Exploring cognitive style and task-specific preferences for process representations

Process models describe someone’s understanding of processes. Processes can be described using unstructured, semi-formal or diagrammatic representation forms. These representations are used in a variety of task settings, ranging from understanding processes to executing or improving processes, with the implicit assumption that the chosen representation form will be appropriate for all task settings. We explore the validity of this assumption by examining empirically the preference for different process representation forms depending on the task setting and cognitive style of the user. Based on data collected from 120 business school students, we show that preferences for process representation formats vary dependent on application purpose and cognitive styles of the participants. However, users consistently prefer diagrams over other representation formats. Our research informs a broader research agenda on task-specific applications of process modeling. We offer several recommendations for further research in this area.

CoWs, Google and online possibilities : creating digital posters to enhance student engagement

A new Bachelor of Science (BSc) course was introduced at Queensland University of Technology (QUT) in 2013 and focused on inquiry-based, collaborative and active learning. Two of the first year units required that students carry out a group poster assessment task. This poster provides a preliminary evaluation from an academic staff perspective of the assessment approach used, whereby students created digital posters to utilise the affordances of new learning spaces. The digital posters approach was first introduced to a group of academic staff from the Science and Engineering Faculty (SEF) in 2012 during a professional development program to explicitly develop skills and shared understandings of teaching in collaborative learning spaces (Steel & Andrews, 2012). Considerations were given to the pedagogical requirements of a poster assessment task, the affordances of the learning space and an identification of possible benefits of using Google Sites to create digital posters. Positive feedback from this group (as highlighted in the quotes shown) and subsequent approval from unit coordinators for two of the new first year BSc units meant that the approach was adopted for Semester 1, 2013 with approximately 360 students in each unit.

Multiple perspectives on quality of life for residents with dementia in long term care facilities: Protocol for a comprehensive Australian study

Background Dementia is a chronic illness without cure or effective treatment, which results in declining mental and physical function and assistance from others to manage activities of daily living. Many people with dementia live in long term care facilities, yet research into their quality of life (QoL) was rare until the last decade. Previous studies failed to incorporate important variables related to the facility and care provision or to look closely at the daily lives of residents. This paper presents a protocol for a comprehensive, multi-perspective assessment of QoL of residents with dementia living in long term care in Australia. A secondary aim is investigating the effectiveness of self-report instruments for measuring QoL. Methods The study utilizes a descriptive, mixed methods design to examine how facility, care staff, and resident factors impact QoL. Over 500 residents with dementia from a stratified, random sample of 53 facilities are being recruited. A sub-sample of 12 residents is also taking part in qualitative interviews and observations. Conclusions This national study will provide a broad understanding of factors underlying QoL for residents with dementia in long term care. The present study uses a similar methodology to the US-based Collaborative Studies of Long Term Care (CS-LTC) Dementia Care Study, applying it to the Australian setting.

A case study of teaching English and multimodality with ICTs: Constraints and possibilities

Current Australian policies and curricular frameworks demand that teachers and students use technology creatively and meaningfully in classrooms to develop students into 21C technological citizens. English teachers and students also have to learn new metalanguage around visual grammar since multimodal tasks often combine creative with critical General Capabilities (GC) with that of the of ICTs and literacy in the Australian Curriculum: English (AC:E). Both teachers and learners come to these tasks with varying degrees of techno-literacy, skills and access to technologies. This paper reports on case-study research following a technology based collaborative professional development (PD) program between a university Lecturer facilitator and English Teachers in a secondary Catholic school. The study found that the possibilities for creative and critical engagement are rich, but there are real grounded constraints such as lack of time, impeding teachers’ ability to master and teach new technologies in classrooms. Furthermore, pedagogical approaches are affected by technical skill levels and school infrastructure concerns which can militate against effective use of ICTs in school settings. The research project was funded by the Brisbane Catholic Education Office and focused on how teachers can be supported in these endeavours in educational contexts as they prepare students of English to be creative global citizens who use technology creatively.

