Computers, Volume 10, Issue 8 (August 2021) – 15 articles

A lot of deep learning applications are desired to be run on mobile devices. Both accuracy and inference time are meaningful for a lot of them. While the number of FLOPs is usually used as a proxy for neural network latency, it may not be the best choice. In order to obtain a better approximation of latency, the research community uses lookup tables of all possible layers for the calculation of the inference on a mobile CPU. It requires only a small number of experiments. Unfortunately, on a mobile GPU, this method is not applicable in a straightforward way and shows low precision. In this work, we consider latency approximation on a mobile GPU as a data- and hardware-specific problem. Our main goal is to construct a convenient Latency Estimation Tool for Investigation (LETI) of neural network inference and building robust and accurate latency prediction models for each specific task. To achieve this goal, we make tools that provide a convenient way to conduct massive experiments on different target devices focusing on a mobile GPU. After evaluation of the dataset, one can train the regression model on experimental data and use it for future latency prediction and analysis. We experimentally demonstrate the applicability of such an approach on a subset of the popular NAS-Benchmark 101 dataset for two different mobile GPU. Full article

The use of gradient descent training to optimize the performance of a rule-fact network expert system via updating the network’s rule weightings was previously demonstrated. Along with this, four training techniques were proposed: two used a single path for optimization and two use multiple paths. The performance of the single path techniques was previously evaluated under a variety of experimental conditions. The multiple path techniques, when compared, outperformed the single path ones; however, these techniques were not evaluated with different network types, training velocities or training levels. This paper considers the multi-path techniques under a similar variety of experimental conditions to the prior assessment of the single-path techniques and demonstrates their effectiveness under multiple operating conditions. Full article

Currently, society is going through a health event with devastating results. In their desire to control the 2019 coronavirus disease, large organizations have turned over the execution of their activities to the use of information technology. These tools, adapted to the use of the Internet, have been presented as an effective solution to the measures implemented by the majority of nations where quarantines are generalized. However, the solution given by information technologies has several disadvantages that must be solved. The most important in this regard is with the serious security incidents that exist, where many organizations have been compromised and their data has been exposed. As a solution, this work proposes the design of a guide that allows for the implementation of a computer incident response team on a university campus. Universities are optimal environments for the generation of new technologies; they also serve as the ideal test bed for the generation of security policies and new treatments for incidents in an organization. In addition, with the implementation of the computer incident response team in a university, it is proposed to be part of a response group to any security incident at the national level. Full article

The article deals with the use of context-sensitive policies in the building of data acquisition systems in large scale distributed cyber-physical systems built on fog computing platforms. It is pointed out that the distinctive features of modern cyber-physical systems are their high complexity and constantly changing structure and behavior, which complicates the data acquisition procedure. To solve this problem, it is proposed to use an approach according to which the data acquisition procedure is divided into two phases: model construction and data acquisition, which allows parallel realization of these procedures. A distinctive feature of the developed approach is that the models are built in runtime automatically. As a top-level model, a multi-level relative finite state operational automaton is used. The automaton state is described using a multi-level structural-behavioral model, which is a superposition of four graphs: the workflow graph, the data flow graph, the request flow graph and the resource graph. To implement the data acquisition procedure using the model, the context-sensitive policy mechanism is used. The article discusses possible approaches to implementation of suggested mechanisms and describes an example of application. Full article

The increasing digitalization and advancement in information communication technologies has greatly changed how humans interact with digital information. Nowadays, it is not sufficient to only display relevant data in production activities, as the enormous amount of data generated from smart devices can overwhelm operators without being fully utilized. Operators often require extensive knowledge of the machines in use to make informed decisions during processes such as maintenance and production. To enable novice operators to access such knowledge, it is important to reinvent the way of interacting with digitally enhanced smart devices. In this research, a mobile augmented reality remote monitoring system is proposed to help operators with low knowledge and experience level comprehend digital twin data of a device and interact with the device. It analyses both historic logs as well as real-time data through a cloud server and enriches 2D data with 3D models and animations in the 3D physical space. A cloud-based machine learning algorithm is applied to transform learned knowledge into live presentations on a mobile device for users to interact with. A scaled-down case study is conducted using a tower crane model to demonstrate the potential benefits as well as implications when the system is deployed in industrial environments. This user study verifies that the proposed solution yields consistent measurable improvements for novice users in human-device interaction that is statistically significant. Full article

