Text mining and probabilistic language modeling for online review spam detecting

In the era of Web 2.0, huge volumes of consumer reviews are posted to the Internet every day. Manual approaches to detecting and analyzing fake reviews (i.e., spam) are not practical due to the problem of information overload. However, the design and development of automated methods of detecting fake reviews is a challenging research problem. The main reason is that fake reviews are specifically composed to mislead readers, so they may appear the same as legitimate reviews (i.e., ham). As a result, discriminatory features that would enable individual reviews to be classified as spam or ham may not be available. Guided by the design science research methodology, the main contribution of this study is the design and instantiation of novel computational models for detecting fake reviews. In particular, a novel text mining model is developed and integrated into a semantic language model for the detection of untruthful reviews. The models are then evaluated based on a real-world dataset collected from amazon.com. The results of our experiments confirm that the proposed models outperform other well-known baseline models in detecting fake reviews. To the best of our knowledge, the work discussed in this article represents the first successful attempt to apply text mining methods and semantic language models to the detection of fake consumer reviews. A managerial implication of our research is that firms can apply our design artifacts to monitor online consumer reviews to develop effective marketing or product design strategies based on genuine consumer feedback posted to the Internet.

Mining a data reasoning model for personalized text classification

It is a big challenge to acquire correct user profiles for personalized text classification since users may be unsure in providing their interests. Traditional approaches to user profiling adopt machine learning (ML) to automatically discover classification knowledge from explicit user feedback in describing personal interests. However, the accuracy of ML-based methods cannot be significantly improved in many cases due to the term independence assumption and uncertainties associated with them. This paper presents a novel relevance feedback approach for personalized text classification. It basically applies data mining to discover knowledge from relevant and non-relevant text and constraints specific knowledge by reasoning rules to eliminate some conflicting information. We also developed a Dempster-Shafer (DS) approach as the means to utilise the specific knowledge to build high-quality data models for classification. The experimental results conducted on Reuters Corpus Volume 1 and TREC topics support that the proposed technique achieves encouraging performance in comparing with the state-of-the-art relevance feedback models.

Running dry : Brisbane's public fountains

"Does heat have a cooling effect on culture? Sweat argues the reverse: culture thrives in the subtropical zones. While acknowledging that the subtropical generates ambivalence—being cast as alternately idyllic or hellish—Sweat nonetheless seeks to develop the specific voices of subtropical cultures. The uneasy place of this sweaty discourse is explored across art, literature, architecture, and the built environment. In particular, Sweat focuses on the most commonly experienced situation, the everyday house. While it addresses subjects from Japan, Brazil, and France, Sweat centres on Brisbane, Queensland—long in the shadow of Sydney and Melbourne in the Australian cultural psyche—due to its enduring and self-conscious attention to subtropical living..." -- online book description

Loss and grief at the heart of text messages [REVIEW]

WHAT if you lost someone you loved? What if you had to let go for the sake of your own sanity? Lachlan Philpott's Colder and Dennis Kelly's Orphans, playing as part of La Boite's and Queensland Theatre Company's independents programs, are emotionally and textually dense theatrical works...

Unsupervised multi-label text classification using a world knowledge ontology

The development of text classification techniques has been largely promoted in the past decade due to the increasing availability and widespread use of digital documents. Usually, the performance of text classification relies on the quality of categories and the accuracy of classifiers learned from samples. When training samples are unavailable or categories are unqualified, text classification performance would be degraded. In this paper, we propose an unsupervised multi-label text classification method to classify documents using a large set of categories stored in a world ontology. The approach has been promisingly evaluated by compared with typical text classification methods, using a real-world document collection and based on the ground truth encoded by human experts.

Recognising the capacities of dynamic reconfiguration for the QoS assurance of running systems in concurrent and parallel environments

Recognizing the impact of reconfiguration on the QoS of running systems is especially necessary for choosing an appropriate approach to dealing with dynamic evolution of mission-critical or non-stop business systems. The rationale is that the impaired QoS caused by inappropriate use of dynamic approaches is unacceptable for such running systems. To predict in advance the impact, the challenge is two-fold. First, a unified benchmark is necessary to expose QoS problems of existing dynamic approaches. Second, an abstract representation is necessary to provide a basis for modeling and comparing the QoS of existing and new dynamic reconfiguration approaches. Our previous work [8] has successfully evaluated the QoS assurance capabilities of existing dynamic approaches and provided guidance of appropriate use of particular approaches. This paper reinvestigates our evaluations, extending them into concurrent and parallel environments by abstracting hardware and software conditions to design an evaluation context. We report the new evaluation results and conclude with updated impact analysis and guidance.

