volume-17
Latest Articles
Some Results on Colored Network Contraction
JUSPN, volume-17 , Issue 2 (2022), PP 91 - 98
Published: 20 Dec 2022
DOI: 10.5383/JUSPN.17.02.006
by Flavio Lombardi, Elia Onofri from Istituto per le Applicazioni del Calcolo, Consiglio Nazionale delle Ricerche (IAC-CNR), Rome (00185), Italy, Dipartimento di Matematica e Fisica, Roma Tre University, L. S. Murialdo, 1, Rome (00146), Italy and Member of the INdAM-GNCS research group
Abstract: Networks are pervasive in computer science and in real world applications. It is often useful to leverage distinctive node features to regroup such data in clusters, by making use of a single representative node per cluster. Such contracted graphs can help identify features of the original networks that were not visible before. As an example, we can identify contiguous nodes having the same discrete property in a social network. Contracting a graph allows a more scalable analysis of the interactions and structure of the network nodes. This paper delves into the problem of contracting possibly large colored networks into smaller and more easily manageable representatives. It also describes a simple but effective algorithm to perform this task. Extended performance plots are given for a range of graphs and results are detailed and discussed with the aim of providing useful use cases and application scenarios for the approach. read more... read less...
Keywords: Colored Networks, Graph Contraction, Greedy Algorithm, Graph Analysis
An Architecture for Cognitive Computing in Healthcare
JUSPN, volume-17 , Issue 2 (2022), PP 83 - 90
Published: 20 Dec 2022
DOI: 10.5383/JUSPN.17.02.005
by Ronald Tombe, F Mzee Awuor, Serestina Viriri from Kisii University, Kisii, Kenya and University of KwaZulu-Natal, Durban, South Africa
Abstract: The integrated impact of computing techniques and resources with big-data processing transforms human lifestyles by providing quality services ranging from healthcare to smart homes and effective interactions. However, many healthcare systems fail to consider patient emergencies and cannot provide a customized resource service. Cognitive computing is a requisite technology to create these intelligent systems based on artificial intelligence algorithms. This paper presents technologies for personalized healthcare services through cognitive computing. This paper investigates cognitive computing developments from discovering knowledge, cognitive science, and big-data analytics at the onset. Then, the system architecture for a cognitive computing system is given. Furthermore, this paper presents the technologies for cognitive computing healthcare improvement opportunities and their challenges. Finally, this paper discusses the representative intelligent systems of cognitive computing, including medical, robotic, and cognitive-communication systems. read more... read less...
Keywords: cognitive computing, big-data, healthcare, machine learning, pervasive computing, ubiquitous systems
Design and Specification of a Privacy-Preserving Registration for Blockchain-Based Energy Markets
JUSPN, volume-17 , Issue 2 (2022), PP 73 - 81
Published: 20 Dec 2022
DOI: 10.5383/JUSPN.17.02.004
by Michell Boerger, Philipp Lämmel, Nikolay Tcholtchev, and Manfred Hauswirth from Fraunhofer Institute for Open Communication Systems (FOKUS), Berlin, Germany and Technical University of Berlin, Berlin, Germany
Abstract: The challenges of climate change and the related demand to integrate non-plannable and weather-dependent renewable energy resources pose enormous challenges for the entire energy domain, e.g. in the context of grid control. These challenges reveal the need for new technical solutions and new business models while they indicate the required and inevitable transition to smart grids. Many blockchain-based solutions are being discussed in this context, ranging from peer-to-peer energy trading to grid-serving applications. However, especially in connection with public blockchains, clear security privacy challenges arise since the security and privacy of private data must be guaranteed while traceability must be avoided. Therefore, in this paper, we will specify privacy-protecting registration processes for blockchain-based flexibility markets that enable pseudonymous access to the latter. Furthermore, in collaboration with a governmental regulating institution named DGA, we will show that using an existing X.509-based PKI and RSA-based cryptographic processes, the integrity of all market participants can be guaranteed. This integrity is essential for the security-critical use of operating reserve. In addition, we will evaluate the specified processes in terms of efficiency, scalability, security, and privacy protection. read more... read less...
