EXEHDA-HM: A compositional approach to explore contextual information on hybrid models

The current proposals of hybrid context modeling bring new challenges, an important one is how applications can access and process data stored on these models. Thinking about that, this paper proposes a solution to deal with this challenge through a compositional approach that explores the context information on hybrid models, called EXEHDA-HM. The proposed approach stands out by the design of a repository that supports three database models and by the compositional processing strategy based on rules. In our proposal, the applications can combine data stored on different bases in a single rule, which could enhance the identification of contextual situations. For the evaluation we designed and implemented some case studies on information security area, exploring the hybrid repository composed of relational, non-relational, and triple storage models. Our results demonstrate that was possible to identify richer situations with the data composition across more than one model and there are situations that can only be found through this composition.     

Recognizing machines with parametric and nonparametric learning methods using contextual information

Pattern recognition systems using information from the pattern which follows the present pattern are discussed. Parametric learning methods in the supervised and unsupervised machines are proposed and compare favorably with conventional methods. Furthermore, a semilinear machine with a nonparametric learning method is considered. The results of computer experiments with artificially generated data and with handprinted alphanumeric characters are given to show that the approach we adopt is quite useful for recognition of Markovian patterns.     

The affective outcomes of using influence tactics in embodied conversational agents

In this study, we highlight the theoretical underpinnings of human impression management tactics and link them to current research in embodied conversational agents. Specifically, we incorporated impression management behaviors into an embodied conversational agent in order to show that different influence strategies affect user perceptions, and how those perceptions might be moderated by user gender. We programmed the agent to use two human impression management techniques (ingratiation and self-promotion) and had the agent interact with 88 users. After the interaction, users reported their perceptions of the system’s power, trustworthiness, expertise, and attractiveness. The impression management techniques altered users’ perceptions and these perceptions were moderated by gender differences.     

Generalization of recommender systems: Collaborative filtering extended to groups of users and restricted to groups of items

In this paper we present a collaborative filtering method which opens up the possibilities of traditional collaborative filtering in two aspects: (1) it enables joint recommendations to groups of users and (2) it enables the recommendations to be restricted to items similar to a set of reference items. By way of example, a group of four friends could request joint recommendations of films similar to “Avatar” or “Titanic”. In the paper, using experiments, we show that the traditional approach of collaborative filtering does not satisfactorily resolve the new possibilities contemplated; we also provide a detailed formulation of the method proposed and an extensive set of experiments and comparative results which show the superiority of designed collaborative filtering compared to traditional collaborative filtering in: (a) number of recommendations obtained, (b) quality of the predictions, (c) quality of the recommendations. The experiments have been carried out on the databases Movielens and Netflix.     

The category FGrp of fuzzy groups is not equivalent to the category Grp of classical groups

In this paper, the concept of strong homomorphism of fuzzy group in the sense of [⁴] is given. The fundamental homomorphism theorem of fuzzy group, in the sense of strong homomorphism, is also obtained. We built a category FGrp of fuzzy groups and strong homomorphism and show that category FGrp and the category Grp of group and homomorphism are not equivalent. These discussions show that fuzzy group is different from classical group.     

A new and practical influence measure for subsets of covariance matrix sample principal components with applications to high dimensional datasets

Principal Component Analysis (PCA) is an important tool in multivariate analysis, in particular when faced with high dimensional data. There has been much done with regard to sensitivity analysis and the development of influence diagnostics for the eigenvector estimators that define the sample principal components. However, little, if any, has been done in this setting with regard to the sample principal components themselves. In this paper we develop a sensitivity measure for principal components associated with the covariance matrix that is very much related to the influence function (Hampel, 1974). This influence measure is based on the average squared canonical correlation and differs from the existing measures in that it assesses the influence of certain observational types on the sample principal components. We use this measure to derive an influence diagnostic that satisfies two key criteria being (i) it detects influential observations with respect to subsets of sample principal components and (ii) is efficient to calculate even in high dimensions. We use several microarray datasets to show that our measure satisfies both criteria.     

Examining learning in relation to the contexts of use of ICT

Abstract Although ICT resources are commonly expected to produce uniform benefits, they are necessarily employed within pre-existing contexts of educational and social activity, and the outcome in terms of both pattern of use and learning depends on how they fit in with these. As a result, the same technology or software may have unexpectedly diverse effects, according to specific setting. If the object is to exercise control over outcome, then the conditions of use need to be planned for within the design and implementation of the technology. In order to do this, it is crucial that research gathers data on how outcomes are affected by the interplay between technology and context. This raises questions about the methods that would be appropriate for the conduct and dissemination of such research. These points are discussed in relation to three studies, one each at primary, secondary and university levels of education.     

