This paper focuses on the potential value and future prospects of using virtual reality (VR) technology in online education. In detailing online education and the latest VR technology, we focus on metaverse construction and artificial intelligence (AI) for educational VR use. In particular, we describe a virtual university campus in which on-demand VR lectures are conducted in virtual lecture halls, automated evaluations of student learning and training using machine learning, and the linking of multiple digital campuses.

During the COVID-19 pandemic there was a rapid shift to emergency remote teaching practices and online tools for education have already gained further attention. While eLearning initiatives are developed and its implementation at scale are widely discussed, this research focuses on the utilization of data which can be logged in such eLearning systems. We demonstrate the need and potential of utilizing learning logs to create services supporting sustainable quality improvement of education. Learning and Evidence Analytics Framework (LEAF), is the overarching technology framework with affordances to adopt evidence-based practices for education. It aims to promote learning for all by introducing data-driven services for personalized approaches.

Self-directed learning in an appropriately designed environment can help learners retain knowledge tied to experience and motivate them to learn more. For teachers, however, it is difficult to design an environment to give to learners and to give feedback that reflects respect for their independent efforts, while for learners, it is difficult to set learning objectives on their own and to construct knowledge correctly based on their own efforts. In this research, we developed a learning support system that provides a mechanism for constructing an observational learning environment using virtual space and that encourages self-directed knowledge discovery. We confirmed that this system contributes to a learner's structural understanding and its retention and to a greater desire to learn at a level comparable to that of concept map creation, another active learning method.

This study examined whether distance learning in a first-year PBL courses in the first unit of instruction improves the effectiveness of subsequent group work learning over face-to-face learning. The first-year PBL consisted of three units: an input unit, a group work unit and an outcomes presentation unit. In 2017/2018, the input unit was conducted in the classroom with face-to-face learning. In 2017, a workshop was held in addition to face-to-face learning in classroom. In 2020/2021, the input unit was conducted with distance learning. In the years, approximately 100 people completed the questionnaire. A preliminary check confirmed that the average score of students' self-assessment of their own social skills were not significantly different among the four years. Analysis showed that in 2018, the perceived efficacy in the group work unit depended on learners' high social skills. Alternatively, in 2017/2020/2021, the perceived efficacy in group work was not dependent on learners' social skills. This suggests that distance learning and face-to-face learning with workshop learning, instead of full face-to-face learning for the units placed before the group work unit facilitates the learning efficacy of the group work unit, even for students with social skill concerns.

This study focuses on creating arithmetical stories as a sub-task of problem posing and proposes a game named “Tri-prop scrabble” as a learning environment based on a fusion method of learning and game. The problem-posing ability has a positive relationship with mathematics achievement and understanding the mathematical structure of problems. In the proposed game, learners are expected to experience creating and concatenating various arithmetical stories by integrating simple sentences. The result of a preliminary feasibility study shows that the participants were able to pose and concatenate a variety of types of arithmetic stories and accept this game is helpful for learning arithmetic word problems.

The package manager (PM) is crucial to most technology stacks, acting as a broker to ensure that a verified dependency package is correctly installed, configured, or removed from an application. Diversity in technology stacks has led to dozens of PMs with various features. While our recent study indicates that package management features of PM are related to end-user experiences, it is unclear what those issues are and what information is required to resolve them. In this paper, we have investigated PM issues faced by end-users through an empirical study of content on Stack Overflow (SO). We carried out a qualitative analysis of 1,131 questions and their accepted answer posts for three popular PMs (i.e., Maven, npm, and NuGet) to identify issue types, underlying causes, and their resolutions. Our results confirm that end-users struggle with PM tool usage (approximately 64-72%). We observe that most issues are raised by end-users due to lack of instructions and errors messages from PM tools. In terms of issue resolution, we find that external link sharing is the most common practice to resolve PM issues. Additionally, we observe that links pointing to useful resources (i.e., official documentation websites, tutorials, etc.) are most frequently shared, indicating the potential for tool support and the ability to provide relevant information for PM end-users.

Contemporary development projects benefit from code review as it improves the quality of a project. Large ecosystems of inter-dependent projects like OpenStack generate a large number of reviews, which poses new challenges for collaboration (improving patches, fixing defects). Review tools allow developers to link between patches, to indicate patch dependency, competing solutions, or provide broader context. We hypothesize that such patch linkage may also simulate cross-collaboration. With a case study of OpenStack, we take a first step to explore collaborations that occur after a patch linkage was posted between two patches (i.e., cross-patch collaboration). Our empirical results show that although patch linkage that requests collaboration is relatively less prevalent, the probability of collaboration is relatively higher. Interestingly, the results also show that collaborative contributions via patch linkage are non-trivial, i.e, contributions can affect the review outcome (such as voting) or even improve the patch (i.e., revising). This work opens up future directions to understand barriers and opportunities related to this new kind of collaboration, that assists with code review and development tasks in large ecosystems.