Mining Cell Transition Data

Cell transition data is obtained from a cellular phone that switches its current serving cell tower. The data consists of a sequence of transition events, which are pairs of cell identifiers and transition times. The focus of this thesis is applying data mining methods to such data, developing new algorithms, and extracting knowledge that will be a solid foundation on which to build location-aware applications. In addition to a thorough exploration of the features of the data, the tools and methods developed in this thesis provide solutions to three distinct research problems. First, we develop clustering algorithms that produce a reliable mapping between cell transitions and physical locations observed by users of mobile devices. The main clustering algorithm operates in online fashion, and we consider also a number of offline clustering methods for comparison. Second, we define the concept of significant locations, known as bases, and give an online algorithm for determining them. Finally, we consider the task of predicting the movement of the user, based on historical data. We develop a prediction algorithm that considers paths of movement in their entirety, instead of just the most recent movement history. All of the presented methods are evaluated with a significant body of real cell transition data, collected from about one hundred different individuals. The algorithms developed in this thesis are designed to be implemented on a mobile device, and require no extra hardware sensors or network infrastructure. By not relying on external services and keeping the user information as much as possible on the user s own personal device, we avoid privacy issues and let the users control the disclosure of their location information.

Nonlinear dynamic state estimation in instrumented structures with conditionally linear Gaussian substructures

Many problems of state estimation in structural dynamics permit a partitioning of system states into nonlinear and conditionally linear substructures. This enables a part of the problem to be solved exactly, using the Kalman filter, and the remainder using Monte Carlo simulations. The present study develops an algorithm that combines sequential importance sampling based particle filtering with Kalman filtering to a fairly general form of process equations and demonstrates the application of a substructuring scheme to problems of hidden state estimation in structures with local nonlinearities, response sensitivity model updating in nonlinear systems, and characterization of residual displacements in instrumented inelastic structures. The paper also theoretically demonstrates that the sampling variance associated with the substructuring scheme used does not exceed the sampling variance corresponding to the Monte Carlo filtering without substructuring. (C) 2012 Elsevier Ltd. All rights reserved.

Dynamic state estimation for identifying earthquake support motions in instrumented structures

The problem of identification of multi-component and (or) spatially varying earthquake support motions based on measured responses in instrumented structures is considered. The governing equations of motion are cast in the state space form and a time domain solution to the input identification problem is developed based on the Kalman and particle filtering methods. The method allows for noise in measured responses, imperfections in mathematical model for the structure, and possible nonlinear behavior of the structure. The unknown support motions are treated as hypothetical additional system states and a prior model for these motions are taken to be given in terms of white noise processes. For linear systems, the solution is developed within the Kalman filtering framework while, for nonlinear systems, the Monte Carlo simulation based particle filtering tools are employed. In the latter case, the question of controlling sampling variance based on the idea of Rao-Blackwellization is also explored. Illustrative examples include identification of multi-component and spatially varying support motions in linear/nonlinear structures.

Aprender colaborativamente Fundamentos de Tecnología de Computadores en base a problemas y proyectos

Capítulo 3 del libro: Guisasola, Jenaro ; Garmendia, Mikel (eds.) "Aprendizaje basado en problemas, proyectos y casos: diseño e implementación de experiencias en la universidad" (ISBN: 978-84-9860-959-2)

Bayesian and L1 Approaches to Sparse Unsupervised Learning

The use of L1 regularisation for sparse learning has generated immense research interest, with successful application in such diverse areas as signal acquisition, image coding, genomics and collaborative filtering. While existing work highlights the many advantages of L1 methods, in this paper we find that L1 regularisation often dramatically underperforms in terms of predictive performance when compared with other methods for inferring sparsity. We focus on unsupervised latent variable models, and develop L1 minimising factor models, Bayesian variants of "L1", and Bayesian models with a stronger L0-like sparsity induced through spike-and-slab distributions. These spike-and-slab Bayesian factor models encourage sparsity while accounting for uncertainty in a principled manner and avoiding unnecessary shrinkage of non-zero values. We demonstrate on a number of data sets that in practice spike-and-slab Bayesian methods outperform L1 minimisation, even on a computational budget. We thus highlight the need to re-assess the wide use of L1 methods in sparsity-reliant applications, particularly when we care about generalising to previously unseen data, and provide an alternative that, over many varying conditions, provides improved generalisation performance.

