Using Web Search Logs to Identify Query Classification Terms

Purpose – The work presented in this paper aims to provide an approach to classifying web logs by personal properties of users. Design/methodology/approach – The authors describe an iterative system that begins with a small set of manually labeled terms, which are used to label queries from the log. A set of background knowledge related to these labeled queries is acquired by combining web search results on these queries. This background set is used to obtain many terms that are related to the classification task. The system then ranks each of the related terms, choosing those that most fit the personal properties of the users. These terms are then used to begin the next iteration. Findings – The authors identify the difficulties of classifying web logs, by approaching this problem from a machine learning perspective. By applying the approach developed, the authors are able to show that many queries in a large query log can be classified. Research limitations/implications – Testing results in this type of classification work is difficult, as the true personal properties of web users are unknown. Evaluation of the classification results in terms of the comparison of classified queries to well known age-related sites is a direction that is currently being exploring. Practical implications – This research is background work that can be incorporated in search engines or other web-based applications, to help marketing companies and advertisers. Originality/value – This research enhances the current state of knowledge in short-text classification and query log learning. Classification schemes, Computer networks, Information retrieval, Man-machine systems, User interfaces

Quo Vadis? Reliable and Practical Rule Extraction from Neural Networks

Rule extraction from neural network algorithms have been investigated for two decades and there have been significant applications. Despite this level of success, rule extraction from neural network methods are generally not part of data mining tools, and a significant commercial breakthrough may still be some time away. This paper briefly reviews the state-of-the-art and points to some of the obstacles, namely a lack of evaluation techniques in experiments and larger benchmark data sets. A significant new development is the view that rule extraction from neural networks is an interactive process which actively involves the user. This leads to the application of assessment and evaluation techniques from information retrieval which may lead to a range of new methods.

Improved facial expression recognition via uni-hyperplane classification

Large margin learning approaches, such as support vector machines (SVM), have been successfully applied to numerous classification tasks, especially for automatic facial expression recognition. The risk of such approaches however, is their sensitivity to large margin losses due to the influence from noisy training examples and outliers which is a common problem in the area of affective computing (i.e., manual coding at the frame level is tedious so coarse labels are normally assigned). In this paper, we leverage the relaxation of the parallel-hyperplanes constraint and propose the use of modified correlation filters (MCF). The MCF is similar in spirit to SVMs and correlation filters, but with the key difference of optimizing only a single hyperplane. We demonstrate the superiority of MCF over current techniques on a battery of experiments.

Providing a law degree for the real world : perspective of an Australian law school

As a result of a two year curriculum review, QUT’s undergraduate law degree has a focus on first year student transition, integration of law graduate capabilities throughout the degree and work integrated learning. A ‘whole-degree’ approach was adopted to ensure that capabilities were appropriately embedded and scaffolded throughout the degree, that teaching and learning approaches met the needs of students as they transitioned from first year through to final year, and that students in final year were provided with a capstone experience to assist them with transition into the work place. The revised degree commenced implementation in 2009. This paper focuses on the ‘real world’ approach to the degree achieved through the first year program, embedding and scaffolding law graduate capabilities through authentic and valid assessment and work integrated learning to assist graduates with transition into the workplace.

Semantic service discovery in the service ecosystem

Electronic services are a leitmotif in ‘hot’ topics like Software as a Service, Service Oriented Architecture (SOA), Service oriented Computing, Cloud Computing, application markets and smart devices. We propose to consider these in what has been termed the Service Ecosystem (SES). The SES encompasses all levels of electronic services and their interaction, with human consumption and initiation on its periphery in much the same way the ‘Web’ describes a plethora of technologies that eventuate to connect information and expose it to humans. Presently, the SES is heterogeneous, fragmented and confined to semi-closed systems. A key issue hampering the emergence of an integrated SES is Service Discovery (SD). A SES will be dynamic with areas of structured and unstructured information within which service providers and ‘lay’ human consumers interact; until now the two are disjointed, e.g., SOA-enabled organisations, industries and domains are choreographed by domain experts or ‘hard-wired’ to smart device application markets and web applications. In a SES, services are accessible, comparable and exchangeable to human consumers closing the gap to the providers. This requires a new SD with which humans can discover services transparently and effectively without special knowledge or training. We propose two modes of discovery, directed search following an agenda and explorative search, which speculatively expands knowledge of an area of interest by means of categories. Inspired by conceptual space theory from cognitive science, we propose to implement the modes of discovery using concepts to map a lay consumer’s service need to terminologically sophisticated descriptions of services. To this end, we reframe SD as an information retrieval task on the information attached to services, such as, descriptions, reviews, documentation and web sites - the Service Information Shadow. The Semantic Space model transforms the shadow's unstructured semantic information into a geometric, concept-like representation. We introduce an improved and extended Semantic Space including categorization calling it the Semantic Service Discovery model. We evaluate our model with a highly relevant, service related corpus simulating a Service Information Shadow including manually constructed complex service agendas, as well as manual groupings of services. We compare our model against state-of-the-art information retrieval systems and clustering algorithms. By means of an extensive series of empirical evaluations, we establish optimal parameter settings for the semantic space model. The evaluations demonstrate the model’s effectiveness for SD in terms of retrieval precision over state-of-the-art information retrieval models (directed search) and the meaningful, automatic categorization of service related information, which shows potential to form the basis of a useful, cognitively motivated map of the SES for exploratory search.