An approach to measuring influence and cognitive similarity in computer-mediated communication

While many explanations of influence have been proposed there is still debate over which is correct even though most are supported by empirical evidence. This uncertainty has been attributed to there being too little evidence of real-world influence networks, and an inability to separate influence from cognitive similarity, that is, a pre-existing like-mindedness, attitude or way of thinking shared among participants. This paper proposes theme resonance, a new metric for measuring both influence and cognitive similarity between and among participants in the same online conversation. Theme resonance is derived from two textual content analysis systems: Centering Resonance Analysis and qualitative thematic modeling. The use of theme resonance is demonstrated by constructing influence networks using online conversations in ten weblogs, allowing the propagation of new conversational themes to be traced from initiator though subsequent propagators. A method of separating influence from like-mindedness is also demonstrated. Depending on the metric chosen influence and its susceptibility were found both to be opposite ends of the same spectrum, and distinct attributes. In either case the majority of blog participants are close to the low end of each characteristic. However, those at the higher ends are shown to be easily and distinctly identified.     

Maximizing three-hop influence spread in social networks using discrete comprehensive learning artificial bee colony optimizer

Aiming at resolving the influence maximization (IM) problem in social networks, this paper proposes a three-layer-comprehensive-influence evaluation (TLCIE) model to measure the spread range of combinational nodes in the independent or weighted cascade models. The TLCIE is an enhanced three-hop influence spread model by integrating the intra- and inter-layer’s propagation effect, which improves the accuracy of propagation simulation and the reliability of parameter estimation. Then, an adaptive discrete artificial bee colony algorithm (ADABC) is devised to resolve the TLCIE model efficiently. In ADABC, the comprehensive-learning guided (CLG) updating rules, the degree-improvement initialization method and the semi-abandonment scout bee strategy are incorporated to enhance the search ability. Finally, the proposed model and algorithm are tested on a set of real-world social network instances, and the experimental results validate their effectiveness and efficiency.     

The loss in efficiency from using grouped data to estimate coefficients of group level variables

We derive the efficiency loss from using grouped data to estimate coefficients of variables that vary across groups but not individuals within a group (e.g., state unemployment rates) when micro data are unavailable on the dependent variable. We present an empirical example of our theoretical results, and show that the efficiency loss in this application is small. JEL Classification Number C10.     

OptAssign—A web-based tool for assigning students to groups

A frequent task in educational settings is to assign students to groups based on stated preferences for some projects or topics. This paper introduces a web-based tool supporting both the work flow of collecting student data and the group formation. The latter is based on finding optimal solutions to suitable mathematical assignment problems, allowing for a number of constraints regarding size and structure of the groups. Evaluation results show advantages compared to manual procedures in terms of time savings for lecturers, and higher fairness and correctness as perceived by students.     

The influence of disposition and social ties on trust in new virtual teammates

With the increased presence of social media tools such as LinkedIn and Facebook, social network information is now commonplace. Social media websites prominently display the social distance or so-called “degrees of separation” among users, effectively allowing people to view their shared social ties with others, including prospective teammates they have not met. Through the presentation and manipulation of social network information, this longitudinal experiment investigated whether dispositional and relational variables contribute to “swift trust” among new virtual teammates. Data from 74 participants were collected to test a path analytic model predicting that social ties and propensity to trust influence perceptions of a new teammate’s trustworthiness (ability, benevolence, and integrity) as well as the willingness to trust that new teammate when given the opportunity to do so. Path analysis indicated good model fit, but showed no significant evidence that social ties or propensity to trust affect perceived trustworthiness at the initial point of team engagement. Additionally, only one component of perceived trustworthiness (perceived ability) and propensity to trust were found to predict trusting behavior towards a new, unknown, teammate.     

Electronic [re]constitution of groups: Group dynamics from face-to-face to an online setting

The authors work as online tutors for a BSc (Hons) physiotherapy programme at Coventry University in the United Kingdom. This paper represents a stage in our developing understanding, over a 3 year period, of the impact of group dynamics on online interaction among physiotherapy students engaged in sharing with their peers their first experiences of clinical practice. The literature exploring online interaction tends to situate meaning either in theories borrowed from conventional face-to-face interaction or on virtual interaction. Research focusing on ‘blended learning’ that combines face-to-face and online interaction is limited in terms of considering how group dynamics impact groups that are constituted and reconstituted in the two very different learning contexts. Using a case study approach, the authors consider how group dynamics change as groups move from face-to-face to online collaboration in pursuit of learning objectives. We characterize typical features of the cases and draw conclusions based on similarities and differences. Findings suggest that group learning is linked to group cohesion, which appears to be mediated by social and cognitive factors that students bring with them. Social presence appears vital to positive group dynamics and is a precursor to cognitive presence, which develops when groups rise above their desire to be sociable and supportive. Group dynamics, whether positive or negative, and their consequent impact on interaction appear to be relatively stable across contexts once the group scene is set through face-to-face interaction. Engagement and interaction of individual students, however, can alter when face-to-face interaction moves online.     

