Finding short patterns to classify text documents

Many classification methods have been proposed to find patterns in text documents. However, according to Occam's razor principle, "the explanation of any phenomenon should make as few assumptions as possible", short patterns usually have more explainable and meaningful for classifying text documents. In this paper, we propose a depth-first pattern generation algorithm, which can find out short patterns from text document more effectively, comparing with breadth-first algorithm

Reducing performance bias for unbalanced text mining

In text categorization applications, class imbalance, which refers to an uneven data distribution where one class is represented by far more less instances than the others, is a commonly encountered problem. In such a situation, conventional classifiers tend to have a strong performance bias, which results in high accuracy rate on the majority class but very low rate on the minorities. An extreme strategy for unbalanced, learning is to discard the majority instances and apply one-class classification to the minority class. However, this could easily cause another type of bias, which increases the accuracy rate on minorities by sacrificing the majorities. This paper aims to investigate approaches that reduce these two types of performance bias and improve the reliability of discovered classification rules. Experimental results show that the inexact field learning method and parameter optimized one-class classifiers achieve more balanced performance than the standard approaches.

Translating text into motion : performance analysis for singers and directors

The study proposes a method of interpretation, which gives the contemporary singer-actor and director a method of performance analysis considering both the visual and the auditory aspects of vocal performance. The analysis begins with an analysis of the structure, which maps out the key ideas and translates them into main states of mind. The text analysis is followed by an analysis of the musical treatment of the text, also scanning for extra dramatic meaning inherent in vocal line and musical accompaniment. The so isolated dramatic ideas in text and musical structure and content are finally discussed in terms of their physical aspects, using the system of gesture as practiced by actors and singers before, during and beyond the Baroque period. The method is exemplified by the obbligato recitative "In quali eccessi" from Donna Elvira’s scene in the second act of the opera "Don Giovanni" by L. Da Ponte and W. A. Mozart. The musical example includes gestural notation and a table gestural notation symbols.

A virtual environment for cultural and linguistic inclusivity : teaching in a global environment

This paper is about a project that uses Information and communication technologies in a virtual environment where students can communicate in their own language in text and audio.

Case study of electronic networking in an initial teacher education course

Electronic networking ('computer-mediated communication1), considered to be ‘unique domain for educational activity’ (Harasim, 1989:50) and ‘new educational paradigm’ (Mason & Kaye, 1989:23), has been widely used and researched in K-12 schooling, place-based undergraduate subjects and distance education courses. However, only a limited number of reports of usage with experienced teachers (professional development), beginning teachers (induction support) and trainee teachers (initial training) have been published. Hence, little is known about the ways in which this new medium might contribute to the acquisition and maintenance of professional knowledge in the field of teacher education. The purpose of this study was to document an application of electronic networking in an initial 'school-based1 teacher education course. Three factors which were considered to be important in the adoption of electronic networking were specifically addressed: (a) the potential of the medium to attract and maintain a representative and comprehensive audience', (b) the willingness of participants to use the medium for the notation of ideas about teaching; and, (c) the extent to which reflection on practice was evident in network messages. This study also identified and investigated other effects which emerged as participants attempted to negotiate personal relationships with new technology. A case study was selected to investigate audience, notation, reflection, and other effects, in a particular application. Data were collected using participant observation, software-generated statistics, printed documentation, university records, questionnaires, interviews and content analysis of messages. These data were used to describe and analyse network participation by trainee teachers, classroom teachers and university staff. The data revealed that an audience did exist on the electronic network but that this was not comprehensive. Teachers had difficulty accessing the network because of other school commitments, access to equipment and personal competence with microcomputers. These difficulties indicated that developing and maintaining the teacher audience may be a major problem with electronic networking in initial teacher education. This case study revealed that deeply held concerns about notation of ideas by trainee teachers and classroom teachers can be powerful reasons for limited network participation. For trainee teachers, recording ideas publicly presented special difficulties associated with written communication. They were concerned about writing for an audience; about what to write about and how to write it. The loss of visual and verbal cues which form part of face-to-face communication was also a problem leading to concerns about how messages would be received by others. However, the overwhelming concern of almost all trainee teachers about presenting their own ideas was Tear of criticism' from peers (in particular), and other participants on the network. Trainee teachers expressed concerns about the 'dangers' of putting their thoughts in writing, the scrutiny their messages might have received from others, and the public 'criticism' about what they wrote which might have appeared on the network. Knowing that messages were stored on the network, and could be retrieved at some later date, heightened anxiety about the vulnerability of written communication; what was written on one occasion may have to be defended at some later date when the views expressed initially were no longer held. Classroom teachers were also unsure about recording their own ideas in an electronic form. Like trainee teachers, they were concerned about the scrutiny their contributions might receive from other users, and the lack of visual and verbal cues which they had learnt to use in face-to-face communication. Notating ideas in text-based messages which were archived (by the software), and retrievable by others later, was also daunting to many teachers. Another major 'danger' for teachers was the possible repercussions of 'public comment' about curriculum policy and initiatives which they thought might get them into 'trouble' with their employer. Since very few messages were contributed to conferences, there was little evidence of reflection in network communication. In the main, the network was not used to share information and ideas about curriculum and teaching. Public examples of collaboration between participants were not evident, and the 'special knowledge' held by members in each distinct group of users was not elaborated and discussed. Messages were not used to request information or clarification about issues, to outline the processes by which decisions about teaching were reached, or to synthesis ideas from different sources. The potential of the medium to operationalise reflective practice was not realised. Among the effects observed, the use of an anonymous account to access the network, and the impact this had on participation (in one conference) was considered to be a particularly significant finding. While the opportunity to systematically investigate the effects of anonymity on network participation and message contributions was not realised (by the author) while the research was in progress, the effects observed and discussed are considered to be important and worthy of further investigation. In this case study, the anonymous account helped trainee teachers mask concerns about personal writing skills and fear of criticism from others, indicating that anonymity may alter communication patterns, particularly in the early stages of network use. Given the data collected in this case study, and the interpretations placed on it by the author, a pessimistic assessment of the place of electronic networking in initial teacher education courses was presented. For this situation to change - that is, for electronic discussions to become more fully integrated into course activities - four issues which need to be addressed were identified and discussed. These included clarification of the role of collaboration amongst participants in initial teacher education, the ways in which collaboration can be facilitated using electronic networking, the problems of notation - such as the difficulty of expressing ideas about teaching in written form, and the concerns about permanently archived messages - for teachers and trainee teachers, and the lack of skills which many trainee teachers bring to electronic discussions. In the context of initial teacher education, it was suggested that these four aspects require clarification and development before the potential advantages of electronic networking can be realised. Some specific suggestions about how these issues might be resolved were presented.