Keywords: Blockchain, privacy, security, encryption, distributed ledger, energy market
Low Rank Graph Regularization Embedding for 2D+3D Facial Expression Recognition
JUSPN, volume-17 , Issue 2 (2022), PP 67 - 72
Published: 14 Dec 2022
DOI: 10.5383/JUSPN.17.02.003
by Yunfang Fu, Yujuan Deng ,Yuekui Zhang, Zhengyan Yang, Ruili Qi from School of Computer Science & Engineering, Shijiazhuang University, Shijiazhuang, China, 050035 and Hebei Province Internet of Things Intelligent Perception and Application Technology Innovation Center,Shijiazhuang, China, 050035
Abstract: In this paper, a novel low rank graph regularization embedding for 3D facial expression recognition (LRGREFER) approach is proposed, in which the core tensor is utilized to characterize the low-rank attribute among the samples combined with the factor matrices with the graph regularization embedding. At first, a model based on a 4D tensor is constructed from the facial expression data. By Tucker decomposing the constructed 4D tensor, a resulting core tensor and factor matrices in different tensor modes are utilized to characterize the low-rankness among samples. Because of the loss of information during modelling the 4D tensor, the missing data from partly observed facial expression data are recovered by embedding the tensor completion. Finally, the proposed model is handled and solved by adopting the alternating direction method of multipliers (ADMM). Meanwhile, the classification prediction of facial expressions are implemented by Multi-class-SVM. Numerical experiments are conducted on BU-3DFE database. The experiment results have been verified that our proposed approach is more competitive. read more... read less...
Keywords: Low rank, Graph regularization embedding, Facial expression recognition, Tucker decomposition
An Optimized Kappa Architecture for IoT Data Management in Smart Farming
JUSPN, volume-17 , Issue 2 (2022), PP 59 - 65
Published: 07 Dec 2022
DOI: 10.5383/JUSPN.17.02.002
by Jean Bertin Nkamla Penka, Saïd Mahmoudi, Olivier Debauche from University of Mons, Faculty of Engineering - ILIA Lab / InforTech, Place du parc 20, Mons, Belgium, 7000, University of Liège - GxABT, Terra, Passage des Déportés 2, Gembloux, Belgium, 5030, University of Liège - GxABT, BioDynE - DEAL, Passage des Déportés 2, Gembloux, Belgium, 5030
Abstract: Agriculture 4.0 is a domain of IoT in full growth which produces large amounts of data from machines, robots, and sensors networks. This data must be processed very quickly, especially for the systems that need to make real-time decisions. The Kappa architecture provides a way to process Agriculture 4.0 data at high speed in the cloud, and thus meets processing requirements. This paper presents an optimized version of the Kappa architecture allowing fast and efficient data management in Agriculture. The goal of this optimized version of the classical Kappa architecture is to improve memory management and processing speed. the Kappa architecture parameters are fine tuned in order to process data from a concrete use case. The results of this work have shown the impact of parameters tweaking on the speed of treatment. We have also proven that the combination of Apache Samza with Apache Druid offers the better performances read more... read less...
Keywords: Agriculture 4.0, IoT, Internet of Things, Kappa Architecture, Smart Farming, Smart Agriculture
New and Reliable Points Shifting - Based Algorithm for Indoor Location Services
JUSPN, volume-17 , Issue 2 (2022), PP 51 - 58
Published: 07 Dec 2022
DOI: 10.5383/JUSPN.17.02.001
by Tarek El Salti, Nelson Shaw, Joseph Chun-Chung Cheung, Edward R. Sykes from School of Applied Computing, Sheridan College, Oakville, Ontario, Canada, L6H 2L1 TELUS Communication Inc., 200 Consilium Place, Scarborough, Ontario, Canada, M1H 3E4 Centre for Mobile Innovation, Sheridan College, Oakville, Ontario, Canada, L6H 2L1
Abstract: Indoor localization is of great importance to several fields such as healthcare and asset tracking. However, many factors (e.g., multipath propagations) impact the quality of signals which are used to perform localizations. As a consequence, the precision and accuracy of the computed locations are heavily influenced. Therefore, the methodologies to compute indoor locations always need continuous refinements in terms of those metrics including the time complexity. For the last metric, It impacts the performance of mobile devices due to their limited resources. To address these challenges, a new set of fingerprinting algorithms was presented in this paper called Fingerprinting Line-Based Nearest Neighbour. This set shifts grid points potentially towards targets via a deterministic percentage. The running time of the set is upper bounded. Moreover, this paper presents the following: 1) an upper bound in terms of distance errors for the proposed algorithms, and 2) based on real experiments, the new algorithms (e.g., 90% shifting) improved the accuracy and precision, and had lower distance errors probabilities compared to those for the nearest neighbour-based algorithms (e.g., by 106% and 76%, respectively). read more... read less...
