Query expansion using term relationships in language models for information retrieval

Language Modeling (LM) has been successfully applied to Information Retrieval (IR). However, most of the existing LM approaches only rely on term occurrences in documents, queries and document collections. In traditional unigram based models, terms (or words) are usually considered to be independent. In some recent studies, dependence models have been proposed to incorporate term relationships into LM, so that links can be created between words in the same sentence, and term relationships (e.g. synonymy) can be used to expand the document model. In this study, we further extend this family of dependence models in the following two ways: (1) Term relationships are used to expand query model instead of document model, so that query expansion process can be naturally implemented; (2) We exploit more sophisticated inferential relationships extracted with Information Flow (IF). Information flow relationships are not simply pairwise term relationships as those used in previous studies, but are between a set of terms and another term. They allow for context-dependent query expansion. Our experiments conducted on TREC collections show that we can obtain large and significant improvements with our approach. This study shows that LM is an appropriate framework to implement effective query expansion.

Virtual world process perspective visualization

Product Lifecycle Management has been developed as an approach to providing timely engineering information. However, the number of domain specializations within manufacturing makes such information communication disjointed, inefficient and error-prone. In this paper we propose an immersive 3D visualization of linked domain- specific information views for improving and accelerating communication processes in Product Lifecycle Management. With a common and yet understandable visualization of several domain views, interconnections and dependencies become obvious. The conceptual framework presented here links domain-specific information extracts from Product Lifecycle Management systems with each other and displays them via an integrated 3D representation scheme. We expect that this visualization framework should support holistic tactical decision making processes between domain-experts in operational and tactical manufacturing scenarios.

Using Bayesian methods for the estimation of uncertainty in complex statistical models

The research objectives of this thesis were to contribute to Bayesian statistical methodology by contributing to risk assessment statistical methodology, and to spatial and spatio-temporal methodology, by modelling error structures using complex hierarchical models. Specifically, I hoped to consider two applied areas, and use these applications as a springboard for developing new statistical methods as well as undertaking analyses which might give answers to particular applied questions. Thus, this thesis considers a series of models, firstly in the context of risk assessments for recycled water, and secondly in the context of water usage by crops. The research objective was to model error structures using hierarchical models in two problems, namely risk assessment analyses for wastewater, and secondly, in a four dimensional dataset, assessing differences between cropping systems over time and over three spatial dimensions. The aim was to use the simplicity and insight afforded by Bayesian networks to develop appropriate models for risk scenarios, and again to use Bayesian hierarchical models to explore the necessarily complex modelling of four dimensional agricultural data. The specific objectives of the research were to develop a method for the calculation of credible intervals for the point estimates of Bayesian networks; to develop a model structure to incorporate all the experimental uncertainty associated with various constants thereby allowing the calculation of more credible credible intervals for a risk assessment; to model a single day’s data from the agricultural dataset which satisfactorily captured the complexities of the data; to build a model for several days’ data, in order to consider how the full data might be modelled; and finally to build a model for the full four dimensional dataset and to consider the timevarying nature of the contrast of interest, having satisfactorily accounted for possible spatial and temporal autocorrelations. This work forms five papers, two of which have been published, with two submitted, and the final paper still in draft. The first two objectives were met by recasting the risk assessments as directed, acyclic graphs (DAGs). In the first case, we elicited uncertainty for the conditional probabilities needed by the Bayesian net, incorporated these into a corresponding DAG, and used Markov chain Monte Carlo (MCMC) to find credible intervals, for all the scenarios and outcomes of interest. In the second case, we incorporated the experimental data underlying the risk assessment constants into the DAG, and also treated some of that data as needing to be modelled as an ‘errors-invariables’ problem [Fuller, 1987]. This illustrated a simple method for the incorporation of experimental error into risk assessments. In considering one day of the three-dimensional agricultural data, it became clear that geostatistical models or conditional autoregressive (CAR) models over the three dimensions were not the best way to approach the data. Instead CAR models are used with neighbours only in the same depth layer. This gave flexibility to the model, allowing both the spatially structured and non-structured variances to differ at all depths. We call this model the CAR layered model. Given the experimental design, the fixed part of the model could have been modelled as a set of means by treatment and by depth, but doing so allows little insight into how the treatment effects vary with depth. Hence, a number of essentially non-parametric approaches were taken to see the effects of depth on treatment, with the model of choice incorporating an errors-in-variables approach for depth in addition to a non-parametric smooth. The statistical contribution here was the introduction of the CAR layered model, the applied contribution the analysis of moisture over depth and estimation of the contrast of interest together with its credible intervals. These models were fitted using WinBUGS [Lunn et al., 2000]. The work in the fifth paper deals with the fact that with large datasets, the use of WinBUGS becomes more problematic because of its highly correlated term by term updating. In this work, we introduce a Gibbs sampler with block updating for the CAR layered model. The Gibbs sampler was implemented by Chris Strickland using pyMCMC [Strickland, 2010]. This framework is then used to consider five days data, and we show that moisture in the soil for all the various treatments reaches levels particular to each treatment at a depth of 200 cm and thereafter stays constant, albeit with increasing variances with depth. In an analysis across three spatial dimensions and across time, there are many interactions of time and the spatial dimensions to be considered. Hence, we chose to use a daily model and to repeat the analysis at all time points, effectively creating an interaction model of time by the daily model. Such an approach allows great flexibility. However, this approach does not allow insight into the way in which the parameter of interest varies over time. Hence, a two-stage approach was also used, with estimates from the first-stage being analysed as a set of time series. We see this spatio-temporal interaction model as being a useful approach to data measured across three spatial dimensions and time, since it does not assume additivity of the random spatial or temporal effects.

