experience of HTTP video streaming

Abstract: HTTP video streaming, such as Flash video, is widely deployed to deliver stored media. Owing to TCP's reliable service, the picture and sound quality would not be degraded by network impairments, such as high delay and packet loss. However, the network impairments can cause rebuffering events which would result in jerky playback and deform the video's temporal structure. These quality degradations could adversely affect users' quality of experience (QoE). In this paper, we investigate the relationship among three levels of quality of service (QoS) of HTTP video streaming: network QoS, application QoS, and user QoS (i.e., QoE). Our ultimate goal is to understand how the network QoS affects the QoE of HTTP video streaming. Our approach is to first characterize the correlation between the application and network QoS using analytical models and empirical evaluation. The second step is to perform subjective experiments to evaluate the relationship between application QoS and QoE. Our analysis reveals that the frequency of rebuffering is the main factor responsible for the variations in the QoE.

Social ties and international entrepreneurship: Opportunities and constraints affecting firm internationalization

Abstract: International entrepreneurship involves the identification and exploitation of opportunities for international exchange. Yet little is known about the entrepreneurial methods used for opportunity recognition. While previous work emphasizes effects operating at the level of the business network, I propose that the recognition of exchange opportunities is a highly subjective process, shaped by entrepreneurs’ existing ties with others. Based on interview data collected from 41 managers, I develop a comprehensive measure for classifying different methods of opportunity recognition. I then use this measure to classify 665 international exchange ventures set up by entrepreneurs in four Chinese cities. In contrast with past research I find virtually no role for blind luck. Although the majority of exchange opportunities were discovered rather than sought, these discoveries were intentional rather than accidental. I also find that entrepreneurs’ idiosyncratic connections with others both promote and inhibit international exchange. Tie-based opportunities lead to higher-quality and more valuable exchanges that are constrained in terms of geographic, psychic and linguistic distance. From this I conclude that entrepreneurial networks have distinct opportunity horizons that limit the reach of tie-based exchanges and potentially lead to sub-optimal internationalization trajectories.

Comparisons of natural and recycled aggregate concretes prepared with the addition of different mineral admixtures

Abstract: This paper presents the results of a laboratory study on the performance of natural and recycled aggregate concrete prepared with the incorporation of different mineral admixtures including silica fumes (SF), metakaolin (MK), fly ash (FA) and Ground granulated blast slag (GGBS). The compressive and splitting tensile strength, drying shrinkage, chloride ion penetration and ultrasonic pulse velocity (UPV) of the concrete mixtures were determined. The test results, in general, showed that the incorporation of mineral admixtures improved the properties of the recycled aggregate concretes. SF and MK contributed to both the short and long-term properties of the concrete, whereas FA and GGBS showed their beneficial effect only after a relatively long curing time. As far as the compressive strength is concerned, the replacement of cement by 10% of SF or 15% of MK improved both mechanical and durability performance, while the replacement of cement by 35% FA or 55% GGBS decreased the compressive strength, but improved the durability properties of the recycled aggregate concretes. Moreover, the results show that the contributions of the mineral admixtures to performance improvement of the recycled aggregate concrete are higher than that to the natural aggregate concrete.

Investigation of Social Media Marketing: How Does the Hotel Industry in Hong Kong Perform in Marketing on Social Media Websites?

Abstract: The purpose of this study is to investigate the extent to which social media marketing is being utilized in the Hong Kong hotel industry. Marketing performance of 67 hotels in Hong Kong on 23 social media sites was evaluated according to 18 criteria adapted from past studies. The results indicate that hotels generally have a poor performance in using social media to learn about customers. Major problems regarding the hotels' social media marketing efforts are identified. Implications behind these problems and recommendations for improvement are made accordingly.