Non-governmental organizations (NGOs) in under-developed countries are receiving funds from donor agencies for various purposes, including relief from natural disasters and other emergencies, promoting education, women empowerment, economic development, and many more. Some donor agencies have lost their trust in NGOs in under-developed countries, as some NGOs have been involved in the misuse of funds. This is evident from irregularities in the records. For instance, in education funds, on some occasions, the same student has appeared in the records of multiple NGOs as a beneficiary, when in fact, a maximum of one NGO could be paying for a particular beneficiary. Therefore, the number of actual beneficiaries would be smaller than the number of claimed beneficiaries. This research proposes a blockchain-based solution to ensure trust between donor agencies from all over the world, and NGOs in under-developed countries. The list of National IDs along with other keys would be available publicly on a blockchain. The distributed software would ensure that the same set of keys are not entered twice in this blockchain, preventing the problem highlighted above. The details of the fund provided to the student would also be available on the blockchain and would be encrypted and digitally signed by the NGOs. In the case that a record inserted into this blockchain is discovered to be fake, this research provides a way to cancel that record. A cancellation record is inserted, only if it is digitally signed by the relevant donor agency. Full article

Being able to understand, express, and communicate emotions is widely recognized as a fundamental competence. For the younger generation entering the professional market, this is particularly relevant as, in this context, emotions are managed and communicated in ways (and channels) that are different from what they are used to and that can easily lead to misunderstandings. Therefore, it is important to analyze how young people deal with, understand, and interpret emotions, particularly in the context of a professional career where the ability to dialogue with different people and how to get around problems in a healthy and resilient way is essential. This analysis will allow one to design and create tools that allow the younger generation to improve their emotional intelligence and competence. This article introduces the League of Emotions Learners (LoEL) project, an innovative initiative that, through a game app, develops the emotional competence and intelligence of young people. The article then presents the results obtained in the initial validation that led to the positive understanding of its impact. Full article

In this article, we test the hypothesis that computer science papers asking questions (i.e., those with a question mark at the end of their title) are cited more frequently than those that do not have this property. To this end, we analyze a data set of almost two million records on computer science papers indexed in the Web of Science database and focus our investigation on the mean number of citations per paper of its specific subsets. The main finding is that the average number of citations per paper of the so-called “asking papers” is greater by almost 20% than that of other papers, and that this difference is statistically significant. Full article

This paper presents a comparative analysis of four semi-supervised machine learning (SSML) algorithms for detecting malicious nodes in an optical burst switching (OBS) network. The SSML approaches include a modified version of K-means clustering, a Gaussian mixture model (GMM), a classical self-training (ST) model, and a modified version of self-training (MST) model. All the four approaches work in semi-supervised fashion, while the MST uses an ensemble of classifiers for the final decision making. SSML approaches are particularly useful when a limited number of labeled data is available for training and validation of the classification model. Manual labeling of a large dataset is complex and time consuming. It is even worse for the OBS network data. SSML can be used to leverage the unlabeled data for making a better prediction than using a smaller set of labelled data. We evaluated the performance of four SSML approaches for two (Behaving, Not-behaving), three (Behaving, Not-behaving, and Potentially Not-behaving), and four (No-Block, Block, NB- wait and NB-No-Block) class classifications using precision, recall, and F1 score. In case of the two-class classification, the K-means and GMM-based approaches performed better than the others. In case of the three-class classification, the K-means and the classical ST approaches performed better than the others. In case of the four-class classification, the MST showed the best performance. Finally, the SSML approaches were compared with two supervised learning (SL) based approaches. The comparison results showed that the SSML based approaches outperform when a smaller sized labeled data is available to train the classification models. Full article

Convolutional neural networks raised the bar for machine learning and artificial intelligence applications, mainly due to the abundance of data and computations. However, there is not always enough data for training, especially when it comes to historical collections of cultural heritage where the original artworks have been destroyed or damaged over time. Transfer Learning and domain adaptation techniques are possible solutions to tackle the issue of data scarcity. This article presents a new method for domain adaptation based on Knowledge graph embeddings. Knowledge Graph embedding forms a projection of a knowledge graph into a lower-dimensional where entities and relations are represented into continuous vector spaces. Our method incorporates these semantic vector spaces as a key ingredient to guide the domain adaptation process. We combined knowledge graph embeddings with visual embeddings from the images and trained a neural network with the combined embeddings as anchors using an extension of Fisher’s linear discriminant. We evaluated our approach on two cultural heritage datasets of images containing medieval and renaissance musical instruments. The experimental results showed a significant increase in the baselines and state-of-the-art performance compared with other domain adaptation methods. Full article

