Peter W Eklund - Profile on Academia.edu (original) (raw)

Papers by Peter W Eklund

SN Computer Science/SN computer science, Mar 14, 2024

This paper presents a novel approach to video super-resolution (VSR) by focusing on the selection... more This paper presents a novel approach to video super-resolution (VSR) by focusing on the selection of input frames, a process critical to VSR. VSR methods typically rely on deep learning techniques, those that are able to learn features from a large dataset of low-resolution (LR) and corresponding high-resolution (HR) videos and generate high-quality HR frames from any new LR input frames using the learned features. However, these methods often use as input the immediate neighbouring frames to a given target frame without considering the importance and dynamics of the frames across the temporal dimension of a video. This work aims to address the limitations of the conventional sliding-window mechanisms by developing input frame selection algorithms. By dynamically selecting the most representative neighbouring frames based on content-aware selection measures, our proposed algorithms enable VSR models to extract more informative and accurate features that are better aligned with the target frame, leading to improved performance and higher-quality HR frames. Through an empirical study, we demonstrate that the proposed dynamic content-aware selection mechanism improves super-resolution results without any additional architectural overhead, offering a counter-intuitive yet effective alternative to the long-established trend of increasing architectural complexity to improve VSR results.

Neural Computing and Applications, May 31, 2022

Omnidirectional Videos (or 360° videos) are widely used in Virtual Reality (VR) to facilitate imm... more Omnidirectional Videos (or 360° videos) are widely used in Virtual Reality (VR) to facilitate immersive and interactive viewing experiences. However, the limited spatial resolution in 360° videos does not allow for each degree of view to be represented with adequate pixels, limiting the visual quality offered in the immersive experience. Deep learning Video Super-Resolution (VSR) techniques used for conventional videos could provide a promising software-based solution; however, these techniques do not tackle the distortion present in equirecentagular projections of 360° video signals. An additional obstacle is the limited 360° video datasets to study. To address these issues, this paper creates a novel 360° Video Dataset (360VDS) with a study of the extensibility of conventional VSR models to 360° videos. This paper further proposes a novel deep learning model for 360° Video Super-Resolution (360° VSR), called Spherical Signal Super-resolution with Proportioned Optimisation (S3PO). ...

IEEE Transactions on Computational Social Systems, 2021

Emergency Management and Beyond S INCE early 2020, the COVID-19 global pandemic has significantly... more Emergency Management and Beyond S INCE early 2020, the COVID-19 global pandemic has significantly impacted almost every aspect of the human society throughout the world. Until now, middle of 2021, although with all the efforts on pandemic intervention and vaccination, COVID-19 is still hovering around the world, resulting in more than 177 million confirmed cases and 3.8 million deaths. In June 2020, IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS (TCSS) launched the Special Issue on Computational Social Systems for COVID-19 Emergency Management and Beyond, aiming to provide the report of state-of-the-art research work from the global that addresses innovative techniques, applications, and results. The special issue received submissions all around the globe, and 13 articles were reviewed and recommended for publication, among them two from North America, two from China, two from India, three from Australia, one from Japan, and one from Pakistan. A quick scanning of the accepted articles is presented in the following. The modeling of the COVID-19 spread is always a central issue to investigate, and we received a number of related works. Rastgoftar and Atkins proposed a novel pandemic spread model by integrating historical data of infected cases, deaths, and recoveries, and obtained a data-driven model for spread dynamics among the human population. Wang et al. analyzed the impact of related multisource urban data, such as local temperature, relative humidity, air quality, and inflow rate, from Hubei province, and proposed a simple but effective short-term prediction model for COVID-19 cases, considering the human mobility from Hubei province to the target cities. Nie et al. proposed a new dynamic susceptible-exposed-infected-recovered (SEIR) model to simulate the epidemic situation in China considering and incorporating internet messages' information entropy. Kumar and Viral utilized the autoregressive moving average model to forecast the situation of the following few months in India, and the forecasting results show that India will also face a worse condition if the rules and regulations by the government are not strict. Mishra et al. utilized empirical data and tried to figure out the impact of various parameters, such as the probability of transmission, contact rate, infectious, and susceptible populations. The data analysis approximated the reproduction number of COVID-19 between 1.5 and 4 with a low fatality rate. Iqbal et al. used the long short-term memory (LSTM) model to predict the volume of COVID-19 patients in Pakistan using COVID-19 data from March 2020 to May 2020, and results were used to validate the advantage of the model. The COVID-19 pandemic impacts directly on human beings' sentiment, resulting in a large body of corresponding Digital Object Identifier 10.