A toolbox for animal call recognition

Monitoring the natural environment is increasingly important as habit degradation and climate change reduce theworld’s biodiversity.We have developed software tools and applications to assist ecologists with the collection and analysis of acoustic data at large spatial and temporal scales.One of our key objectives is automated animal call recognition, and our approach has three novel attributes. First, we work with raw environmental audio, contaminated by noise and artefacts and containing calls that vary greatly in volume depending on the animal’s proximity to the microphone. Second, initial experimentation suggested that no single recognizer could dealwith the enormous variety of calls. Therefore, we developed a toolbox of generic recognizers to extract invariant features for each call type. Third, many species are cryptic and offer little data with which to train a recognizer. Many popular machine learning methods require large volumes of training and validation data and considerable time and expertise to prepare. Consequently we adopt bootstrap techniques that can be initiated with little data and refined subsequently. In this paper, we describe our recognition tools and present results for real ecological problems.

Leadership as dance : a consideration of the applicability of the ‘mother’ of all arts as the basis for establishing connoisseurship

This article follows the lead of several researchers who claim there is an urgent need to utilize insights from the arts, aesthetics and the humanities to expand our understanding of leadership. It endeavours to do this by exploring the metaphor of dance. It begins by critiquing current policy metaphors used in the leadership literature that present a narrow and functional view of leadership. It presents and discusses a conceptual model of leadership as dance that incorporates key dimensions such as context, dance and music and includes Polyani’s concept of connoisseurship. This article identifies some of the tensions that are inherent in both notions of dance and leadership. The final part of the article discusses the implications the model raises for broadening our understanding of leadership and school leadership preparation programmes. Three core implications raised here are (i) making space for alternative metaphors in leadership preparation programmes; (ii) providing opportunities to students of leadership to understand through alternative learning approaches and (iii) providing opportunities for engagement in alternative research agendas.

Human–computer interaction : the impact of users’ cognitive styles on query reformulation behaviour during web searching

This paper discusses users’ query reformulation behaviour while searching information on the Web. Query reformulations have emerged as an important component of Web search behaviour and human-computer interaction (HCI) because a user’s success of information retrieval (IR) depends on how he or she formulates queries. There are various factors, such as cognitive styles, that influence users’ query reformulation behaviour. Understanding how users with different cognitive styles formulate their queries while performing Web searches can help HCI researchers and information systems (IS) developers to provide assistance to the users. This paper aims to examine the effects of users’ cognitive styles on their query reformation behaviour. To achieve the goal of the study, a user study was conducted in which a total of 3613 search terms and 872 search queries were submitted by 50 users who engaged in 150 scenario-based search tasks. Riding’s (1991) Cognitive Style Analysis (CSA) test was used to assess users’ cognitive style as wholist or analytic, and verbaliser or imager. The study findings show that users’ query reformulation behaviour is affected by their cognitive styles. The results reveal that analytic users tended to prefer Add queries while all other users preferred New queries. A significant difference was found among wholists and analytics in the manner they performed Remove query reformulations. Future HCI researchers and IS developers can utilize the study results to develop interactive and user-cantered search model, and to provide context-based query suggestions for users.