Defect prediction approaches have been greatly contributing to software quality assurance activities such as code review or unit testing. Just-in-time defect prediction approaches are developed to predict whether a commit is a defect-inducing commit or not. Prior research has shown that commit-level prediction is not enough in terms of effort, and a defective commit may contain both defective and non-defective files. As the defect prediction community is promoting fine-grained granularity prediction approaches, we propose our novel class-level prediction, which is finer-grained than the file-level prediction, based on the files of the commits in this research. We designed our model for Python projects and tested it with ten open-source Python projects. We performed our experiment with two settings: setting with product metrics only and setting with product metrics plus commit information. Our investigation was conducted with three different classifiers and two validation strategies. We found that our model developed by random forest classifier performs the best, and commit information contributes significantly to the product metrics in 10-fold cross-validation. We also created a commit-based file-level prediction for the Python files which do not have the classes. The file-level model also showed a similar condition as the class-level model. However, the results showed a massive deviation in time-series validation for both levels and the challenge of predicting Python classes and files in a realistic scenario.

This paper focuses on the “data collection period” for training a better Just-In-Time (JIT) defect prediction model — the early commit data vs. the recent one —, and conducts a large-scale comparative study to explore an appropriate data collection period. Since there are many possible machine learning algorithms for training defect prediction models, the selection of machine learning algorithms can become a threat to validity. Hence, this study adopts the automatic machine learning method to mitigate the selection bias in the comparative study. The empirical results using 122 open-source software projects prove the trend that the dataset composed of the recent commits would become a better training set for JIT defect prediction models.

Network-on-Chips (NoCs) are important components for scalable many-core processors. Because the performance of parallel applications is usually sensitive to the latency of NoCs, reducing it is a primary requirement. In this study, a compression router that hides the (de)compression-operation delay is proposed. The compression router (de)compresses the contents of the incoming packet before the switch arbitration is completed, thus shortening the packet length without latency penalty and reducing the network injection-and-ejection latency. Evaluation results show that the compression router improves up to 33% of the parallel application performance (conjugate gradients (CG), fast Fourier transform (FT), integer sort (IS), and traveling salesman problem (TSP)) and 63% of the effective network throughput by 1.8 compression ratio on NoC. The cost is an increase in router area and its energy consumption by 0.22mm² and 1.6 times compared to the conventional virtual-channel router. Another finding is that off-loading the decompressor onto a network interface decreases the compression-router area by 57% at the expense of the moderate increase in communication latency.

We present a mathematical formulation of a trust metric using a quality and quantity pair. Under a certain assumption, we regard trust as an additive value and define the soundness of a trust computation as not to exceed the total sum. Moreover, we point out the importance of not only soundness of each computed trust but also the stability of the trust computation procedure against changes in trust value assignment. In this setting, we define trust composition operators. We also propose a trust computation protocol and prove its soundness and stability using the operators.

A new method for hiding information in digital images is proposed. Our method differs from existing techniques in that the information is hidden in a mixture of colors carefully tuned on a specific device according to the device's signal-to-luminance (gamma) characteristics. Because these reproduction characteristics differ in general from device to device and even from model to model, the hidden information appears when the cover image is viewed on a different device, and hence the hiding property is device-dependent. To realize this, we modulated a cover image using two identically-looking checkerboard patterns and switched them locally depending on the hidden information. Reproducing these two patterns equally on a different device is difficult. A possible application of our method would be secure printing where an image is allowed to be viewed only on a screen but a warning message appears when it is printed.

Currently, NDN-based VANETs protocols have several problems with packet overhead of rebroadcasting, control packet, and the accuracy of next-hop selection due to the dynamic topology. To deal with these problems in this paper, we propose a robust and lightweight forwarding protocol in Vehicular ad-hoc Named Data Networking. The concept of our forwarding protocol is adopting a packet-free approach. A vehicle collects its neighbor's visual identification by a pair of cameras (front and rear) to assign a unique visual ID for each node. Based on these IDs, we construct a hop-by-hop FIB-based forwarding strategy effectively. Furthermore, the Face duplication [1] in the wireless environment causes an all-broadcast problem. We add the visual information to Face to distinguish the incoming and outgoing Face to prevent broadcast-storm and make FIB and PIT work more accurate and efficiently. The performance evaluation results focusing on the communication overhead show that our proposal has better results in overall network traffic costs and Interest satisfaction ratio than previous works.

Explainable artificial intelligence (AI) technology enables us to quantitatively analyze the whole prediction logic of AI as a global explanation. However, unwanted relationships learned by AI due to data sparsity, high dimensionality, and noise are also visualized in the explanation, which deteriorates confidence in the AI. Thus, methods for correcting those unwanted relationships in explanation has been developed. However, since these methods are applicable only to differentiable machine learning (ML) models but not to non-differentiable models such as tree-based models, they are insufficient for covering a wide range of ML technology. Since these methods also require re-training of the model for correcting its explanation (i.e., in-processing method), they cannot be applied to black-box models provided by third parties. Therefore, we propose a method called ensemble-based explanation correction (EBEC) as a post-processing method for correcting the global explanation of a prediction model in a model-agnostic manner by using the Rashomon effect of statistics. We evaluated the performance of EBEC with three different tasks and analyzed its function in more detail. The evaluation results indicate that EBEC can correct global explanation of the model so that the explanation aligns with the domain knowledge given by the user while maintaining its accuracy. EBEC can be extended in various ways and combined with any method to improve correction performance since it is a post-processing-type correction method. Hence, EBEC would contribute to high-productivity ML modeling as a new type of explanation-correction method.

