A tag-based personalized item recommendation system using tensor modeling and topic model approaches

This research falls in the area of enhancing the quality of tag-based item recommendation systems. It aims to achieve this by employing a multi-dimensional user profile approach and by analyzing the semantic aspects of tags. Tag-based recommender systems have two characteristics that need to be carefully studied in order to build a reliable system. Firstly, the multi-dimensional correlation, called as tag assignment , should be appropriately modelled in order to create the user profiles [1]. Secondly, the semantics behind the tags should be considered properly as the flexibility with their design can cause semantic problems such as synonymy and polysemy [2]. This research proposes to address these two challenges for building a tag-based item recommendation system by employing tensor modeling as the multi-dimensional user profile approach, and the topic model as the semantic analysis approach. The first objective is to optimize the tensor model reconstruction and to improve the model performance in generating quality rec-ommendation. A novel Tensor-based Recommendation using Probabilistic Ranking (TRPR) method [3] has been developed. Results show this method to be scalable for large datasets and outperforming the benchmarking methods in terms of accuracy. The memory efficient loop implements the n-mode block-striped (matrix) product for tensor reconstruction as an approximation of the initial tensor. The probabilistic ranking calculates the probabil-ity of users to select candidate items using their tag preference list based on the entries generated from the reconstructed tensor. The second objective is to analyse the tag semantics and utilize the outcome in building the tensor model. This research proposes to investigate the problem using topic model approach to keep the tags nature as the “social vocabulary” [4]. For the tag assignment data, topics can be generated from the occurrences of tags given for an item. However there is only limited amount of tags availa-ble to represent items as collection of topics, since an item might have only been tagged by using several tags. Consequently, the generated topics might not able to represent the items appropriately. Furthermore, given that each tag can belong to any topics with various probability scores, the occurrence of tags cannot simply be mapped by the topics to build the tensor model. A standard weighting technique will not appropriately calculate the value of tagging activity since it will define the context of an item using a tag instead of a topic.

A case study of utilising concept knowledge in a topic specific document collection

The use of ‘topic’ concepts has shown improved search performance, given a query, by bringing together relevant documents which use different terms to describe a higher level concept. In this paper, we propose a method for discovering and utilizing concepts in indexing and search for a domain specific document collection being utilized in industry. This approach differs from others in that we only collect focused concepts to build the concept space and that instead of turning a user’s query into a concept based query, we experiment with different techniques of combining the original query with a concept query. We apply the proposed approach to a real-world document collection and the results show that in this scenario the use of concept knowledge at index and search can improve the relevancy of results.

When sexuality is in a research topic! The methodological challenges in sexuality and street healing research in Bangladesh

This paper is a bridge between two studies by the author: (i) completed MA research; and (ii) on-going PhD research, on male sexual health and the street healing system in Bangladesh. Street healing, a traditional healing system in Bangladesh, is at the centre of the studies. This is a popular form of folk healing in Bangladesh, where male impotency is a central issue. The author has been researching street healing to understand male sexual health-seeking behaviour in Bangladesh. In this paper, the author brings in experiences from his MA research to explore the challenges of studying sexuality and street healing in Bangladesh and concludes by describing his plan to address those issues in his on-going PhD research.

On the topic of competency and excellence of our graduate engineers

Problem solving is an essential element of civil engineering education. It has been I observed that students are best able to understand civil engineering theory when there is a ' practical application of it. Teaching theory alone has led to lower levels of comprehension and motivation and a correspondingly higher rate of failure and "drop-out". This paper analyses the effectiveness of introducing practical design projects at an early stage within a civil engineering undergraduate program at Queensland University of Technology. In two of the essential basic subjects, Engineering Mechanics and Steel Structures, model projects which simulate realistic engineering exercises were introduced. Students were required to work in small groups to analyse, design and build the lightest I most efficient model bridges made of specific materials such as spaghetti, drinking straw, paddle pop sticks and balsa wood and steel columns for a given design loading/target capacity. The paper traces the success of the teaching strategy at each stage from its introduction through to the final student and staff evaluation.

Pattern-based topic modelling and its application for information filtering and information retrieval

This thesis targets on a challenging issue that is to enhance users' experience over massive and overloaded web information. The novel pattern-based topic model proposed in this thesis can generate high-quality multi-topic user interest models technically by incorporating statistical topic modelling and pattern mining. We have successfully applied the pattern-based topic model to both fields of information filtering and information retrieval. The success of the proposed model in finding the most relevant information to users mainly comes from its precisely semantic representations to represent documents and also accurate classification of the topics at both document level and collection level.

ACE Research Vignette 045: Are “Entrepreneurial Opportunities” a meaningful topic for entrepreneurship research?