Text mining in negative relevance feedback

It is a big challenge to clearly identify the boundary between positive and negative streams. Several attempts have used negative feedback to solve this challenge; however, there are two issues for using negative relevance feedback to improve the effectiveness of information filtering. The first one is how to select constructive negative samples in order to reduce the space of negative documents. The second issue is how to decide noisy extracted features that should be updated based on the selected negative samples. This paper proposes a pattern mining based approach to select some offenders from the negative documents, where an offender can be used to reduce the side effects of noisy features. It also classifies extracted features (i.e., terms) into three categories: positive specific terms, general terms, and negative specific terms. In this way, multiple revising strategies can be used to update extracted features. An iterative learning algorithm is also proposed to implement this approach on RCV1, and substantial experiments show that the proposed approach achieves encouraging performance.

Four years on – are the Gazelles still running? A longitudinal study of firm performance after a period of rapid growth

Gazelles, or very rapidly growing firms, are important because they contribute disproportionately to economic growth. There is a concern that some of these firms pursue growth too aggressively resulting in lower subsequent performance. We investigate the relationship between growth and subsequent profitability for gazelle firms, and how this is moderated by firm strategy. Previous empirical research regarding the growth-profitability relationship for firms in general is rather inconclusive, with only one study specifically investigating gazelle firms. Likewise, there are theoretical arguments both for and against growth leading to profitability that equally apply to gazelle firms. Further, while contingency theory might suggest the relationship depends on the firm’s strategy, earlier studies have not investigated this relationship. We address these questions using longitudinal data (seven years) for a sample of 964 Danish Gazelle firms. Our study finds a clear positive relationship between growth and subsequent profitability among gazelle firms. Moreover, this relationship is stronger for firms pursuing a broad market strategy rather than a focus or niche strategy. An important managerial implication is that the growth strategy should be clearly integrated with the general strategic orientation of the firm.

Bicep femoris voluntary activation deficits contribute to eccentric knee flexor weakness following intermittent running

INTRODUCTION: Hamstring strain injuries (HSI) are the predominant non-contact injury in many sports. Eccentric hamstring muscle weakness following intermittent running has been implicated within the aetiology of HSI. This weakness following intermittent running is often greater eccentrically than concentrically, however the cause of this unique, contraction mode specific phenomenon is unknown. AIM: To determine if this preferential eccentric decline in strength is caused by declines in voluntary hamstring muscle activation. METHODS: Fifteen recreationally active males completed 18 × 20m overground sprints. Maximal strength (concentric and eccentric knee flexor and concentric knee extensor) was determined isokinetically at the velocities of ±1800.s-1 and ±600.s- while hamstring muscle activation was assessed using surface electromyography, before and 15 minutes after the running protocol. RESULTS: Overground intermittent running caused greater eccentric (27.2 Nm; 95% CI = 11.2 to 43.3; p=0.0001) than concentric knee flexor weakness (9.3 Nm; 95% CI = -6.7 to 25.3; P=0.6361). Following the overground running, voluntary activation levels of the lateral hamstrings showed a significant decline (0.08%; 95% CI = 0.045 to 0.120; P<0.0001). In comparison, medial hamstring activation showed no change following intermittent running. CONCLUSION: Eccentric hamstring strength is decreased significantly following intermittent overground running. Voluntary activation deficits in the biceps femoris muscle are responsible for some portion of this weakness. The implications of this finding are significant because the biceps femoris muscle is the most frequently strained of all the hamstring muscles and because fatigue appears to play an important part in injury occurrence.