arXiv (Cornell University), Jan 3, 2023

Supervised Question Answering systems (QA systems) rely on domain-specific humanlabeled data for ... more Supervised Question Answering systems (QA systems) rely on domain-specific humanlabeled data for training. Unsupervised QA systems generate their own questionanswer training pairs, typically using secondary knowledge sources to achieve this outcome. Our approach (called PIE-QG) uses Open Information Extraction (OpenIE) to generate synthetic training questions from paraphrased passages and uses the question-answer pairs as training data for a language model for a state-of-the-art QA system based on BERT. Triples in the form of <subject, predicate, ob-ject> are extracted from each passage, and questions are formed with subjects (or objects) and predicates while objects (or subjects) are considered as answers. Experimenting on five extractive QA datasets demonstrates that our technique achieves on-par performance with existing state-of-the-art QA systems with the benefit of being trained on an order of magnitude fewer documents and without any recourse to external reference data sources.

2p-D-7 遠心分離法により濃縮した単層カーボンナノチューブのラマン散乱

日本物理学会講演概要集年会, Mar 15, 1996

Springer eBooks, 1994

As it knows most of the quotidian types of control, problems are not simply to evaluate and consi... more As it knows most of the quotidian types of control, problems are not simply to evaluate and considerate formal modelling based in traditional techniques. The process control makes the evaluation and executions more efficient in the industry. This article was made with the purpose to compare two types of control, one with FUZZY logic and second one PID control. Here we have developed temperature control system using fuzzy logic. The flyback converter with voltage doubler rectifier acts as an output module. To overcome the efficiency degradation during lightload due to load dependent soft switching of the ZVS, a control method using pulse width modulation (PWM)proportional to the load current is used. Comparison between Fuzzy logic controller & PID controller based on pulse width modulation is proposed and the results are analyzed. Thus comparing the result PID controller gives more accurate results than first Fuzzy logic controller.

Lecture Notes in Computer Science, 2000

International Semantic Web Conference, 2013

This project supports application for a practice-based [4] PhD by Peter Goodall. The project will... more This project supports application for a practice-based [4] PhD by Peter Goodall. The project will produce a system architecture and proof-of-concept laboratory implementation to model, instrument, prototype, and evaluate the effectiveness of several alternate system designs which are intended to enable small end-user communities to evolve specialized ontologies and annotations for entities important to them. Depending on available resources, the laboratory may also be used to study bridging-subset ontologies for interchange, federation and seeding of community ontologies. There will be a discussion of the strengths and weaknesses of various approaches tried by others, and a reflection on the usefulness of the system resulting from innovative work of this project.

Collaboration and Networked Technology

Journal of Computing in Teacher Education, Mar 10, 2014

AbstractThis article reports details of the implementation and results of an investigation into t... more AbstractThis article reports details of the implementation and results of an investigation into the use of the World Wide Web as a teaching tool in Computing Studies Methods (educational technology) at the University of Sydney (SU) and in Computer Literacy at the University of Adelaide (AU). Using the Web to learn about topics in educational technology, courses at both universities undertook a restructuring of content and delivery to examine the viability of the Web as a medium for collaborative learning through browsing, indexing, and publication. A student-centered delivery was negotiated at both SU and AU, with course leaders and laboratory demonstrators acting as facilitators. Students at SU were graduates with computer science training studying teaching methodology in Computing Studies, while those at AU were first-year students without computer science training studying Computer Literacy. At each university, students worked both individually and cooperatively in small groups to produce projects on a...