Keywords: : Indoor Location Services, Fingerprinting, Wi-Fi, Path Loss exponent, K-Nearest Neighbor
Towards Low-Cost IoT and LPWAN-Based Flood Forecast and Monitoring System
JUSPN, volume-17 , Issue 1 (2022), PP 43 - 49
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.006
by Nassima Tadrist, Olivier Debauche, Saïd Mahmoudi, Adriano Guttadauria from University of Blida 1 Saâd Dahlab, Department of Water and Environmental Sciences, BP 270 Route Soumâa, Blida, Algeria, 09000 , University of Liège, TERRA / BioDynE - DEAL, Passage des Déportés 2, Gembloux, Belgium, 5030, University of Mons, Infortech Institute / Faculty of Engineering - ILIA, Place du Parc 20, Mons, Belgium, 7000
Abstract: t The recent floods have shown that the classic monitoring systems for watercourses are no longer adapted because other phenomena such as the insufficient capacity and/or obstruction of drainage networks, the modification of cultivation practices and rotations, the increase in the size of plots linked to the reparcelling, the urbanization of floodable areas, etc. The combination of all these causes, plus the modification of the water regime, implies an increase in the risk of flooding and an adapted monitoring that is no longer limited to watercourses in order to give early warning of the risk of flooding by runoff. The Internet of Things (IoT) and the availability of microcontrollers and sensors with low data rates and long ranges, as well as low-power wide area networks (LPWANs), allow for much more advanced monitoring systems. read more... read less...
Keywords: Monitoring System, Warning System, Flood, Runoff, Moody Flood, Flash Flood, Runoff Flooding, Sewer, IoT, LPWAN, LoRaWan, NB-IoT
Towards Performance of NLP Transformers on URL-Based Phishing Detection for Mobile Devices
JUSPN, volume-17 , Issue 1 (2022), PP 35 - 42
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.005
by Hossein Shirazi, Katherine Haynes, Indrakshi Ray from Computer Science Department, Colorado State University, Fort Collins, CO, USA and Cooperative Institute for Research in the Atmosphere, Colorado State University, Fort Collins, CO, USA
Abstract: Hackers are increasingly launching phishing attacks via SMS and social media. Games and dating apps introduce yet another attack vector. However, current deep learning-based phishing detection applications do not apply to mobile devices due to the computational burden. We propose a lightweight phishing detection algorithm that distinguishes phishing from legitimate websites solely from URLs to be used in mobile devices. As a baseline performance, we apply Artificial Neural Networks (ANNs) to URL-based and HTML-based website features. A model search results in 15 ANN models with accuracies >96%, comparable to state-of-the-art approaches. Next, we test the performance of deep ANNs on URLbased features only; however, all models perform poorly with the highest accuracy of 86.2%, indicating that URL-based features alone are not adequate to detect phishing websites even with deep ANNs. Since language transformers learn to represent context-dependent text sequences, we hypothesize that they will be able to learn directly from the text in URLs to distinguish between legitimate and malicious websites. We apply three state-of-the-art deep transformers (BERT, ELECTRA, and RoBERTa) for phishing detection. Testing custom and standard vocabularies, we find that pre-trained transformers available for immediate use (with fine-tuning) outperform the model trained with the custom URL-based vocabulary. In addition, we test a thinner BERT transformer which is suitable for lightweight devices like mobiles, called MobileBERT. Our results emphasize that evaluation metrics of this model are competitive to other models in this study, yet the testing time is significantly less, making this model a choice for embedding phishing detection algorithms in mobile phones. Using pre-trained transformers to predict phishing websites from only URLs has five advantages: 1) requires little training time (230 to 320 s), 2) is more easily updatable than feature-based approaches because no pre-processing of URLs is required, 3) is safer to use because phishing websites can be predicted without physically visiting the malicious sites, 4) is easily deployable for real-time detection and is applicable to run on mobile devices, and 5) using a mobile specific transformer yields comparable performance and predicts 3 times faster than the other transformer models in this study. read more... read less...
Keywords: Social Engineering Attack, Phishing Detection, Deep learning, Transformers, Mobile Application
The way it made me feel – Creating and evaluating an in-app feedback tool for mobile apps
JUSPN, volume-17 , Issue 1 (2022), PP 27 - 34
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.004
by Simon André Scherr, Mher Ter-Tovmasyan, Frauke Neugebauer, Steffen Hupp, Frank Elberzhager from Fraunhofer IESE, Kaiserslautern, Germany, 67663
Abstract: Mobile apps are becoming increasingly important in everyone's daily life. The success of an app is linked to high user acceptance. Therefore, it is necessary to capture users' expectations, needs, and problems regarding an app in any situation. By continuously capturing and analyzing user feedback, developers can evaluate the level of user acceptance. There are various feedback channels, such as app stores, social networks, and within the app, which can be used to capture user feedback. As we already have experience with feedback from app stores and social networks, we wanted to investigate inapp feedback approaches and thus conducted a mapping study to understand the state of the art of these approaches.We analyzed 36 publications and derived requirements for in-app feedback tools. Based on that, we defined requirements for an in-app feedback tool to describe its prototypical realization. Then we performed an evaluation regarding user acceptance of our tool with 33 participants. The evaluation showed a high rate of acceptance for the tool among the participants. The results also highlighted improvement areas for our tool, such as optimizing the rate of requests for feedback. We plan to address these aspects in future work and to continue improving our tool. read more... read less...