This series of research vignettes is aimed at sharing current and interesting research findings from our team of international entrepreneurship researchers. In this vignette, Professor Per Davidsson discusses research on “entrepreneurial opportunities”. A “Government Health Warning” is in place for this particular vignette: it mainly concerns matters internal to entrepreneurship research; however, reflective practitioners may find it to be of interest.

Anatomy of a trending topic : Methods for the visualisation of retweet chains [Anatomie eines Trending Topics: Methoden zur Visualisierung von Retweet-Ketten]

Twitter and other social networking sites play an ever more present role in the spread of current events. The dynamics of information dissemination through digital network structures are still relatively unexplored, however. At what time an issue is taken up by whom? Who forwards a message when to whom else? What role do individual communication participants, existing digital communities or the technical foundations of each network platform play in the spread of news? In this chapter we discuss, using the example of a video on a current sociopolitical issue in Australia that was shared on Twitter, a number of new methods for the dynamic visualisation and analysis of communication processes. Our method combines temporal and spatial analytical approaches and provides new insights into the spread of news in digital networks. [Social media dienen immer häufger als Disseminationsmechanismen für Medieninhalte. Auf Twitter ermöglicht besonders die Retweet-Funktion den schnellen und weitläufgen Transfer von Nachrichten. In diesem Beitrag etablieren neue methodische Ansätze zur Erfassung, Visualisierung und Analyse von Retweet-Ketten. Insbesondere heben wir hervor, wie bestehende Netzwerkanalysemethoden ergänzt werden können, um den Ablauf der Weiterleitung sowohl temporal als auch spatial zu erfassen. Unsere Fallstudie demonstriert die verbreitung des videoclips einer am 9. Oktober 2012 spontan gehaltenen Wutrede der australischen Premierministerin Julia Gillard, in der sie Oppositionsführer Tony Abbott als Frauenhasser brandmarkte. Durch die Erfassung von Hintergrunddaten zu den jeweiligen NutzerInnen, die sich an der Weiterleitung des Videoclips beteiligten, erstellen wir ein detailliertes Bild des Disseminationsablaufs im vorliegenden Fall. So lassen sich die wichtigsten AkteurInnen und der Ablauf der Weiterleitung darstellen und analysieren. Daraus entstehen Einblicke in die allgemeinen verbreitungsmuster von Nachrichten auf Twitter].

Semantic Classes in Topic Detection and Tracking

Topic detection and tracking (TDT) is an area of information retrieval research the focus of which revolves around news events. The problems TDT deals with relate to segmenting news text into cohesive stories, detecting something new, previously unreported, tracking the development of a previously reported event, and grouping together news that discuss the same event. The performance of the traditional information retrieval techniques based on full-text similarity has remained inadequate for online production systems. It has been difficult to make the distinction between same and similar events. In this work, we explore ways of representing and comparing news documents in order to detect new events and track their development. First, however, we put forward a conceptual analysis of the notions of topic and event. The purpose is to clarify the terminology and align it with the process of news-making and the tradition of story-telling. Second, we present a framework for document similarity that is based on semantic classes, i.e., groups of words with similar meaning. We adopt people, organizations, and locations as semantic classes in addition to general terms. As each semantic class can be assigned its own similarity measure, document similarity can make use of ontologies, e.g., geographical taxonomies. The documents are compared class-wise, and the outcome is a weighted combination of class-wise similarities. Third, we incorporate temporal information into document similarity. We formalize the natural language temporal expressions occurring in the text, and use them to anchor the rest of the terms onto the time-line. Upon comparing documents for event-based similarity, we look not only at matching terms, but also how near their anchors are on the time-line. Fourth, we experiment with an adaptive variant of the semantic class similarity system. The news reflect changes in the real world, and in order to keep up, the system has to change its behavior based on the contents of the news stream. We put forward two strategies for rebuilding the topic representations and report experiment results. We run experiments with three annotated TDT corpora. The use of semantic classes increased the effectiveness of topic tracking by 10-30\% depending on the experimental setup. The gain in spotting new events remained lower, around 3-4\%. The anchoring the text to a time-line based on the temporal expressions gave a further 10\% increase the effectiveness of topic tracking. The gains in detecting new events, again, remained smaller. The adaptive systems did not improve the tracking results.

The Hot Topic Community: Videoconferencing to reduce the professional isolation of teachers in rural Western Australia

Access to quality professional learning and the opportunity to collaborate with other educators can be limited for teachers in rural and remote areas of Western Australia. A recognised need to enhance the skills of rural teaching professionals and encourage teachers in small communities to join collegial networks was established by the members of several professional organisations. A working group consisting of representatives from the Australian College of Educators-WA (ACE-WA), the Rural and Remote Education Advisory Council (RREAC), the Society for the Provision of Education in Rural Australia (SPERA) and the School of Isolated and Distance Education (SIDE) provided teachers in rural areas with the opportunity to reduce professional isolation through the provision of relevant, convenient, and cost effective in-service education. Through a videoconferencing system, accessed within the Western Australian Telecentre Network and other educational organisations, the audience connected and participated with the presenter and studio based audience for two Hot Topics Seminars in 2008. This paper reports on the challenges and successes encountered by the working group and the findings of the research conducted throughout 2008.

