Turku Center for Computer Science

Turku, Finland

Turku Center for Computer Science

Turku, Finland
SEARCH FILTERS
Time filter
Source Type

Salomaa A.,Turku Center for Computer Science
Theoretical Computer Science | Year: 2017

We investigate reaction systems introduced in , in particular, the subclass of minimal reaction systems added with a feature of duration. It turns out that the model is computationally strong. Moreover, in some cases the lengths of the resulting sequences and cycles can be found out directly by arithmetical properties of the duration values. © 2017.


Salomaa A.,Turku Center for Computer Science
International Journal of Foundations of Computer Science | Year: 2013

We investigate formal properties, mainly issues connected with propositional logic, of reaction systems introduced by Ehrenfeucht and Rozenberg. We are concerned only with the most simple variant of the systems. Basic properties of propositional formulas are expressed in terms of reaction systems. This leads to NP-completeness (resp. co-NP-completeness) of many problems concerning reaction systems. Among such problems are: (i) deciding whether the function defined by the system is total, (ii) determining the inverse of the function, (iii) deciding whether state sequences always end with a loop. Propositional formulas with monotonic truth-functions yield a particularly simple representation in terms of reaction systems. © 2013 World Scientific Publishing Company.


Timonen V.,Turku Center for Computer Science
Computer Graphics Forum | Year: 2013

Screen-space ambient occlusion and obscurance have become established methods for rendering global illumination effects in real-time applications. While they have seen a steady line of refinements, their computational complexity has remained largely unchanged and either undersampling artefacts or too high render times limit their scalability. In this paper we show how the fundamentally quadratic per-pixel complexity of previous work can be reduced to a linear complexity. We solve obscurance in discrete azimuthal directions by performing line sweeps across the depth buffer in each direction. Our method builds upon the insight that scene points along each line can be incrementally inserted into a data structure such that querying for the largest occluder among the visited samples along the line can be achieved at an amortized constant cost. The obscurance radius therefore has no impact on the execution time and our method produces accurate results with smooth occlusion gradients in a few milliseconds per frame on commodity hardware. © 2013 The Author(s) Computer Graphics Forum © 2013 The Eurographics Association and John Wiley & Sons Ltd.


Sarlin P.,Turku Center for Computer Science
Pattern Recognition Letters | Year: 2013

A key starting point for financial stability surveillance is understanding past, current and possible future risks and vulnerabilities. Through temporal data and dimensionality reduction, or visual dynamic clustering, this paper aims to present a holistic view of cross-sectional macro-financial patterns over time. The Self-Organizing Time Map (SOTM) is a recent adaptation of the Self-Organizing Map for exploratory temporal structure analysis, which disentangles cross-sectional data structures over time. We apply the SOTM, as well as its combination with classical cluster analysis, in financial stability surveillance. Thus, this paper uses the SOTM for decomposing and identifying temporal structural changes in macro-financial data before, during and after the global financial crisis of 2007-2009. © 2013 Elsevier B.V. All rights reserved.


Sarlin P.,Turku Center for Computer Science
Neurocomputing | Year: 2013

This paper adopts and adapts Kohonen's standard self-organizing map (SOM) for exploratory temporal structure analysis. The self-organizing time map (SOTM) implements SOM-type learning to one-dimensional arrays for individual time units, preserves the orientation with short-term memory and arranges the arrays in an ascending order of time. The two-dimensional representation of the SOTM attempts thus twofold topology preservation, where the horizontal direction preserves time topology and the vertical direction data topology. This enables discovering the occurrence and exploring the properties of temporal structural changes in data. For representing qualities and properties of SOTMs, we adapt measures and visualizations from the standard SOM paradigm, as well as introduce a measure of temporal structural changes. The functioning of the SOTM, and its visualizations and quality and property measures, are illustrated on artificial toy data. The usefulness of the SOTM in a real-world setting is shown on poverty, welfare and development indicators. © 2012 Elsevier B.V.