Keywords: Mobile Apps, Software Quality, User Feedback, User Experience, In-App Feedback
From Collective Memory to Map Services
JUSPN, volume-17 , Issue 1 (2022), PP 19 - 26
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.003
by Konstantinos Koukoulis, Dimitrios Koukopoulos from University of Patras, Department of History and Archaeology, Agrinio, Greece and Hellenic Open University, Patras, Greece
Abstract: The route followed by a refugees’ group towards its destination can, in many cases, be regarded as the reference point around which the collective memory of such a group of people is intertwined. Such a route enriches people's memories with common experiences, targeting places and interactions among refugees and locals and may affect the collective memory of such people positively or negatively. A crucial point in the modern paradigm of smart cities is the quality of life. To achieve quality of life for its citizens a smart city should establish ways to reduce alienation among the different groups that constitute the city's palimpsest. Understanding the different cultural identities and improvement of social cohesion between different people groups is one of the basic vehicles towards this goal. In this paper, we attempt to give a first answer to such problems proposing and implementing specific services in the context of a crowdsourcing system for collective memory management using interactive maps. We demonstrate a basic usage scenario to show the strength of the implemented services, along with a two-step evaluation showing positive results. read more... read less...
Keywords: Crowdsourcing, Collective Memory Management, Big Data, Mobile Services and Platforms
Small Towns and Regional Municipalities Implement SMART Solutions, Identified Issues, and Challenges
JUSPN, volume-17 , Issue 1 (2022), PP 11 - 18
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.002
by Peter Balco, Dorota Košecká, Peter Bajzík from Comenius University, Faculty of Management, Odbojárov 10, P. O. Box 95, Bratislava, 82005, Slovakia, ATOS IT Solutions and Services s.r.o., Pribinova 19, Bratislava Slovakia
Abstract: In the last decade, SMART services and solutions projects have been concentrated mainly in large and economically strong cities where large populations are concentrated. This is place where the potential is found that predicts return on investment as well as further development. As not all cities are predestined for this type of project, we were interested in how small towns and cities perceive their potential to engage in the implementation of such projects. We believe that the topic of SMART solutions should not be a significant priority only for large cities. We decided to analyze the needs of small cities in terms of implementing SMART solutions. We also tried to identify the challenges as well as the requirements to accelerate this process. In our analysis, we focused on the Slovak Republic, which is a good candidate for such research due to its structure of cities and municipalities. In the process of data collection, we approached more than 2,744 s mall towns and municipalities with a population of up to 5,000 with a request for information, and 547 town and municipality representatives responded to the questionnaire. The results of the research show an interesting and clear finding, s mall towns and rural areas also want SMART. In the research, we identified several not simple problems that need to be solved for the successful implementation of these goals read more... read less...
Keywords: SMART CITY, clusters, SMART Services, SMART Villages, SMART regions
Remote Collaboration Needs for New Work: Concepts, Procedure and Evaluation
JUSPN, volume-17 , Issue 1 (2022), PP 01 - 09
Published: 04 Aug 2022
DOI: 10.5383/JUSPN.17.01.001
by Sven Storck, Kathleen Späth, Claudia Nass Bauer, Frank Elberzhager from Fraunhofer IESE, Kaiserslautern, Germany, 67663
Abstract: More freedom, more flexibility, and reduced travel time for knowledge workers are just a few advantages of new work models, which have been discussed for several years now. Moreover, the problem of rural depopulation can be addressed by this concept. In the research project “Digital Teams”, we aim to develop a digital open-source platform to support and optimize the digital work environment for distributed teams in rural areas, especially in the knowledge work context. In this article, we focus on the research and design aspects of the project. We provide insights on how we have used the design thinking approach for our research and the development of the UX- and UI-design concepts. We are focusing on an ecosystem concept, which provides all relevant services for knowledge workers in their daily work life, rather than focusing on a specific remote collaboration purpose. We present initial evaluation results, which tend to be positive and give an outlook on future work. read more... read less...
Keywords: New Work, Design Thinking, Collaboration, Groupware, Rural Depopulation