Wireless Multimedia Sensor Networks (WMSNs) based on IEEE 802.11 mesh networks are effective and suitable solutions for video surveillance systems in detecting intrusions in selected monitored areas. The IEEE 802.11-based WMSNs offer high bit rate video transmissions but are challenged by energy inefficiency issues and concerns. To resolve the energy inefficiency challenges, the salient research studies proposed a hybrid architecture. This newly evolved architecture is based on the integration of IEEE 802.11-based mesh WMSNs along with the LoRa network to form an autonomous and high bitrate, energy-efficient video surveillance system. This paper proposes an energy-aware and Quality of Service (QoS) routing mechanism for mesh-connected visual sensor nodes in a hybrid Internet of Things (IoT) network. The routing algorithm allows routing a set of video streams with guaranteed bandwidth and limited delay using as few visual sensor nodes as possible in the network. The remaining idle visual sensor nodes can be turned off completely, and thus it can significantly minimize the overall energy consumption of the network. The proposed algorithm is numerically simulated, and the results show that the proposed approach can help in saving a significant amount of energy consumption while guaranteeing bandwidth and limited delay. Full article

The emergence of the Learning Analytics (LA) field contextualised the connections in various disciplines and the educational sector, acted as a steppingstone toward the reformation of the educational scenery, thus promoting the importance of providing users with adaptive and personalised learning experiences. At the same time, the use of Augmented Reality (AR) applications in education have been gaining a growing interest across all the educational levels and contexts. However, the efforts to integrate LA techniques in immersive technologies, such as AR, are limited and scarce. This inadequacy is mainly attributed to the difficulties that govern the collection and interpretation of the primary data. To deal with this shortcoming, we present the “Augmented Reality Learning Analytics” (ARLEAN) ethical framework, tailored to the specific characteristics that AR applications have, and focused on various learning subjects. The core of this framework blends the technological, pedagogical, and psychological elements that influence the outcome of educational interventions, with the most widely adopted LA techniques. It provides concrete guidelines to educational technologists and instructional designers on how to integrate LA into their practices to inform their future decisions and thus, support their learners to achieve better results. Full article

Archives, both analogue and digital, are primarily concerned with preserving records as originals. Because of this, immutable data as used in a blockchain data structure seem a logical choice when designing such systems. At the same time, archives maintain records which may need to change over the long term. It is a requirement of archival preservation to be able to update records’ metadata in order not only to guarantee authenticity after digital preservation actions but also to ensure that relationships to other records, which might be created after an original record has entered the archive (and has been registered in a blockchain), can be maintained. The need to maintain an archival bond, which represents a network of relationships between aggregation of records, i.e., the relationship connecting previous and subsequent records belonging to the same activity, is a prime example of this requirement. This paper explores realisation of the archival bond in the context of blockchain-based archival system by proposing a supporting database system which enables metadata to be changed as required but also significantly simplifies searching compared to searching on-chain information, while keeping the immutability characteristic of blockchain. Full article

In 2019, the National Archives of Korea (NAK) developed a blockchain recordkeeping platform to conduct R&D on recordkeeping approaches. This paper introduces two types of R&D studies that have been conducted thus far. The first is the use of blockchain transaction audit trail technology to ensure the authenticity of audiovisual archives, i.e., the application of blockchain to a new system. The second uses blockchain technology to verify whether the datasets of numerous information systems built by government agencies are managed without forgery or tampering, i.e., the application of blockchain to an existing system. Government work environments globally are rapidly shifting from paper records to digital. However, the traditional recordkeeping methodology has not adequately kept up with these digital changes. Despite the importance of responding to digital changes by incorporating innovative technologies such as blockchain in recordkeeping practices, it is not easy for most archives to invest funds in experiments on future technologies. Owing to the Korean government’s policy of investing in digital transformation, NAK’s blockchain recordkeeping platform has been developed, and several R&D tasks are underway. Hopefully, the findings of this study will be shared with archivists around the world who are focusing on the future of recordkeeping. Full article