Teaching creatively, teaching for creativity : QUT’s Creative Industries Faculty Showcase Panel

In this panel, we showcase approaches to teaching for creativity in disciplines of the Media, Entertainment and Creative Arts School and the School of Design within the Creative Industries Faculty (CIF) at QUT. The Faculty is enormously diverse, with 4,000 students enrolled across a total of 20 disciplines. Creativity is a unifying concept in CIF, both as a graduate attribute, and as a key pedagogic principle. We take as our point of departure the assertion that it is not sufficient to assume that students of tertiary courses in creative disciplines are ‘naturally’ creative. Rather, teachers in higher education must embrace their roles as facilitators of development and learning for the creative workforce, including working to build creative capacity (Howkins, 2009). In so doing, we move away from Renaissance notions of creativity as an individual genius, a disposition or attribute which cannot be learned, towards a 21st century conceptualisation of creativity as highly collaborative, rhizomatic, and able to be developed through educational experiences (see, for instance, Robinson, 2006; Craft; 2001; McWilliam & Dawson, 2008). It has always been important for practitioners of the arts and design to be creative. Under the national innovation agenda (Bradley et al, 2008) and creative industries policy (e.g., Department for Culture, Media and Sport, 2008; Office for the Arts, 2011), creativity has been identified as a key determinant of economic growth, and thus developing students’ creativity has now become core higher education business across all fields. Even within the arts and design, professionals are challenged to be creative in new ways, for new purposes, in different contexts, and using new digital tools and platforms. Teachers in creative disciplines may have much to offer to the rest of the higher education sector, in terms of designing and modelling innovative and best practice pedagogies for the development of student creative capability. Information and Communication Technologies such as mobile learning, game-based learning, collaborative online learning tools and immersive learning environments offer new avenues for creative learning, although analogue approaches may also have much to offer, and should not be discarded out of hand. Each panelist will present a case study of their own approach to teaching for creativity, and will address the following questions with respect to their case: 1. What conceptual view of creativity does the case reflect? 2. What pedagogical approaches are used, and why were these chosen? What are the roles of innovative learning approaches, including ICTs, if any? 3. How is creativity measured or assessed? How do students demonstrate creativity? We seek to identify commonalities and contrasts between and among the pedagogic case studies, and to answer the question: what can we learn about teaching creatively and teaching for creativity from CIF best practice?

An English-translated parallel corpus for the CJK Wikipedia collections

In this paper, we describe a machine-translated parallel English corpus for the NTCIR Chinese, Japanese and Korean (CJK) Wikipedia collections. This document collection is named CJK2E Wikipedia XML corpus. The corpus could be used by the information retrieval research community and knowledge sharing in Wikipedia in many ways; for example, this corpus could be used for experimentations in cross-lingual information retrieval, cross-lingual link discovery, or omni-lingual information retrieval research. Furthermore, the translated CJK articles could be used to further expand the current coverage of the English Wikipedia.

How activation, entanglement, and searching a semantic network contribute to event memory

Free association norms indicate that words are organized into semantic/associative neighborhoods within a larger network of words and links that bind the net together. We present evidence indicating that memory for a recent word event can depend on implicitly and simultaneously activating related words in its neighborhood. Processing a word during encoding primes its network representation as a function of the density of the links in its neighborhood. Such priming increases recall and recognition and can have long lasting effects when the word is processed in working memory. Evidence for this phenomenon is reviewed in extralist cuing, primed free association, intralist cuing, and single-item recognition tasks. The findings also show that when a related word is presented to cue the recall of a studied word, the cue activates it in an array of related words that distract and reduce the probability of its selection. The activation of the semantic network produces priming benefits during encoding and search costs during retrieval. In extralist cuing recall is a negative function of cue-to-distracter strength and a positive function of neighborhood density, cue-to-target strength, and target-to cue strength. We show how four measures derived from the network can be combined and used to predict memory performance. These measures play different roles in different tasks indicating that the contribution of the semantic network varies with the context provided by the task. We evaluate spreading activation and quantum-like entanglement explanations for the priming effect produced by neighborhood density.

Context-aware device self-configuration using self-organizing maps

Modern mobile computing devices are versatile, but bring the burden of constant settings adjustment according to the current conditions of the environment. While until today, this task has to be accomplished by the human user, the variety of sensors usually deployed in such a handset provides enough data for autonomous self-configuration by a learning, adaptive system. However, this data is not fully available at certain points in time, or can contain false values. Handling potentially incomplete sensor data to detect context changes without a semantic layer represents a scientific challenge which we address with our approach. A novel machine learning technique is presented - the Missing-Values-SOM - which solves this problem by predicting setting adjustments based on context information. Our method is centered around a self-organizing map, extending it to provide a means of handling missing values. We demonstrate the performance of our approach on mobile context snapshots, as well as on classical machine learning datasets.

Semantic labelling for document feature patterns using ontological subjects

Finding and labelling semantic features patterns of documents in a large, spatial corpus is a challenging problem. Text documents have characteristics that make semantic labelling difficult; the rapidly increasing volume of online documents makes a bottleneck in finding meaningful textual patterns. Aiming to deal with these issues, we propose an unsupervised documnent labelling approach based on semantic content and feature patterns. A world ontology with extensive topic coverage is exploited to supply controlled, structured subjects for labelling. An algorithm is also introduced to reduce dimensionality based on the study of ontological structure. The proposed approach was promisingly evaluated by compared with typical machine learning methods including SVMs, Rocchio, and kNN.