Manipulating user resistance to large-scale information systems through influence tactics

This study examines how influence tactics can be used to manipulate user resistance and subsequent usage behaviors in the context of large-scale, newly implemented information systems (IS). To do so, we integrate the theory of interpersonal influence and leadership, adaptive structuration theory, and the literature on user resistance and behavioral IS theories, and develop a comprehensive theoretical model anchored in (a) social influence tactics, (b) IT artifact perceptions, and (c) user resistance and subsequent behaviors. The resultant theoretical model explains how influence tactics modulate user perceptions, user resistance to the system, and ultimately the unfaithful use of the system. This model is tested in the context of an electronic medical record (EMR) system implementation, with a sample of 442 clinicians who were exposed to various influence tactics and with the use of SEM techniques. The results indicate that certain influence tactics are beneficial in reducing user resistance while others can unintentionally increase resistance. Resistance, in turn, drives adaptive user behaviors in the form of unfaithful use of the system. Thus, it is possible to manipulate resistance and resultant adaptive behaviors through influence tactics, without necessarily altering the IT artifact. This is particularly important, as many off-the-shelf IT artifacts cannot be substantially modified.     

Personalizing influence diagrams: applying online learning strategies to dialogue management

We consider the problem of adapting the parameters of an influence diagram in an online fashion for real-time personalization. This problem is important when we use the influence diagram repeatedly to make decisions and we are uncertain about its parameters. We describe learning algorithms to solve this problem. In particular, we show how to modify various explore-versus-exploit strategies that are known to work well for Markov decision processes to the more general influence-diagram model. As an illustration, we describe how our techniques for online personalization allow a voice-enabled browser to adapt to a particular speaker for spoken dialogue management. We evaluate all the explore-versus-exploit strategies in this domain.     

Alternative approaches to include exogenous variables in DEA measures: A comparison using Monte Carlo

The theory for measuring efficiency of producers has developed alternative approaches to correct for the effect of non-discretionary variables in the analysis. A review of different options in the specific literature of Data envelopment analysis (DEA) allows us to identify three main approaches: one-stage, two-stage and multi-stage models. Recently, some of these models have been improved through the development of bootstrap methods making it possible to make inference and to avoid bias in the estimation of efficiency scores. The aim of this paper is to test the performance of these recent models and to compare among them using simulated data from a Monte Carlo experimental design.     

DIANA: A computer-supported heterogeneous grouping system for teachers to conduct successful small learning groups

Teachers interested in small-group learning can benefit from using psychological factors to create heterogeneous groups. In this paper we describe a computer-supported grouping system named DIANA that uses genetic algorithms to achieve fairness, equity, flexibility, and easy implementation. Grouping was performed so as to avoid the creation of exceptionally weak groups. We tested DIANA with 66 undergraduate computer science students assigned to groups of three either randomly (10 groups) or using an algorithm reflecting [Sternberg, R. J. (1994). Thinking styles: theory and assessment at the interface between intelligence and personality. In R. J. Sterberg, & P. Ruzgis (Eds.), Personality and Intelligence (pp. 169–187). New York: Cambridge University Press.] three thinking styles (12 groups). The results indicate that: (a) the algorithm-determined groups were more capable of completing whatever they were “required to do” at a statistically significant level, (b) both groups were equally capable of solving approximately 80% of what they “chose to do,” and (c) the algorithm-determined groups had smaller inter-group variation in performance. Levels of satisfaction with fellow group member attitudes, the cooperative process, and group outcomes were also higher among members of the algorithm-determined groups. Suggestions for applying computer-supported group composition systems are offered.     

Probabilistic measures of efficiency and the influence of contextual variables in nonparametric production models: an application to agricultural research in Brazil

In a research institution it is important to identify which management practices have influence on production efficiency. In this paper we assess the statistical significance of contextual variables type, size, financial resources acquisition, intensity of partnerships, processes improvements and management change. The analysis is carried out for the Brazilian Agricultural Research Corporation over the period 1999–2006. The statistical analysis uses a balanced dynamic panel data model. We conclude that only financial resources acquisition is statistically significant. The association with the production process is positive. We also found the two lag inertial components of the conditional FDH to unconditional FDH ratio statistically significant, indicating a 2‐year effort to improve efficiency.     

Providing developmental feedback to individuals from different ethnic minority groups using expert systems

Abstract: The paper is an account of a commercially successful and viable expert system entitled FRED (Feedback Reporter Evaluating Development). Expert systems offer a consistent approach to providing feedback to individuals that people with personal biases and emotions may find difficult to achieve. The text used in the expert system is sensitive and carefully worded by a team of skilled professionals, many of whom were from ethnic and cultural minorities. This is especially important when dealing with internal applicants from ethnic minorities as feedback given is consistent, fair, non-judgemental and devoid of possible ethnic prejudices and emotional statements. FRED offers a sensitive, comprehensive and cost-effective method of producing feedback reports for individuals from an ethnic minority and the paper details FRED's design, use and evaluation in an organizational setting. Future research is recommended in utilizing expert systems to provide feedback in assessment and development settings.