增量式个性化推荐系统研究

互联网个性化推荐系统(Internet personal recommender systems)是根据用户的兴趣推荐最相关的互联网信息给用户的系统。在网上信息过载矛盾越来越严重、用户信息检索的个性化需求日益增强的现状下，推荐系统已经在搜索引擎、电子商务、网上社区等互联网关键应用中起到了关键性的作用，并且越来越受到重视。 然而，在大型网站上部署一个成熟推荐系统的代价依然很大，需要大量的计算和存储资源，推荐的准确性也依然有很大提升空间和需求，这就为推荐系统的研究提供了很多挑战。在这些挑战中推荐算法的准确性和可扩展性一直是该领域最为关注的两个问题，所谓推荐的准确性是指推荐的信息中用户真正感兴趣的比例，而可扩展性指的是系统能否在可容忍的时间和空间复杂度内处理海量的数据。如何在提高算法推荐准确性的同时增强算法的可扩展性是推荐系统改进的主要研究目标。然而，目前学术界的研究更多侧重于提高推荐算法的准确性，而对于可扩展性，很多准确性很高的算法由于需要比较复杂的计算，处理大规模动态数据的能力往往比较有限，并且它们的评测实验中并没有将可扩展性纳入到评价范畴，导致这些算法目前还很难在工业界大规模应用。 本论文的研究试图解决这一问题。通过在推荐算法中借鉴增量学习(Incremental learning)的思想，即考虑最新的训练数据来更新原有的机器学习模型，不需要或仅需要参考部分旧的训练数据，相对于使用全部数据也即批量的处理方式，增量式改进可以大大降低模型更新的复杂度，从而可以大幅度提高推荐算法在遇到新的训练数据时推荐模型更新的效率，降低计算代价，使得推荐模型的更新可以更加及时，进而提高推荐结果的准确性。具体来说，我们在提出了两种新的增量式协同过滤算法的同时，采用增量式学习的方法对目前准确性最好的若干推荐算法进行加速，特别是提高这些算法面对新的训练数据的更新模型的速度和效率，从而为这些算法的大规模的应用提供了可能。另一方面，新的训练数据包含了最新的用户兴趣，因此相对于旧的训练数据，算法在做更新时应给予更高的权重，这样才能做到推荐的结果在考虑到用户长期兴趣的同时，特别考虑用户近期的兴趣，从而使得推荐结果更加准确。这两方面归纳起来，我们旨在通过增量式学习使得推荐算法在更新时更加高效和精确，真正适用于互联网上海量数据的推荐，同时对其他增量式推荐系统方面的研究也具有借鉴意义。我们的改进工作主要包括以下几个方面： 基于主题模型的增量式推荐算法。主题模型，特别是概率隐含主题模型(PLSA)是一种广泛应用于推荐系统的主流方法，在文本推荐、图像推荐以及协同过滤推荐领域都有着很好的推荐效果。目前制约PLSA算法取得更大成功的重要因素就是PLSA算法更新的复杂度过高，使得学习模型的更新只能做批量式处理，这样就导致推荐的时效性不高，也没有办法体现用户的最新的兴趣和整体的最新动态。我们提出了一种增量式学习方法，可以应用于文本分析领域和协同过滤领域，当有新的训练数据到来时，对于基于文本的推荐，增量式更新方法仅寻找最相关的用户和文本以及涉及到的单词进行主题分布的更新，并给予新的文本以更高权重；对于协同过滤，我们的方法仅对当前用户所评分过得物品以及当前物品所涉及的用户进行更新，大大降低了更新的运算复杂度，提高了新数据在推荐算法中所占的权重，使得推荐更加准确、及时。