Efficient classification using phrases generated by topic models

There are many popular models available for classification of documents like Naïve Bayes Classifier, k-Nearest Neighbors and Support Vector Machine. In all these cases, the representation is based on the “Bag of words” model. This model doesn't capture the actual semantic meaning of a word in a particular document. Semantics are better captured by proximity of words and their occurrence in the document. We propose a new “Bag of Phrases” model to capture this discriminative power of phrases for text classification. We present a novel algorithm to extract phrases from the corpus using the well known topic model, Latent Dirichlet Allocation(LDA), and to integrate them in vector space model for classification. Experiments show a better performance of classifiers with the new Bag of Phrases model against related representation models.

Testable models of aquatic systems: A NERC special topic for integrating experiments and modelling

A major part of the support for fundamental research on aquatic ecosystems continues to be provided by the Natural Environment Research Council (NERC). Funds are released for ”thematic” studies in a selected special topic or programme. ”Testable Models of Aquatic Ecosystems” was a Special Topic of the NERC, initiated in 1995, the aim of which was to promote ecological modelling by making new links between experimental aquatic biologists and state-of-the-art modellers. The Topic covered both marine and freshwater systems. This paper summarises projects on aspects of the responses of individual organisms to the effects of environmental variability, on the assembly, permanence and resilience of communities, and on aspects of spatial models. The authors conclude that the NERC Special Topic has been highly successful in promoting the development and application of models, most particularly through the interplay between experimental ecologists and formal modellers.

Topic 1. Some steric-height features of the water off the Pacific coast of Canada; Topic 2. Possible warming of the water at station P in the northeast Pacific during 1956-1984 [abstract]

TOPIC 1: In terms of seasonal scale, temperature effect dominates the annual change of steric height in the open ocean whereas salinity effect controls it along the continental shelf. Large portion of the annual change of height relative to the 1000-db surface is contained in the upper 100m layer. However, in interannual scale large anomalies of steric height in the open ocean, are more often than not, caused by halosteric rather than thermosteric effect. At least in the open ocean the heights are almost totally determined by the behavior of deep water. Their interannual variability appears to be related to the cumulative effect of Eckman pumping. TOPIC 2: There is a "trend" that over the past 28 years the water at Station P has warmed. Least-square analysis indicates that this warming may be significant but shortening of the time-series data by approximately 10 years fails to show that this is the case. These "trends" have to be interpreted with care. The warming may be "apparent" in that it is not indicated clearly in the deep isopynal surfaces which, during the above period, have deepened. Thus warming at the isobaric surfaces may be the effect of the downward migration of the isopynal surfaces.

Evaluation of the economic costs and benefits of methods for reducing nutrient loads to the Gulf of Mexico: Topic 6 Report for the Integrated Assessment on Hypoxia in the Gulf of Mexico.

In this report we analyze the Topic 5 report’s recommendations for reducing nitrogen losses to the Gulf of Mexico (Mitsch et al. 1999). We indicate the relative costs and cost-effectiveness of different control measures, and potential benefits within the Mississippi River Basin. For major nonpoint sources, such as agriculture, we examine both national and basin costs and benefits. Based on the Topic 2 economic analysis (Diaz and Solow 1999), the direct measurable dollar benefits to Gulf fisheries of reducing nitrogen loads from the Mississippi River Basin are very limited at best. Although restoring the ecological communities in the Gulf may be significant over the long term, we do not currently have information available to estimate the benefits of such measures to restore the Gulf’s long-term health. For these reasons, we assume that measures to reduce nitrogen losses to the Gulf will ultimately prove beneficial, and we concentrate on analyzing the cost-effectiveness of alternative reduction strategies. We recognize that important public decisions are seldom made on the basis of strict benefit–cost analysis, especially when complete benefits cannot be estimated. We look at different approaches and different levels of these approaches to identify those that are cost-effective and those that have limited undesirable secondary effects, such as reduced exports, which may result in lost market share. We concentrate on the measures highlighted in the Topic 5 report, and also are guided by the source identification information in the Topic 3 report (Goolsby et al. 1999). Nonpoint sources that are responsible for the bulk of the nitrogen receive most of our attention. We consider restrictions on nitrogen fertilizer levels, and restoration of wetlands and riparian buffers for denitrification. We also examine giving more emphasis to nitrogen control in regions contributing a greater share of the nitrogen load.