Velocity-jump models with crowding effects

Velocity jump processes are discrete random walk models that have many applications including the study of biological and ecological collective motion. In particular, velocity jump models are often used to represent a type of persistent motion, known as a “run and tumble”, which is exhibited by some isolated bacteria cells. All previous velocity jump processes are non-interacting, which means that crowding effects and agent-to-agent interactions are neglected. By neglecting these agent-to-agent interactions, traditional velocity jump models are only applicable to very dilute systems. Our work is motivated by the fact that many applications in cell biology, such as wound healing, cancer invasion and development, often involve tissues that are densely packed with cells where cell-to-cell contact and crowding effects can be important. To describe these kinds of high cell density problems using a velocity jump process we introduce three different classes of crowding interactions into a one-dimensional model. Simulation data and averaging arguments lead to a suite of continuum descriptions of the interacting velocity jump processes. We show that the resulting systems of hyperbolic partial differential equations predict the mean behavior of the stochastic simulations very well.

Be my buddy: IM and the future of virtual reference

In 2006, the administrators of the Australian virtual reference service, AskNow, entered the Instant Messaging (IM) arena. One of the first large scale, collaborative IM services in the world, the AskNow IM trial provided a unique opportunity to prove IM virtual reference as a concept, as well as to test the technology itself. This paper will discuss the rationale and impetus for the trial, explore the successes and stumbling blocks encountered during the establishment and evolution of the trial and the service model, examine the lessons learnt throughout the trial, and conclude by discussing the way forward for IM services and virtual reference.

Clustering of web users using the tensor decomposed models

We propose to use the Tensor Space Modeling (TSM) to represent and analyze the user’s web log data that consists of multiple interests and spans across multiple dimensions. Further we propose to use the decomposition factors of the Tensors for clustering the users based on similarity of search behaviour. Preliminary results show that the proposed method outperforms the traditional Vector Space Model (VSM) based clustering.