我们的算法在天涯问答文本数据集上和MovieLens电影推荐数据集、Last.FM歌曲推荐数据集、豆瓣图书推荐数据集等协同过滤数据集上取得了很好的效果。 基于蚁群算法(Ant colony algorithm)的协同过滤推荐方法。受到群体智能(Swarm intelligence)算法的启发，我们提出了一种类似于蚁群算法的协同过滤推荐方法——Ant Collaborative Filtering，初始化阶段该方法给予每个用户或一组用户以全局唯一的单位数量的信息素，当用户对物品评分或者用户表示对该物品感兴趣时，用户所携带的信息素相应的传播到该物品上，同时该物品上已有的信息素（初始化为0）也会相应的传播给该用户；此外，用户和物品所携带的信息素会随着时间的推移有一定速率的挥发，通过挥发机制，可以在推荐时更重视用户近期的兴趣；推荐阶段，按照用户和物品所携带的信息素的种类和数量，我们可以得到相应的相似度，进而通过经典的相似度比较的方法来进行推荐。基于蚁群的协同过滤方法的优势在于可以有效的降低训练数据中的稀疏性，并且推荐算法可以实时的进行更新和推荐，同时考虑了用户兴趣随着时间的变化。我们在MovieLens电影评分、豆瓣书籍推荐、Last.FM音乐推荐数据集上验证了我们的方法。最后，我们建立了一个互联网新闻推荐系统，该系统以Firefox插件形式实现，自动采集用户浏览兴趣和偏好，后端使用不同的推荐算法推荐用户感兴趣的新闻给用户。 基于联合聚类(Co-clustering)的两阶段协同过滤方法。聚类(Clustering)是一种缩小数据规模、降低数据稀疏性的有效方法。对于庞大而稀疏的协同过滤训练数据来说，聚类是一种很自然事实上也的确很有效的预处理方法。因此我们提出了一种两阶段协同过滤框架：首先通过我们提出的一种联合聚类的方法，将原始评分矩阵分解成很多维度很小的块，每一块里面包含相似的用户对相似的物品的评分，然后通过矩阵拟合的方法（我们使用了非负矩阵分解NMF和主题模型PLSA）来对这些小块中的未知评分进行预测。当用户新增了对于某物品的一条评分，我们仅需要更新该用户或该物品所处的数据块进行重新评分预估，大大加快了评分预估的速度。我们在MovieLens电影评分数据集上验证了该算法的效果。 本文的研究成果不仅可以直接应用于大型推荐系统中，而且对于增量式推荐系统的后续研究也具有一定的指导意义。首先基于PLSA的增量式推荐算法对于其他基于图模型的推荐系统具有借鉴价值，其次蚁群推荐算法为一类新的、基于群体智能(Swarm intellignece)的协同过滤算法做出了有价值的探索，最后我们提出的两阶段协同过滤框架对于提高推荐算法的可扩展性和更新效率提出了一个通用的有效解决方案。 推荐系统是一个无止尽的优化的过程，除了推荐精度的不断提高之外，推荐算法的性能随着互联网上数据量的增加也需要进一步提高，增量式学习无疑是提高推荐算法更新速度最重要的方法，本文的研究为这一方向提供了参考。