In order to solve the low-quality problems such as low brightness, poor contrast, noise interference and color imbalance in night images, a night image enhancement algorithm based on MDIFE-Net curve estimation is presented. This algorithm mainly consists of three parts: Firstly, we design an illumination estimation curve (IEC), which adjusts the pixel level of the low illumination image domain through a non-linear fitting function, maps to the enhanced image domain, and effectively eliminates the effect of illumination loss; Secondly, the DCE-Net is improved, replacing the original Relu activation function with a smoother Mish activation function, so that the parameters can be better updated; Finally, illumination estimation loss function, which combines image attributes with fidelity, is designed to drive the no-reference image enhancement, which preserves more image details while enhancing the night image. The experimental results show that our method can not only effectively improve the image contrast, but also make the details of the target more prominent, improve the visual quality of the image, and make the image achieve a better visual effect. Compared with four existing low illumination image enhancement algorithms, the NIQE and STD evaluation index values are better than other representative algorithms, verify the feasibility and validity of the algorithm, and verify the rationality and necessity of each component design through ablation experiments.

Aiming at the problem of spatial focus regions distribution dispersion and dislocation in feature pyramid networks and insufficient feature dependency acquisition in both spatial and channel dimensions, this paper proposes a spatial-temporal aggregated shuffle attention for video instance segmentation (STASA-VIS). First, an mixed subsampling (MS) module to embed activating features from the low-level target area of feature pyramid into the high-level is designed, so as to aggregate spatial information on target area. Taking advantage of the coherent information in video frames, STASA-VIS uses the first ones of every 5 video frames as the key-frames and then propagates the keyframe feature maps of the pyramid layers forward in the time domain, and fuses with the non-keyframe mixed subsampled features to achieve time-domain consistent feature aggregation. Finally, STASA-VIS embeds shuffle attention in the backbone to capture the pixel-level pairwise relationship and dimensional dependencies among the channels and reduce the computation. Experimental results show that the segmentation accuracy of STASA-VIS reaches 41.2%, and the test speed reaches 34FPS, which is better than the state-of-the-art one stage video instance segmentation (VIS) methods in accuracy and achieves real-time segmentation.

With the continued innovation of deep neural networks, spiking neural networks (SNNs) that more closely resemble biological brain synapses have attracted attention because of their low power consumption. Unlike artificial neural networks (ANNs), for continuous data values, they must employ an encoding process to convert the values to spike trains, suppressing the SNN's performance. To avoid this degradation, the incoming analog signal must be regulated prior to the encoding process, which is also realized in living things eg, the basement membranes of humans mechanically perform the Fourier transform. To this end, we combine an ANN and an SNN to build ANN-to-SNN hybrid neural networks (HNNs) that improve the concerned performance. To qualify this performance and robustness, MNIST and CIFAR-10 image datasets are used for various classification tasks in which the training and encoding methods changes. In addition, we present simultaneous and separate training methods for the artificial and spiking layers, considering the encoding methods of each. We find that increasing the number of artificial layers at the expense of spiking layers improves the HNN performance. For straightforward datasets such as MNIST, similar performances as ANN's are achieved by using duplicate coding and separate learning. However, for more complex tasks, the use of Gaussian coding and simultaneous learning is found to improve the accuracy of the HNN while lower power consumption.

Deep neural networks (DNNs) perform well for image recognition, speech recognition, and pattern analysis. However, such neural networks are vulnerable to adversarial examples. An adversarial example is a data sample created by adding a small amount of noise to an original sample in such a way that it is difficult for humans to identify but that will cause the sample to be misclassified by a target model. In a military environment, adversarial examples that are correctly classified by a friendly model while deceiving an enemy model may be useful. In this paper, we propose a method for generating a selective adversarial example that is correctly classified by a friendly gait recognition system and misclassified by an enemy gait recognition system. The proposed scheme generates the selective adversarial example by combining the loss for correct classification by the friendly gait recognition system with the loss for misclassification by the enemy gait recognition system. In our experiments, we used the CASIA Gait Database as the dataset and TensorFlow as the machine learning library. The results show that the proposed method can generate selective adversarial examples that have a 98.5% attack success rate against an enemy gait recognition system and are classified with 87.3% accuracy by a friendly gait recognition system.

We present an interactive modeling system for Japanese castles. We develop an user interface that can generate the fundamental structure of the castle tower consisting of stone walls, turrets, and roofs. By clicking on the screen displaying the 3D space with the mouse, relevant parameters are calculated automatically to generate 3D models of Japanese-style castles. We use characteristic curves that often appear in ancient Japanese architecture for the realistic modeling of the castles. We evaluate the effectiveness of our method by comparing the castle generated by our method with a commercially-available 3D mode of a castle.