Environmental education through listening to children

Environmental education centres contribute to schools and communities in Environmental Education and Education for Sustainability through nature and urban -based, experiential learning and action learning approaches. An underlying assumption of these centres is that intensive, short-term, outdoor/environmental education experiences can change key attitudes and/or actions leading to positive environmental behaviour. This study reflects the interests of a researching professional who investigated aspects of a program that he designed and implemented as principal of an environmental education centre. Most evaluations of similar programs have used quasi-experimental designs to measure the program outcomes. However, this study considered the experiences of the program from the perspectives of a group of key stakeholders often overlooked in the literature; the children who participated in the program. This study examined children’s accounts of their own experiences in order to contribute new understandings of children’s perspectives and how they can be considered when designing and implementing environmental education programs. This research drew on key theoretical assumptions derived from the sociology of childhood. Within sociology of childhood, children are considered to be competent practitioners within their social worlds, who, through their talk and interaction, participate actively in the construction of their own social situations. This approach also views children as capable and competent learners who construct their knowledge through everyday participation in social experiences. This study set out to generate children’s own accounts of their experiences of a five day residential program at the Centre. In total, 54 children participated in the study that used a multi-faceted data collection approach that included conversations, drawings, photographs and journal writing. Using content analysis, data were analysed by means of an inductive approach to develop themes related to the children’s perspectives of their experiences. Three interrelated and co-dependent components of the experience emerged from the analysis; space and place; engagement and participation; and responsiveness and reflection. These components co-exist and construct the conditions for effective experiences in environmental education at the Centre. The first key finding was the emphasis that the children placed on being provided with somewhere where they could feel safe and comfortable to interact with their environment and engage in a range of outdoor experiences. The children identified that place was an outdoor classroom where they could participate in first-hand experiences and, at times, explore out-of-bound spaces; that is, a place where they had previously been limited, often by adults, in their opportunities to interact with nature. A second key finding was the emphasis that the children placed on engagement and participation in environmental experience. The children described participating in a range of new primary experiences that involved first-hand, experiences and also described participating in collaborative experiences that involved interacting with peers and with teachers, who appeared to behave differently to how they behaved at school. Finally, the children described a different type of interactional relationship with teachers, comparing the active educational role they played on camp to a more passive role at school where they sat at a table and the teacher wrote on the board. The final key finding was the emphasis that the children placed on responsiveness and reflection in the experience. In responding to their experiences, the children described the fun and excitement, confidence and satisfaction that they gained from the experience. The children also identified how their experiences contributed to the development of a caring-for-nature attitude and the value of a disorienting dilemma in promoting responsiveness to the environment. This disorienting dilemma was an event that caused the children to reassess their own beliefs and attitudes. From the three main findings, a theoretical framework that represented the children’s accounts of their experiences and a pedagogic approach that respected their accounts was developed. This pedagogic approach showed how a disorienting dilemma could create a disequilibrium in relation to a child’s existing ideas and experiences. As a result, children were challenged to reflect upon their existing environmental beliefs and practices. The findings of this research have implications for the field of environmental education. Adopting sociology of childhood provides an alternative foundation to research and can present a deeper understanding of what children believe, than an approach that relies solely on using scientific methods to undercover and analyse these understandings. This research demonstrates the value of gaining children’s accounts to assist educators to design environmental education programs as it can offer more than adult and educator perspectives. This study also provides understandings of environmental education practice by describing how the children engaged with informal learning situations. Finally, two sets of recommendations, drawn from this study, are made. The first set considers nine recommendations about and for future research and the second relates to redesigning of the environmental educational program at the research site, with six recommendations made.

A pattern based two-stage text classifier

In a classification problem typically we face two challenging issues, the diverse characteristic of negative documents and sometimes a lot of negative documents that are closed to positive documents. Therefore, it is hard for a single classifier to clearly classify incoming documents into classes. This paper proposes a novel gradual problem solving to create a two-stage classifier. The first stage identifies reliable negatives (negative documents with weak positive characteristics). It concentrates on minimizing the number of false negative documents (recall-oriented). We use Rocchio, an existing recall based classifier, for this stage. The second stage is a precision-oriented “fine tuning”, concentrates on minimizing the number of false positive documents by applying pattern (a statistical phrase) mining techniques. In this stage a pattern-based scoring is followed by threshold setting (thresholding). Experiment shows that our statistical phrase based two-stage classifier is promising.