面向多目标追踪的无线传感器网络协同任务分配研究

在基于动态联盟机制的无线传感器网络协同任务分配研究中,为了解决多目标追踪带来的联盟间的资源竞争问题,本文采用分布式约束满足算法解决多动态联盟间的协同问题.根据无线传感器网络多目标追踪的应用需求,建立了基于动态联盟机制的协同任务分配的分布式约束满足模型,并采用分布式随机算法求解满足约束条件的动态联盟集合,实现多动态联盟间的协同.仿真结果表明,分布式约束满足算法有效地解决了多目标追踪中多个动态联盟间的资源竞争问题,能够有效降低系统的能量消耗。

基于动态联盟机制的无线传感器网络任务分配问题研究

微机电系统、先进传感器、无线通信及现代网络等技术的进步，推动了无线传感器网络的产生和发展。集数据采集、处理、无线传输等功能于一体的无线传感器网络扩展了人们的信息获取能力，将逻辑上的信息世界与真实物理世界融合在一起，将改变人类与物理世界的交互方式。 任务分配就是在无线传感器节点协同过程中，确定由哪些节点来完成特定的任务。针对不同的应用领域，需要不同的任务分配方案与之适应，才能得到最佳的资源利用率和检测性能。无线传感器网络是一种分布式网络，需要多个节点协同执行检测任务，因此任务分配问题既是无线传感器网络的基本问题，也是无线传感器网络应用的基础。动态联盟机制是一种事件触发的任务分配机制，对动态环境的适应性相对较好，适用于动态性要求相对较高的目标追踪等无线传感器网络应用领域。 本文针对基于动态联盟机制的无线传感器网络的任务分配问题展开研究。论文的主要工作如下： 综合论述了无线传感器网络的任务分配问题的研究内容、特性和研究现状等。 针对节点能量和能力严格受限的问题，提出了一种基于拍卖的动态联盟组建机制。首先，将拍卖方法引入了动态联盟的组建过程，简化了动态联盟的组建，提高了动态联盟的结盟成功率，在更加有效地利用网络能量资源的同时提升了网络性能。而后，在选择拍卖标的时，综合考虑了节点剩余能量和通信能量消耗，提升了无线传感器网络的生命周期。 针对动态联盟的组织维护和能量均衡性问题，提出了一种基于协商的动态联盟成员更新机制。当动态联盟的成员能量消耗达到一定程度时，采用基于协商的机制对动态联盟成员进行更新，以增强系统能量消耗的均衡性，从而延长网络的生命周期。 针对任务影响区域不断变化的问题，给出了一种基于资源预留的动态联盟检测区域更新机制，以适应对动态联盟的动态性的要求。首先加入了联盟覆盖范围和休眠盟员的概念，以消除针对同一任务的检测传感器节点的冗余，进一步降低网络执行任务期间的能量消耗。而后又加入动态联盟的更新机制，以消除联盟衔接期间网络对任务的暂时“失明”，保证动态联盟执行任务时的连续性，从而在一定程度上保证网络的检测性能。 针对多动态联盟间的协同问题，提出了基于分布式约束满足的多联盟协同机制。根据多个目标经过无线传感器网络监控区域时的任务分配需求，提出了一种基于分布式约束满足的多动态联盟协同机制，建立了基于动态联盟机制的任务分配问题的分布式约束满足模型，采用分布式随机算法进行求解，可以针对未知数量的目标追踪进行分布式的动态协同任务分配，有效解决了动态联盟间的协同问题，从而降低网络的能量消耗，节省网络资源。 总之，论文对基于动态联盟机制的无线传感器网络任务分配问题进行了研究和探讨，旨在对无线传感器网络的应用起到一定的推动作用。

Generalized Low-Rank Approximations

We study the frequent problem of approximating a target matrix with a matrix of lower rank. We provide a simple and efficient (EM) algorithm for solving {\\em weighted} low rank approximation problems, which, unlike simple matrix factorization problems, do not admit a closed form solution in general. We analyze, in addition, the nature of locally optimal solutions that arise in this context, demonstrate the utility of accommodating the weights in reconstructing the underlying low rank representation, and extend the formulation to non-Gaussian noise models such as classification (collaborative filtering).

Inferring and Exploiting Categories for Next Location Prediction

Predicting the next location of a user based on their previous visiting pattern is one of the primary tasks over data from location based social networks (LBSNs) such as Foursquare. Many different aspects of these so-called “check-in” profiles of a user have been made use of in this task, including spatial and temporal information of check-ins as well as the social network information of the user. Building more sophisticated prediction models by enriching these check-in data by combining them with information from other sources is challenging due to the limited data that these LBSNs expose due to privacy concerns. In this paper, we propose a framework to use the location data from LBSNs, combine it with the data from maps for associating a set of venue categories with these locations. For example, if the user is found to be checking in at a mall that has cafes, cinemas and restaurants according to the map, all these information is associated. This category information is then leveraged to predict the next checkin location by the user. Our experiments with publicly available check-in dataset show that this approach improves on the state-of-the-art methods for location prediction.