Learning task specific distributed paragraph representations using a 2-tier convolutional neural network

We introduce a type of 2-tier convolutional neural network model for learning distributed paragraph representations for a special task (e.g. paragraph or short document level sentiment analysis and text topic categorization). We decompose the paragraph semantics into 3 cascaded constitutes: word representation, sentence composition and document composition. Specifically, we learn distributed word representations by a continuous bag-of-words model from a large unstructured text corpus. Then, using these word representations as pre-trained vectors, distributed task specific sentence representations are learned from a sentence level corpus with task-specific labels by the first tier of our model. Using these sentence representations as distributed paragraph representation vectors, distributed paragraph representations are learned from a paragraph-level corpus by the second tier of our model. It is evaluated on DBpedia ontology classification dataset and Amazon review dataset. Empirical results show the effectiveness of our proposed learning model for generating distributed paragraph representations.

Ingroup and outgroup minorities:differential impact upon public and private responses

The experiment which is presented in this paper was designed to overcome some of the problems associated with previous research investigating the effects of social categorization and minority influence. Sixty-eight fourteen-year-old British Secondary School pupils indicated their attitudes towards a 'grant for pupils' before and after reading a text which advocated a minority position. The text was attributed as being the work of either pupils from their own school (ingroup minority) or from a school they discriminated against (outgroup minority). Responses were either made in ‘public’ (by telling subjects that other pupils would see their responses) or in ‘private’ (by subjects putting their responses into a ‘ballot box’). The results showed that on public responses ingroup minorities had more influence than outgroup minorities while there was no difference on private responses. Also, greater change occurred when responses were made in private than in public. These results are compatible with the intergroup analysis of minority influence.

How weak categorizers based upon different principles strengthen performance

Combining the results of classifiers has shown much promise in machine learning generally. However, published work on combining text categorizers suggests that, for this particular application, improvements in performance are hard to attain. Explorative research using a simple voting system is presented and discussed in the light of a probabilistic model that was originally developed for safety critical software. It was found that typical categorization approaches produce predictions which are too similar for combining them to be effective since they tend to fail on the same records. Further experiments using two less orthodox categorizers are also presented which suggest that combining text categorizers can be successful, provided the essential element of ‘difference’ is considered.

Peripheral vision and pattern recognition:a review

We summarize the various strands of research on peripheral vision and relate them to theories of form perception. After a historical overview, we describe quantifications of the cortical magnification hypothesis, including an extension of Schwartz's cortical mapping function. The merits of this concept are considered across a wide range of psychophysical tasks, followed by a discussion of its limitations and the need for non-spatial scaling. We also review the eccentricity dependence of other low-level functions including reaction time, temporal resolution, and spatial summation, as well as perimetric methods. A central topic is then the recognition of characters in peripheral vision, both at low and high levels of contrast, and the impact of surrounding contours known as crowding. We demonstrate how Bouma's law, specifying the critical distance for the onset of crowding, can be stated in terms of the retinocortical mapping. The recognition of more complex stimuli, like textures, faces, and scenes, reveals a substantial impact of mid-level vision and cognitive factors. We further consider eccentricity-dependent limitations of learning, both at the level of perceptual learning and pattern category learning. Generic limitations of extrafoveal vision are observed for the latter in categorization tasks involving multiple stimulus classes. Finally, models of peripheral form vision are discussed. We report that peripheral vision is limited with regard to pattern categorization by a distinctly lower representational complexity and processing speed. Taken together, the limitations of cognitive processing in peripheral vision appear to be as significant as those imposed on low-level functions and by way of crowding.

Categorically speaking:reflecting AEC/FM participants' requirements in Groupware

Groupware allows participants in an AEC/FM project to timely share relevant electronic project information at near real-time speed. One challenge of existing groupware tools is the varying information categorization needs of different AEC/FM project participants within the same groupware environment. This paper introduces initial results of a survey of several AEC/FM companies' project information categorizations. These initial results reveal differences in the ways AEC/FM project participants organize project information within each company. These differences will result in the need for a comprehensive information categorization solution when companies move their organized company specific project information to a project-wide repository in groupware. In the future, as a result of this research, a comprehensive project categorization scheme will be introduced which would allow participant-specific views of the AEC/FM project information to be rendered.