Understanding business process models : the costs and benefits of structuredness

Previous research has put forward a number of properties of business process models that have an impact on their understandability. Two such properties are compactness and(block-)structuredness. What has not been sufficiently appreciated at this point is that these desirable properties may be at odds with one another. This paper presents the results of a two-pronged study aimed at exploring the trade-off between compactness and structuredness of process models. The first prong of the study is a comparative analysis of the complexity of a set of unstructured process models from industrial practice and of their corresponding structured versions. The second prong is an experiment wherein a cohort of students was exposed to semantically equivalent unstructured and structured process models. The key finding is that structuredness is not an absolute desideratum vis-a-vis for process model understandability. Instead, subtle trade-offs between structuredness and other model properties are at play.

AskNow Instant Messaging: innovation in virtual reference

In late 2006, the National Library of Australia implemented a trial Instant Messaging service that ran in parallel with the AskNow chat reference service for a six month period. The trial was a resounding success, proving both a demand for an IM service and the suitability of the medium for reference service provision in a collaborative environment. The trial also allowed the collection of a significant body of data on user expectations, librarian experience and the nature of enquiries. This article begins by introducing the concept of IM and discusses the impetus for its use as a channel for reference service provision. It presents and analyses data collected from user surveys, session transcripts, usage statistics, staff surveys and other staff feedback mechanisms, and explores the issues arising from the data analysis. The article concludes by discussing the IM system architecture that the NLA is currently developing, which will allow the Library to move forward with an ongoing IM service.

An examination of the microscopic simulation models to identify traffic safety indicators

Evaluating the safety of different traffic facilities is a complex and crucial task. Microscopic simulation models have been widely used for traffic management but have been largely neglected in traffic safety studies. Micro simulation to study safety is more ethical and accessible than the traditional safety studies, which only assess historical crash data. However, current microscopic models are unable to mimic unsafe driver behavior, as they are based on presumptions of safe driver behavior. This highlights the need for a critical examination of the current microscopic models to determine which components and parameters have an effect on safety indicator reproduction. The question then arises whether these safety indicators are valid indicators of traffic safety. The safety indicators were therefore selected and tested for straight motorway segments in Brisbane, Australia. This test examined the capability of a micro-simulation model and presents a better understanding of micro-simulation models and how such models, in particular car following models can be enriched to present more accurate safety indicators.

‘Please leave your mobile phone on’ : Social educational networking in a social society : Encouraging in-class engagement at QUT across physical and virtual learning environments

Though stadium style seating in large lecture theatres may suggest otherwise, effective teaching and learning is a not a spectator sport. A challenge in creating effective learning environments in both physical and virtual spaces is to provide optimal opportunity for student engagement in active learning. Queensland University of Technology (QUT) has developed the Open Web Lecture (OWL), a new web-based student response application, which seamlessly integrates a virtual learning environment within the physical learning space. The result is a blended learning experience; a fluid collaboration between academic and students connected to OWL via the University’s Wi-Fi using their own laptop or mobile web device. QUT is currently piloting the OWL application to encourage student engagement. OWL offers opportunities for participants to: • Post comments and questions • Reply to comments
 • "Like" comments
 • Poll students and review data • Review archived sessions. Many of these features instinctively appeal to student users of social networking media, yet avail the academic of control within the University network. Student privacy is respected through a system of preserving peer-peer anonymity, a functionality that seeks to address a traditional reluctance to speak up in large classes. The pilot is establishing OWL as an opportunity for engaging students in active learning opportunities by enabling • virtual learning in physical spaces for large group lectures, seminar groups, workshops and conferences • live collaborative technology connecting students and the academic via the wireless network using their own laptop or mobile device • an non- intimidating environment in which to ask questions • promotion of a sense of community • instant feedback • problem based learning. The student and academic response to OWL has been overwhelmingly positive, crediting OWL as an easy to use application, which creates effective learning opportunities though interactivity and immediate feedback. This poster and accompanying online presentation of the technology will demonstrate how OWL offers new possibilities for active learning in physical spaces by: • providing increased opportunity for student engagement • supporting a range of learners and learning activities • fostering blended learning experiences. The presentation will feature visual displays of the technology, its various interfaces and feedback including clips from interviews with students and academics participating in the early stages of the pilot.