Bjorne J.,Turku Center for Computer Science
BMC bioinformatics | Year: 2012

We present a system for extracting biomedical events (detailed descriptions of biomolecular interactions) from research articles, developed for the BioNLP'11 Shared Task. Our goal is to develop a system easily adaptable to different event schemes, following the theme of the BioNLP'11 Shared Task: generalization, the extension of event extraction to varied biomedical domains. Our system extends our BioNLP'09 Shared Task winning Turku Event Extraction System, which uses support vector machines to first detect event-defining words, followed by detection of their relationships. Our current system successfully predicts events for every domain case introduced in the BioNLP'11 Shared Task, being the only system to participate in all eight tasks and all of their subtasks, with best performance in four tasks. Following the Shared Task, we improve the system on the Infectious Diseases task from 42.57% to 53.87% F-score, bringing performance into line with the similar GENIA Event Extraction and Epigenetics and Post-translational Modifications tasks. We evaluate the machine learning performance of the system by calculating learning curves for all tasks, detecting areas where additional annotated data could be used to improve performance. Finally, we evaluate the use of system output on external articles as additional training data in a form of self-training. We show that the updated Turku Event Extraction System can easily be adapted to all presently available event extraction targets, with competitive performance in most tasks. The scope of the performance gains between the 2009 and 2011 BioNLP Shared Tasks indicates event extraction is still a new field requiring more work. We provide several analyses of event extraction methods and performance, highlighting potential future directions for continued development.


Salomaa A.,Turku Center for Computer Science
Natural Computing | Year: 2013

In reaction systems introduced by Ehrenfeucht and Rozenberg the number of resources is, by definition, at least 2. If it is exactly 2, the system is referred to as minimal. We compare minimal reaction systems with almost minimal ones, where the number of resources equals 3. The difference turns out to be huge. While many central problems for minimal systems are of low polynomial complexity, the same problems in the almost minimal case are NP- or co-NP-complete. The situation resembles the difference between 2-SAT and 3-SAT, also from the point of view of techniques used. We also compare maximal sequence lengths obtainable in the two cases. We are concerned only with the most simple variant of reaction systems. © 2013 Springer Science+Business Media Dordrecht.


Salomaa A.,Turku Center for Computer Science
Theoretical Computer Science | Year: 2012

The paper investigates formal properties of reaction systems introduced by Ehrenfeucht and Rozenberg. A reaction system defines a function from the set 2S of subsets of a finite set S into 2S itself. We investigate properties of such functions, and characterize situations when the function is total. We also introduce and characterize the property of functional completeness. Function classes defined by different types of reaction systems are compared. Comparisons are carried out also between different methods of generating long sequences and cycles. © 2012 Elsevier B.V. All rights reserved.


Sarlin P.,Turku Center for Computer Science
International Journal of Machine Learning and Cybernetics | Year: 2012

This paper uses the self-organizing map (SOM), a neural network-based projection and clustering technique, for monitoring the millennium development goals (MDGs). The eight MDGs represent commitments to reduce poverty and hunger, and to tackle ill-health, gender inequality, lack of education, lack of access to clean water and environmental degradation by 2015. This paper presents a SOM model for cross sectional and temporal visual benchmarking of countries and pairs the map with a geospatial dimension by mapping the clustering onto a geographic map. The temporal monitoring is facilitated by fuzzifying the second-level clustering with membership degrees. By creating an MDG index, and associating the SOM model with it, the model enables cross sectional and temporal analysis of the overall MDG progress of countries or regions. Further, the SOM model enables analysis of country-specific as well as regional performance according to a user-specified level of aggregation. The result of this paper is an MDG map for visual tracking and monitoring of the progress of MDG indicators. © 2011 Springer-Verlag.


Liu Y.,Åbo Akademi University | Li H.,Turku Center for Computer Science
Computers in Human Behavior | Year: 2011

Unlike traditional technologies, the use of mobile technology is exposed to shifting use contexts. Use context has frequently been described as an important factor influencing the adoption of mobile innovations. However, empirical evidence about the impact of use context is limited. This paper investigated the effect of use context on the formation of users' perceptions of mobile hedonic services by using mobile gaming as an example. Through the employment of structural equation modelling technology, an adoption model of mobile gaming is proposed and assessed based on results from 267 questionnaires. The results show that use context is the strongest predictor of mobile game adoption. It directly or indirectly affects all different perceptions of mobile gaming in significant ways, including perceived ease of use, perceived usefulness, perceived enjoyment, cognitive concentration, attitude and behavioral intention. Additionally, perceived usefulness, perceived enjoyment and cognitive concentration all have a positive influence on the attitudinal variables of mobile game acceptance. We concluded that the formation of people's perceptions about mobile gaming is conditional and based on the special consideration of certain use contexts. Both theoretical and practical implications are discussed. © 2010 Elsevier Ltd. All rights reserved.

Loading Turku Center for Computer Science collaborators
Loading Turku Center for Computer Science collaborators