Genetic Programming and Evolvable Machines, Jun 9, 2017

This paper describes the MetaCompose music generator, a compositional, extensible framework for a... more This paper describes the MetaCompose music generator, a compositional, extensible framework for affective music composition. In this context 'affective' refers to the music generator's ability to express emotional information. The main purpose of MetaCompose is to create music in real-time that can express different mood-states, which we achieve through a unique combination of a graph traversal-based chord sequence generator, a search-based melody generator, a pattern-based accompaniment generator, and a theory for mood expression. Melody generation uses a novel evolutionary technique combining FI-2POP with multi-objective optimization. This allows us to explore a Pareto front of diverse solutions that are creatively equivalent under the terms of a multi-criteria objective function. Two quantitative user studies were performed to evaluate the system: one focusing on the music generation technique, and the other that explores valence expression, via the introduction of dissonances. The results of these studies demonstrate (i) that each part of the generation system improves the perceived quality of the music produced, and (ii) how valence expression via dissonance produces the perceived affective state. This system, which can reliably generate affect-expressive music, can subsequently be integrated in any kind of interactive application (e.g. games) to create an adaptive and dynamic soundtrack.

International Conference on Formal Concept Analysis, 2019

In this paper an application is presented that allows the construction of conjunctive queries (re... more In this paper an application is presented that allows the construction of conjunctive queries (represented as labeled graphs), the presentation of the results of the queries (as tables), and the selection of options for specialization or generalization of the graph queries. The graph labels correspond to the query vocabulary, which is obtained by conceptual scaling of the database. Each scale provides the specialization and generalization options for a particular kind of graph node. At any time, the user is shown the current query, the result table, and a set of options to modify the query.

MetaCompose is a music generator based on a hybrid evolutionary technique that combines FI-2POP a... more MetaCompose is a music generator based on a hybrid evolutionary technique that combines FI-2POP and multi-objective optimization. In this paper we employ the MetaCompose music generator to create music in real-time that expresses different mood-states in a game-playing environment (Checkers). In particular, this paper focuses on determining if differences in player experience can be observed when: (i) using affective-dynamic music compared to static music, and (ii) the music supports the game's internal narrative/state. Participants were tasked to play two games of Checkers while listening to two (out of three) different setups of game-related generated music. The possible setups were: static expression, consistent affective expression, and random affective expression. During game-play players wore a E4 Wristband, allowing various physiological measures to be recorded such as blood volume pulse (BVP) and electromyographic activity (EDA). The data collected confirms a hypothesis based on three out of four criteria (engagement, music quality, coherency with game excitement, and coherency with performance) that players prefer dynamic affective music when asked to reflect on the current game-state. In the future this system could allow designers/composers to easily create affective and dynamic soundtracks for interactive applications.

Proceedings of the Genetic and Evolutionary Computation Conference Companion, Jul 6, 2018

MetaCompose is a music generator based on a hybrid evolutionary technique combining FI-2POP and m... more MetaCompose is a music generator based on a hybrid evolutionary technique combining FI-2POP and multi-objective optimization. In this paper we employ the MetaCompose music generator to create music in real-time that expresses different mood-states in a gameplaying environment (Checkers) and present preliminary results of an experiment focusing on determining (i) if differences in player experience can be observed when using affective-dynamic music compared to static music; and (ii) if any difference is observed when the music supports the game's internal narrative/state. Participants were tasked to play two games of Checkers while listening to two (out of three) different setups of game-related generated music. The possible setups were: static expression, consistent affective expression, and random affective expression.

Constructing conceptual scales in formal concept analysis

Springer eBooks, 1998

ABSTRACT

This paper describes a work in progress on coevolving Artificial Neural Networks (ANNs) for music... more This paper describes a work in progress on coevolving Artificial Neural Networks (ANNs) for music improvisation. Using this neuro-evolutionary approach the ANNs adapt to the changes in the human player's music as input, while still maintaining some of the structure of the musical piece previously evolved. The system is called P R I M A L-I M P R O V and evolves modules that are composed of two ANNs, one controlling pitch and one controlling rhythm. The results of a quantitative study show that, by only introducing simple rules as fitness functions, the system is able to generate more interesting arrangements than ANNs evolved without a specific objective. The emerging and interesting musical patterns that are produced by the evolved ANNs hint at the promising potential of the system.

Proceedings of the Workshops on Fuzzy Logic and Fuzzy Control

Google, Inc. (search). ...