The impact of fatigue caused by sprint running on the median power frequency of the hamstring muscle group

Hamstring strain injuries (HSI) are the predominant non-contact injury in many sports. Intermittent running has been shown to result in preferential reductions in eccentric hamstring strength, which increase the risk of sustaining a HSI. The eccentric specific nature of this decline in hamstring function implicates central mechanisms, as peripheral fatigue mechanisms tend to impact upon both concentric and eccentric contractions modes. However, neural function of the hamstrings, such as the median power frequency (MPF) of the surface electromyography signal has yet to be examined in the fatigued hamstring following intermittent sprint running. AIM: To determine the impact of fatigue induced by intermittent sprinting on the MPF of the medial and lateral hamstring muscles. METHODS: Fifteen recreationally active males completed 18 × 20m overground sprints. Maximal strength (concentric and eccentric knee flexor and concentric knee extensor) was determined isokinetically at the velocities of ±180.s-1 and ±60.s- while hamstring muscle activation was assessed using surface electromyography, before and 15 minutes after the running protocol. RESULTS: Overground intermittent running caused a significant reduction in eccentric knee flexor strength (27.2 Nm; 95% CI = 11.2 to 43.3; p=0.0001) but not concentric strength (9.3 Nm; 95% CI = -6.7 to 25.3; P=0.6361). Following the overground running, MPF of the lateral hamstrings showed a significant decline eccentrically (0.86; 95% CI = 0.59 to 1.54; P=0.038) and concentrically (0.76; 95%CI = 0.66 to 0.83; P=0.039). Similar declines in MPF were also noted in the medial hamstrings eccentrically (1.54; 95% CI = 0.59 to 7.9; P=0.005) and concentrically (1.18; 95% CI = 0.44 to 6.8; P=0.040). CONCLUSION: Whilst sprint running induced fatigue led to a eccentric specific reduction in knee flexor torque, MPF was suppressed across both contraction modes. This would indicate that factors associated with the decline in MPF do not appear to explain the contraction mode-specific loss of strength after intermittent sprints. This would implicate other central mechanisms, such as declines in voluntary activation, in explaining the eccentric specific decline in strength seen following sprint running.

Do voluntary activation deficits contribute to eccentric weakness following intermittent running?

Hamstring strain injuries (HSI) are the predominant non-contact injury in many sports. Eccentric hamstring muscle weakness following intermittent running has been implicated within the aetiology of HSI. This weakness following intermittent running is often greater eccentrically than concentrically, however the cause of this unique, contraction mode specific phenomenon is unknown. PURPOSE: To determine if this preferential eccentric decline in strength is caused by declines in voluntary hamstring muscle activation. METHODS: Fifteen recreationally active males completed 18 × 20m overground sprints. Maximal strength (concentric and eccentric knee flexor and concentric knee extensor) was determined isokinetically at the velocities of ±1800.s-1 and ±600.s- while hamstring muscle activation was assessed using surface electromyography, before and 15 minutes after the running protocol. RESULTS: Overground intermittent running caused greater eccentric (27.2 Nm; 95% CI = 11.2 to 43.3; p=0.0001) than concentric knee flexor weakness (9.3 Nm; 95% CI = -6.7 to 25.3; P=0.6361). Following the overground running, voluntary activation levels of the lateral hamstrings showed a significant decline (0.08%; 95% CI = 0.045 to 0.120; P<0.0001). In comparison, medial hamstring activation showed an increased level of activation following intermittent running (0.12%; 95% CI = 0.049 to 0.030; P = 0.0102). CONCLUSION: Eccentric hamstring strength is decreased significantly following intermittent overground running. Voluntary activation deficits in the biceps femoris muscle are responsible for some portion of this weakness. The implications of this finding are significant because the biceps femoris muscle is the most frequently strained of all the hamstring muscles and because fatigue appears to play an important part in injury occurrence.

Declines in eccentric knee flexor weakness following repeat sprint running are related to declines in biceps femoris voluntary activation

