Permanent URI for this collection
Browse
Recent Submissions
Publication History navigation in location-based mobile systems(Working Paper, University of Waikato, Department of Computer Science, 2010-12-01) Müller, Knut; Hinze, AnnikaThe aim of this paper is to provide an overview and comparison of concepts that have been proposed to guide users through interaction histories (e.g. for web browsers). The goal is to gain insights into history design that may be used for designing an interaction history for the location-based Tourist Information Provider (TIP) system [8]. The TIP system consists of several services that interact on a mobile device.Publication Datalog as a parallel general purpose programming language(Working Paper, University of Waikato, Department of Computer Science, 2010-08-27) Cleary, John G.; Utting, Mark; Clayton, RogerThe increasing available parallelism of computers demands new programming languages that make parallel programming dramatically easier and less error prone. It is proposed that datalog with negation and timestamps is a suitable basis for a general purpose programming language for sequential, parallel and distributed computers. This paper develops a fully incremental bottom-up interpreter for datalog that supports a wide range of execution strategies, with trade-offs affecting efficiency, parallelism and control of resource usage. Examples show how the language can accept real-time external inputs and outputs, and mimic assignment, all without departing from its pure logical semantics.Publication The language of certain conflicts of a nondeterministic process(Journal Article, University of Waikato, Department of Computer Science, 2010-07-07) Malik, RobiThe language of certain conflicts is the most general set of behaviours of a nondeterministic process, which certainly lead to a livelock or deadlock when accepted by another process running in parallel. It is of great use in model checking to detect livelocks or deadlocks in very large systems, and in process-algebra to obtain abstractions preserving livelock and deadlock. Unfortunately, the language of certain conflicts is difficult to compute and has only been approximated in previous work. This paper presents an effective algorithm to calculate the language of certain conflicts for any given nondeterministic finite-state process and discusses its properties. The algorithm is shown to be correct and of exponential complexity.Publication Random model trees: an effective and scalable regression method(Working Paper, University of Waikato, Department of Computer Science, 2010-06) Pfahringer, BernhardWe present and investigate ensembles of randomized model trees as a novel regression method. Such ensembles combine the scalability of tree-based methods with predictive performance rivaling the state of the art in numeric prediction. An extensive empirical investigation shows that Random Model Trees produce predictive performance which is competitive with state-of-the-art methods like Gaussian Processes Regression or Additive Groves of Regression Trees. The training and optimization of Random Model Trees scales better than Gaussian Processes Regression to larger datasets, and enjoys a constant advantage over Additive Groves of the order of one to two orders of magnitude.Publication Implementing an event-driven service-oriented architecture in TIP(Working Paper, University of Waikato, Department of Computer Science, 2010-06-17) Rinck, Michael; Hinze, AnnikaMany mobile devices have a density of services, many of which are context or location-aware. To function, many of these services have to collaborate with other services, which may be located in many different places and networks. There is often more then on service suitable for the task at hand. To decide which service to use, quality of service measurements like the accuracy or reliability of a service need to be known. Users do not want third parties to have statistics on how and where they used services. Therefore the collaboration needs to be anonymous. This project implements a model of event-based context-aware service collaboration on a publish/subscribe basis. We compare different implementation designs, with focus on anonymity and quality of service of the services.Publication Efficient multi-label classification for evolving data streams(University of Waikato, Department of Computer Science, 2010-05) Read, Jesse; Bifet, Albert; Holmes, Geoffrey; Pfahringer, BernhardMany real world problems involve data which can be considered as multi-label data streams. Efficient methods exist for multi-label classification in non streaming scenarios. However, learning in evolving streaming scenarios is more challenging, as the learners must be able to adapt to change using limited time and memory. This paper proposes a new experimental framework for studying multi-label evolving stream classification, and new efficient methods that combine the best practices in streaming scenarios with the best practices in multi-label classification. We present a Multi-label Hoeffding Tree with multilabel classifiers at the leaves as a base classifier. We obtain fast and accurate methods, that are well suited for this challenging multi-label classification streaming task. Using the new experimental framework, we test our methodology by performing an evaluation study on synthetic and real-world datasets. In comparison to well-known batch multi-label methods, we obtain encouraging results.Publication Contextual queries and situated information needs for mobile users(Working Paper, University of Waikato, Department of Computer Science, 2010-02-16) Hinze, Annika; Chang, Carole; Nichols, David M.The users of mobile devices increasingly use networked services to address their information needs. Questions asked by mobile users are strongly influenced by contextual factors such as location, conversation and activity. We report on a diary study performed to better understand mobile information needs. Participants’ diary entries are used as a basis for discussing the geographical and situational context in which mobile information behaviour occurs. The suitability of user queries to be answered by a portable knowledge collection and web search are also considered. We find that the type of questions recorded by participants varies across their locations, with differences between home, shopping and in-car contexts. These variations occur both in the query terms and in the form of desired answers. Both the location of queries and the participants’ activities affected participants’ questions. When information needs were affected by both location and activity, they tended to be strongly affected by both factors. The overall picture that emerges is one of multiple contextual influences interacting to shape mobile information needs. Mobile devices that attempt to adapt to users’ context will need to account for a rich variety of situational factors.