CONTEXT-DS 2015 : CONTEXT 2015 Doctorial Symposium ; Proceedings of Doctoral Symposium of the 9th International and Interdisciplinary Conference on Modeling and Using Context (CONTEXT 2015) ; Larnaca, Cyprus, 2 November, 2015

Lecture Notes in Computer Science, 2016

This paper describes a compositional, extensible framework for music composition and a user study... more This paper describes a compositional, extensible framework for music composition and a user study to systematically evaluate its core components. These components include a graph traversal-based chord sequence generator, a search-based melody generator and a pattern-based accompaniment generator. An important contribution of this paper is the melody generator which uses a novel evolutionary technique combining FI-2POP and multi-objective optimization. A participant-based evaluation overwhelmingly confirms that all current components of the framework combine effectively to create harmonious, pleasant and interesting compositions.

SN Computer Science/SN computer science, Mar 14, 2024

Neural Computing and Applications, May 31, 2022

IEEE Transactions on Computational Social Systems, 2021

arXiv (Cornell University), Jan 3, 2023

2p-D-7 遠心分離法により濃縮した単層カーボンナノチューブのラマン散乱

日本物理学会講演概要集年会, Mar 15, 1996

Springer eBooks, 1994

Lecture Notes in Computer Science, 2000

International Semantic Web Conference, 2013

Collaboration and Networked Technology

Journal of Computing in Teacher Education, Mar 10, 2014

Genetic Programming and Evolvable Machines, Jun 9, 2017

International Conference on Formal Concept Analysis, 2019

Proceedings of the Genetic and Evolutionary Computation Conference Companion, Jul 6, 2018

Constructing conceptual scales in formal concept analysis

Springer eBooks, 1998

ABSTRACT

Proceedings of the Workshops on Fuzzy Logic and Fuzzy Control

Google, Inc. (search). ...

Lecture Notes in Computer Science, 2016

Context-aware Applications, Location Privacy and Digital Ecosystems

Location-based services feature prominently in many digital ecosystem designs. In this paper we p... more Location-based services feature prominently in many digital ecosystem designs. In this paper we present the design of a mobility digital ecosystem for public transportation services in a campus area network. We describe four context-aware applications that encourage users to disclose their location: a public transport passenger tracking application; a route-based car-pooling application; an on-campus location-based social networking “assistant” and a virtual art-gallery guided tour. The location-based service applications offered are intended to encourage users to participate in a digital ecosystem and in so doing share data about their location. This location data can be used to provide value-added services to the users and optimize overall system behavior. In this talk, I present the architectures of these four applications and address issues concerning privacy, location-identity and uniform standards developed by the Internet Engineering Task Force (IETF). Our applications provide environments that encourage individual interaction and engagement that are self-regulating and self-evolving: key characteristics of a digital ecosystem.

Information retrieval, social tagging and the virtual museum of the pacific

The Virtual Museum of the Pacific is a social media platform enabling on-line user community enga... more The Virtual Museum of the Pacific is a social media platform enabling on-line user community engagement with the Pacific Collection of the Australian Museum. The success of the system depends on facilitating the development of culturally relevant folksonomies and encouraging a conversation between online communities. In this presentation we explore the relationships between stakeholders, folksonomy and taxonomy, to reveal the design strategies that inform this digital ecosystem. Given its capacity to span both collection management and community access issues, we maintain that the Virtual Museum of the Pacific is a significant model for online community interaction in the contemporary museum environment. The Virtual Museum of the Pacific is accessible at http://epoc.cs.uow.edu.au/vmp and you can watch a youtube presentation of it athttp://www.youtube.com/watch?v=YbSgKvWauP8

Digital Ecosystems

In this talk we present the idea of Web-based Digital Ecosystems as a metaphor for contemporary i... more In this talk we present the idea of Web-based Digital Ecosystems as a metaphor for contemporary information systems development. We motivate and explain the Digital Ecosystem metaphor. To this end, we demonstrate a combined Semantic Web and Web 2.0 approach to semi-automate the development of social media Web sites, in particular for Museum collections. The CollectionWeb framework has been applied in two Web-information systems, the Virtual Museum of the Pacific, developed for the Australian Museum and the Art Collection Ecosystem. Several other Web-based Digital Ecosystems are under development using the framework.