Evaluation of modal analysis techniques using physical models to detect osseointegration of implants in transfemoral amputees

Non-invasive vibration analysis has been used extensively to monitor the progression of dental implant healing and stabilization. It is now being considered as a method to monitor femoral implants in transfemoral amputees. This paper evaluates two modal analysis excitation methods and investigates their capabilities in detecting changes at the interface between the implant and the bone that occur during osseointegration. Excitation of bone-implant physical models with the electromagnetic shaker provided higher coherence values and a greater number of modes over the same frequency range when compared to the impact hammer. Differences were detected in the natural frequencies and fundamental mode shape of the model when the fit of the implant was altered in the bone. The ability to detect changes in the model dynamic properties demonstrates the potential of modal analysis in this application and warrants further investigation.

A Study of XML models for data mining : representations, methods, and issues

With the increasing number of XML documents in varied domains, it has become essential to identify ways of finding interesting information from these documents. Data mining techniques were used to derive this interesting information. Mining on XML documents is impacted by its model due to the semi-structured nature of these documents. Hence, in this chapter we present an overview of the various models of XML documents, how these models were used for mining and some of the issues and challenges in these models. In addition, this chapter also provides some insights into the future models of XML documents for effectively capturing the two important features namely structure and content of XML documents for mining.

A people-to-people recommendation system using Tensor Space Models

Existing recommendation systems often recommend products to users by capturing the item-to-item and user-to-user similarity measures. These types of recommendation systems become inefficient in people-to-people networks for people to people recommendation that require two way relationship. Also, existing recommendation methods use traditional two dimensional models to find inter relationships between alike users and items. It is not efficient enough to model the people-to-people network with two-dimensional models as the latent correlations between the people and their attributes are not utilized. In this paper, we propose a novel tensor decomposition-based recommendation method for recommending people-to-people based on users profiles and their interactions. The people-to-people network data is multi-dimensional data which when modeled using vector based methods tend to result in information loss as they capture either the interactions or the attributes of the users but not both the information. This paper utilizes tensor models that have the ability to correlate and find latent relationships between similar users based on both information, user interactions and user attributes, in order to generate recommendations. Empirical analysis is conducted on a real-life online dating dataset. As demonstrated in results, the use of tensor modeling and decomposition has enabled the identification of latent correlations between people based on their attributes and interactions in the network and quality recommendations have been derived using the 'alike' users concept.

Models of collective cell motion for cell populations with different aspect ratio : diffusion, proliferation and travelling waves

Continuum, partial differential equation models are often used to describe the collective motion of cell populations, with various types of motility represented by the choice of diffusion coefficient, and cell proliferation captured by the source terms. Previously, the choice of diffusion coefficient has been largely arbitrary, with the decision to choose a particular linear or nonlinear form generally based on calibration arguments rather than making any physical connection with the underlying individual-level properties of the cell motility mechanism. In this work we provide a new link between individual-level models, which account for important cell properties such as varying cell shape and volume exclusion, and population-level partial differential equation models. We work in an exclusion process framework, considering aligned, elongated cells that may occupy more than one lattice site, in order to represent populations of agents with different sizes. Three different idealizations of the individual-level mechanism are proposed, and these are connected to three different partial differential equations, each with a different diffusion coefficient; one linear, one nonlinear and degenerate and one nonlinear and nondegenerate. We test the ability of these three models to predict the population level response of a cell spreading problem for both proliferative and nonproliferative cases. We also explore the potential of our models to predict long time travelling wave invasion rates and extend our results to two dimensional spreading and invasion. Our results show that each model can accurately predict density data for nonproliferative systems, but that only one does so for proliferative systems. Hence great care must be taken to predict density data for with varying cell shape.