Introduction: Hamstring strain injuries (HSI) are the predominant non-contact injury in many sports. Eccentric hamstring muscle weakness following intermittent running has been implicated within the aetiology of HSI. This weakness following intermittent running is sometimes greater eccentrically than concentrically, however the cause of this unique, contraction mode specific phenomenon is unknown. The purpose of this research was to determine whether declines in knee flexor strength following overground repeat sprints are caused by declines in voluntary activation of the hamstring muscles. Methods: Seventeen recreationally active males completed 3 sets of 6 by 20m overground sprints. Maximal isokinetic concentric and eccentric knee flexor and concentric knee extensor strength was determined at ±1800.s-1 and ±600.s-1 while hamstring muscle activation was assessed using surface electromyography, before and 15 minutes after the running protocol. Results: Overground repeat sprint running resulted in a significant decline in eccentric knee flexor strength (31.1 Nm; 95% CI = 21.8 to 40.3 Nm; p < 0.001). However, concentric knee flexor strength was not significantly altered (11.1 Nm; 95% CI= -2.8 to 24.9; p=0.2294). Biceps femoris voluntary activation levels displayed a significant decline eccentrically (0.067; 95% CI=0.002 to 0.063; p=0.0325). However, there was no significant decline concentrically (0.025; 95% CI=-0.018 to 0.043; p=0.4243) following sprinting. Furthermore, declines in average peak torque at -1800.s-1 could be explained by changes in hamstring activation (R2 = 0.70). Moreover, it was change in the lateral hamstring muscle activity that was related to the decrease in knee flexor torque (p = 0.0144). In comparison, medial hamstring voluntary activation showed no change for either eccentric (0.06; 95% CI = -0.033 to 0.102; p=0.298) or concentric (0.09; 95% CI = -0.03 to 0.16; p=0.298) muscle actions following repeat sprinting. Discussion: Eccentric hamstring strength is decreased significantly following overground repeat sprinting. Voluntary activation deficits in the biceps femoris muscle explain a large portion of this weakness. The implications of these findings are significant as the biceps femoris muscle is the most frequently strained of the knee flexors and fatigue is implicated in the aetiology of this injury.

Twitter as a technology for audiencing and fandom : the #Eurovision phenomenon

Amongst the most prominent uses of Twitter at present is its role in the discussion of widely televised events: Twitter’s own statistics for 2011, for example, list major entertainment spectacles (the MTV Music Awards, the BET Awards) and sports matches (the UEFA Champions League final, the FIFA Women’s World Cup final) amongst the events generating the most tweets per second during the year (Twitter, 2011). User activities during such televised events constitute a specific, unique category of Twitter use, which differs clearly from the other major events which generate a high rate of tweets per second (such as crises and breaking news, from the Japanese earthquake and tsunami to the death of Steve Jobs), as preliminary research has shown. During such major media events, by contrast, Twitter is used most predominantly as a technology of fandom instead: it serves in the first place as a backchannel to television and other streaming audiovisual media, enabling users offer their own running commentary on the universally shared media text of the event broadcast as it unfolds live. Centrally, this communion of fans around the shared text is facilitated by the use of Twitter hashtags – unifying textual markers which are now often promoted to prospective audiences by the broadcasters well in advance of the live event itself. This paper examines the use of Twitter as a technology for the expression of shared fandom in the context of a major, internationally televised annual media event: the Eurovision Song Contest. It constitutes a highly publicised, highly choreographed media spectacle whose eventual outcomes are unknown ahead of time and attracts a diverse international audience. Our analysis draws on comprehensive datasets for the ‘official’ event hashtags, #eurovision, #esc, and #sbseurovision. Using innovative methods which combine qualitative and quantitative approaches to the analysis of Twitter datasets containing several hundreds of thousands, we examine overall patterns of participation to discover how audiences express their fandom throughout the event. Minute-by-minute tracking of Twitter activity during the live broadcasts enables us to identify the most resonant moments during each event; we also examine the networks of interaction between participants to detect thematically or geographically determined clusters of interaction, and to identify the most visible and influential participants in each network. Such analysis is able to provide a unique insight into the use of Twitter as a technology for fandom and for what in cultural studies research is called ‘audiencing’: the public performance of belonging to the distributed audience for a shared media event. Our work thus contributes to the examination of fandom practices led by Henry Jenkins (2006) and other scholars, and points to Twitter as an important new medium facilitating the connection and communion of such fans.

The product of text and 'Other' statements : discourse analysis and the critical use of Foucault

Much has been written on Michel Foucault’s reluctance to clearly delineate a research method, particularly with respect to genealogy (Harwood 2000; Meadmore, Hatcher, & McWilliam 2000; Tamboukou 1999). Foucault (1994, p. 288) himself disliked prescription stating, “I take care not to dictate how things should be” and wrote provocatively to disrupt equilibrium and certainty, so that “all those who speak for others or to others” no longer know what to do. It is doubtful, however, that Foucault ever intended for researchers to be stricken by that malaise to the point of being unwilling to make an intellectual commitment to methodological possibilities. Taking criticism of “Foucauldian” discourse analysis as a convenient point of departure to discuss the objectives of poststructural analyses of language, this paper develops what might be called a discursive analytic; a methodological plan to approach the analysis of discourses through the location of statements that function with constitutive effects.