In CollectionWeb, digital artefacts of artistic or cultural significance are assembled in a Web-based content management system. A concept clustering technique called Formal Concept Analysis induces page impressions and drives semantic navigation. The object descriptions determine the navigation that can change with the user’s chosen perspective. All that need be done is for the Museum to describe an object to an adequate digital standard. In this way, leveraging existing metadata from a collection management system becomes a possibility for automatically generating Museum sites. Similarly, because association rules can be derived from Formal Concept Analysis, the sites provide a lightweight inference engine without semantic mark-up other than object metadata. The systems allow users to add their own tags to the objects, which in turn can influence the structure and the presentation of the collection. This self-organising capability is a key feature of CollectionWeb and is a key characteristic of the Digital Ecosystem metaphor in general. In doing so we present a compelling way for art institutions, museums and enthusiasts to present, organise and tag their collections.

The Virtual Museum of the Pacific

The Virtual Museum of the Pacific was developed by University of Wollongong Researchers in the Ce... more The Virtual Museum of the Pacific was developed by University of Wollongong Researchers in the Centre of Digital Ecosystems in collaboration with the Australian Museum. It uses rich media and metadata content from the Australian Museum's Pacific Collection. The result is a innovative collection content navigation and social media web site. Anyone is free to register and use the Virtual Museum of the Pacific at http://epoc.cs.uow.edu.au/vmp

The Gong Digital Bus Ecosystem

Passenger information systems can be achieved when there is a single integrated transport authori... more Passenger information systems can be achieved when there is a single integrated transport authority. Such systems are more difficult to implement when combined across multiple companies, transport types and providers. Even more challenging is the provision of services in a ticketless transport network. In the Digital Bus, connecting services, details of the next bus or train, local shops, businesses, hotels and other traveller-related information, are made available via GPS, wireless networks and personal mobile devices as well as in traditional forms such as signs and audible signals. In this seminar, I explain the challenge of planning services in a ticketless transport network using the Gong Shuttle and free University shuttle services as a case study for research on location-based transport services.

Web-based Digital Ecosystems

Managing Digital Libraries on the Web using Digital Ecosystems

This tutorial will present the fundamental metaphors of the Digital Ecosystems for managing Digi... more This tutorial will present the fundamental metaphors of the Digital Ecosystems for managing Digital Libraries on the Web and introduce domain suitable case studies illustrating the application of the opensource CollectionWeb framework for managing Digital Libraries on the Web. In addition it will:

* Examine the business and organisation context for a social media styles of digital library management
* Introduce the sources of data, structured and unstructured, formal and informal
* Examine the data requirements and issues in digital ecosystems used for digital library management, data capture and extensibility
* Present the techniques that deliver a solution to data integration issues for Web-based digital libraries
* Discuss a Service Oriented Architecture as Digital Ecosystem backbone for digital library management
* Introduce the issue of social media tagging as folksonomy maintenance for digital libraries
* Present an access control model that supports the solution to multiple folksonomies in digital libraries
* Examine user interface design and technical requirements issues
* Present evaluation methods for Web-based Digital Ecosystem performance measurement
* Discuss future trends in Digital Ecosystem for the digital library management on the Web

It will be presented by Peter Eklund, Professor of Information Systems and Technology and Director of the Centre for Digital Ecosystems at the University of Wollongong.

Digital Ecosystems

What do Facebook, Twitter, Amazon, Youtube, Flickr all have in common? They are all examples of W... more What do Facebook, Twitter, Amazon, Youtube, Flickr all have in common? They are all examples of Web-based Digital ecosystems. Whether you are writing a book review on Amazon or tagging photos in Flickr, Digital Ecosystems foster interaction, engagement and collaboration within a naturally sustainable environment. Understanding their power is going to be one of the major differential factors for any successful organisation in the future, so come along to our next session where Peter Eklund and Tim Wray from the University of Wollongong, will share their expertise and experiences including:

Digital Ecosystems, Social Media and Social Tagging
How have Digital Ecosystems evolved
Case Study: The Virtual Museum of the Pacific & The Art Collection Ecosystem
What’s in store for the future

This work explores the framework of Conceptual Knowledge Processing using Formal Concept Analysis... more This work explores the framework of Conceptual Knowledge Processing using Formal Concept Analysis (FCA) and Conceptual Graphs (CGs). We hypothesize that the theory of Power Context Families can be used to model data structures for Conceptual Graphs, which can in turn be used to extend Conceptual Knowledge Processing (in particular the FCA framework) to accommodate the more expressive knowledge representation of Conceptual Graphs. Furthermore, we hypothesize that Conceptual Graphs can be mapped onto a relational database model, using the modelled data structures.