Black and white : in search of an ‘apt’ response to Indigenous writing

‘The good editor,’ suggests Thomas McCormack in his Fiction Editor, the Novel and the Novelist, ‘reads, and … responds aptly’ to the writer’s work, ‘where “aptly” means “as the ideal appropriate reader would”.’ McCormack develops an argument that encompasses the dual ideas of sensibility and craft as essential characteristics of the fiction editor. But at an historical juncture that has seen increasing interest in the publication of Indigenous writing, and when Indigenous writers themselves may envisage a multiplicity of readers (writing, for instance, for family and community, and to educate a wider white audience), who is the ‘ideal appropriate reader’ for the literary works of the current generation of Australian Indigenous writers? And what should the work of this ‘good editor’ be when engaging with the text of an Indigenous writer? This paper examines such questions using the work of Margaret McDonell and Jennifer Jones, among others, to explore ways in which non-Indigenous editors may apply aspects of McCormack’s ‘apt response’ to the editing of Indigenous texts.

Identifying annotations for adventure game generation from fiction text

Recent advancements in Text-to-Scene research have lead to the development of systems which automatically extract key concepts from the text of a fiction book and generate computer animated movies depicting the story. Extracting such annotations from raw fiction text is a laborious process and so in this work we evaluate appropriate candidates to serve as the basis for the required annotations for generating interactive virtual worlds.

We validate our choice by generating adventure games: inter-active virtual worlds which create a stylized representation of the environment described in the text, populate it with characters related to the story and define game goals related to the plot of the fiction story. Our prototype produces a fully playable game, making use of an existing open-source game engine.

The process is evaluated using user tests in which participants are asked to measure the accuracy with which the game represents the events, characters and goals described in the story. The response indicates that the chosen annotation set is sufficient to define a game that is a plausibly acceptable representation of the text.

The interaction of discipline and culture in academic writing

Work in contrastive rhetoric has often sought to examine the impact of culturally-based writing conventions on text production and has outlined cultural differences in texts in different languages. At the same time. the study of specialised languages has often claimed a degree of uniformity in text construction both at the level of culture and at the level of the discipline. It appears however that approaches which consider just culture or just discipline miss part of the picture. This paper argues that considerations of discipline and culture are. complex and interrelated and that this complexity and interrelationship can be seen at several different levels in specialised academic texts.

Emotional reactions to real-world events in social networks

A convergence of emotions among people in social networks is potentially resulted by the occurrence of an unprecedented event in real world. E.g., a majority of bloggers would react angrily at the September 11 terrorist attacks. Based on this observation, we introduce a sentiment index, computed from the current mood tags in a collection of blog posts utilizing an affective lexicon, potentially revealing subtle events discussed in the blogosphere. We then develop a method for extracting events based on this index and its distribution. Our second contribution is establishment of a new bursty structure in text streams termed a sentiment burst. We employ a stochastic model to detect bursty periods of moods and the events associated. Our results on a dataset of more than 12 million mood-tagged blog posts over a 4-year period have shown that our sentiment-based bursty events are indeed meaningful, in several ways.