Microarray analysis of expression of cell death-associated genes in spinal cord cells with cyclic tensile strain

Previous studies have described alterations in gene expression following spinal cord injury, but this response to mechanical stimuli is difficult to investigate in vivo. Therefore, we have investigated the effect of cyclic tensile strain on cultured spinal cord cells from E15 Sprague-Dawley rats. Microarray analysis of gene expression and categorization of identified genes were performed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) systems. The application of cyclic tensile strain reduced the viability of cultured spinal cord cells significantly in a dose- and time-dependent manner. GO analysis identified candidate genes related to apoptosis (44) and to response to stimulus (17). KEGG analysis identified changes in the expression levels of 12 genes of the mitogen-activated protein kinase (MAPK) signaling pathway, which were confirmed to be upregulated and validated by RT-PCR analysis. Spinal cord cells undergo cell death in response to cyclic tensile strain, which were dose- and time-dependent, with upregulation of various genes, in particular of the MAPK pathway.

Exploring types and functions of questions in police interviews

There is a proliferation of categorization schemes in the scientific literature that have mostly been developed from psychologists’ understanding of the nature of linguistic interactions. This has a led to problems in defining question types used by interviewers. Based on the principle that the overarching purpose of an interview is to elicit information and that questions can function both as actions in their own right and as vehicles for other actions, a Conversational Analysis approach was used to analyse a small number of police interviews. The analysis produced a different categorization of question types and, in particular, the conversational turns fell into two functional types: (i) Topic Initiation Questions and (ii) Topic Facilitation Questions. We argue that forensic interviewing requires a switch of focus from the ‘words’ used by interviewers in question types to the ‘function’ of conversational turns within interviews.

Harnessing demographic differences in organizations:what moderates the effects of workplace diversity?

To account for the double-edged nature of demographic workplace diversity (i.e,. relational demography, work group diversity, and organizational diversity) effects on social integration, performance, and well-being related variables, research has moved away from simple main effect approaches and started examining variables that moderate these effects. While there is no shortage of primary studies of the conditions under which diversity leads to positive or negative outcomes, it remains unclear which contingency factors make it work. Using the Categorization-Elaboration Model as our theoretical lens, we review variables moderating the effects of workplace diversity on social integration, performance, and well-being outcomes, focusing on factors that organizations and managers have control over (i.e., strategy, unit design, human resource, leadership, climate/culture, and individual differences). We point out avenues for future research and conclude with practical implications.

Effectiveness of performance appraisal:an integrated framework

Based on a robust analysis of the existing literature on performance appraisal (PA), this paper makes a case for an integrated framework of effectiveness of performance appraisal (EPA). To achieve this, it draws on the expanded view of measurement criteria of EPA, i.e. purposefulness, fairness and accuracy, and identifies their relationships with ratee reactions. The analysis reveals that the expanded view of purposefulness includes more theoretical anchors for the purposes of PA and relates to various aspects of human resource functions, e.g. feedback and goal orientation. The expansion in the PA fairness criterion suggests certain newly established nomological networks, which were ignored in the past, e.g. the relationship between distributive fairness and organization-referenced outcomes. Further, refinements in PA accuracy reveal a more comprehensive categorization of rating biases. Coherence among measurement criteria has resulted in a ratee reactions-based integrated framework, which should be useful for both researchers and practitioners.

Dynamic Bayesian network for semantic place classification in mobile robotics

In this paper, the problem of semantic place categorization in mobile robotics is addressed by considering a time-based probabilistic approach called dynamic Bayesian mixture model (DBMM), which is an improved variation of the dynamic Bayesian network. More specifically, multi-class semantic classification is performed by a DBMM composed of a mixture of heterogeneous base classifiers, using geometrical features computed from 2D laserscanner data, where the sensor is mounted on-board a moving robot operating indoors. Besides its capability to combine different probabilistic classifiers, the DBMM approach also incorporates time-based (dynamic) inferences in the form of previous class-conditional probabilities and priors. Extensive experiments were carried out on publicly available benchmark datasets, highlighting the influence of the number of time-slices and the effect of additive smoothing on the classification performance of the proposed approach. Reported results, under different scenarios and conditions, show the effectiveness and competitive performance of the DBMM.