Prose fiction’s infinite playlist : encountering music in fiction narrative

This paper will argue that reference to music in young-adult prose fiction stimulates movement across narrative and artistic boundaries in ways that facilitate a unique reading encounter. The inclusion of musical reference opens up a space for a multisensory experience that is beyond that of the reading experience devoid of musical association, even when the audio is not immediately available at the time of reading. This experience is bound to the role of the reader, however, be it through the remembered or imagined experience of the music that is signaled in-text, or even the reader’s pursuit of the audio in response to the reading. As ‘a threshold literature’ (Eaton 2010, np) that targets a young audience for whom ‘popular music is globally acknowledged as affectively and culturally central’ (Bloustien & Peters 2011, 4), young-adult fiction is an apt space for explorations into the potential that exists when a text includes musical reference. In particular, Gerard Genette’s paratexts (1997), J Hillis Miller’s ‘membranes’ (2005) and T Austin Graham’s ‘literary soundtrack’ (2009) will be used to examine how Rachel Cohn and David Levithan’s young-adult fiction novel Nick and Norah’s infinite playlist (2006) functions as an ‘infinite playlist’ in itself via a series of paratextual and epitextual elements. Discussion of the latticework of music-narrative interaction that exists as a part of this text will facilitate an understanding of how musical reference can encourage movement within and beyond the narrative towards a potentially unique reading experience.

A spectroscopy of texts for effective clustering

For many clustering algorithms, such as k-means, EM, and CLOPE, there is usually a requirement to set some parameters. Often, these parameters directly or indirectly control the number of clusters to return. In the presence of different data characteristics and analysis contexts, it is often difficult for the user to estimate the number of clusters in the data set. This is especially true in text collections such as Web documents, images or biological data. The fundamental question this paper addresses is: ldquoHow can we effectively estimate the natural number of clusters in a given text collection?rdquo. We propose to use spectral analysis, which analyzes the eigenvalues (not eigenvectors) of the collection, as the solution to the above. We first present the relationship between a text collection and its underlying spectra. We then show how the answer to this question enhances the clustering process. Finally, we conclude with empirical results and related work.

A question of genre : de-mystifying the exegesis

In this paper I wish to draw together two on-oing debates which are being played out, at least in part, in TEXT. These debates concern the nature of research for the discipline of Creative Writing and those dealing with problems associated with awarding and assessing research higher degrees - MAs, MA(Hons), MPhils and PhDs. In particular I wish to suggest that the combination of these two debates sheds considerable light on the nature and role of the research higher degree's exegesis which accompanies the creative product.

Proficiency and collaborative learning

This study reports on the effect of different levels of proficiency on the students' achievements in collaborative learning instruction among 30 Persian-speaking EFL college students. Having been divided into dyads with different levels of proficiency, these subjects participated in nine sessions of collaborative instruction based on the form focused collaborative text-editing task. The results of this study are obtained through the analyses of the proficiency gains in formal aspects of language in text-editing pre and post-tests and text-reconstruction post-test. The study highlights the significant effect of this type of instruction on more proficient students. Reasons provided in this study are the language analytic ability of the students, learning styles and strategies, and motivation. The findings of this study are helpful to teachers of the classes with heterogeneous language proficiency.

A bayesian nonparametric joint factor model for learning shared and individual subspaces from multiple data sources

Joint analysis of multiple data sources is becoming increasingly popular in transfer learning, multi-task learning and cross-domain data mining. One promising approach to model the data jointly is through learning the shared and individual factor subspaces. However, performance of this approach depends on the subspace dimensionalities and the level of sharing needs to be specified a priori. To this end, we propose a nonparametric joint factor analysis framework for modeling multiple related data sources. Our model utilizes the hierarchical beta process as a nonparametric prior to automatically infer the number of shared and individual factors. For posterior inference, we provide a Gibbs sampling scheme using auxiliary variables. The effectiveness of the proposed framework is validated through its application on two real world problems - transfer learning in text and image retrieval.

Zeroes of Riemann's zeta function on the critical line with 20000 decimal digits accuracy

The first 12000 zeroes of Riemann's zeta function on the critical line with 20000 decimal digits accuracy. Format: the zeroes are in text file listed consecutively in decimal representation, each zero starts on a new line.

Zeroes of zeta function presented in this file were calculated on MASSIVE cluster (www.massive.org.au) using Python and packages MPmath version 0.17 and gmpy version 2.1, with a Newton based algorithm proposed by Fredrik Johansson with precision set to 20000 decimal digits. Partial recalculation with higher precision didn't show any loss of accuracy so we expect that the values are correct up to, possibly, a few last digits. We express our thanks to Fredrik Johansson for this algorithm and for development of MPmath as well.