Real Tasks Research Articles

An Innovative Instructional Model to Foster Creative Thinking in Primary School Students in Thailand

This research endeavors to address the pressing need for nurturing creative thinking skills among primary school students in the context of Mahasarakham province, Thailand. By systematically analyzing the specific challenges and requirements faced by educators, a comprehensive model for enhancing creative thinking skills was developed. This model integrates multidisciplinary methods, hands-on realistic tasks, and combines STEAM (Science, Technology, Engineering, Arts, and Mathematics) education with the 6E technique. The study demonstrates the model's effectiveness in significantly improving creative thinking skills among students. The key findings underscore the value of multidisciplinary approaches, emphasizing experiential learning through practical tasks that stimulate creativity. The fusion of STEAM education and the 6E technique proved to be particularly impactful, aligning with prior research in the field. The research contributes vital insights and practical recommendations for educators and policymakers, urging them to consider multidisciplinary methods, STEAM education, and the 6E technique as viable strategies to bolster creative thinking skills in primary school students.

Creating Real Situations, Developing Thinking Tools Design and Implementation of the Interdisciplinary Project Autumn Museum for Lower Grades of Primary School

Childrens expressive abilities need to be gradually built. In the lower grades of primary school, real situational tasks should be used as the teaching environment, students task inquiry activities as the main line of learning, and language interaction between teachers and students as the teaching technique. The ultimate goal is to enhance individual language experience centered on students pragmatic abilities. By using situational tasks as the main line, continuously providing students with tools for expression, and achieving the learning objective of childrens creative expression through a structured language construction process. This paper develops and designs an expressive literacy course based on the first semester Chinese textbook for first graders.

Federated learning: Impact of different algorithms and models on prediction results based on fashion-MNIST data set

The realm of federated learning is rapidly advancing amid the era of big data. Therefore, how to select a suitable federated learning algorithm to achieve realistic tasks has become particularly critical. In this study, we explore the impact of different algorithms and models on the prediction results of Federated Learning (FL) using the Fashion-MNIST data set. Federated Learning enhances data privacy and reduces latency by training models directly on local devices since it is a decentralized machine learning approach. We analyze the performance of several FL algorithms including Federated Averaging (FedAvg), Federated Stochastic Gradient Descent (FedSGD), Federated Proximal (FedProx), and SCAFFOLD. Our experiments reveal significant differences in accuracy and stability among these algorithms, highlighting their strengths and weaknesses in handling non-IID (Non-Independent and Identically Distributed) data. FedProx demonstrate superior performance in terms of accuracy and robustness, making them suitable for complex federated learning environments. These discoveries offer crucial insights for choosing suitable FL algorithms and models in practical applications.

Data imbalance bearing fault diagnosis based on fusion attention mechanism and global feature cross GAN network

Abstract As one of the important equipment of motor transmission, bearings play an important role in the production and manufacturing industry, if there are problems in the manufacturing process will bring significant economic losses or even endanger personal safety, so its state prediction and fault diagnosis is of great significance. In bearing fault diagnosis, it is a challenge to eliminate the effect of data imbalance on fault diagnosis. Generative adversarial network (GAN) networks have achieved some success in data imbalance fault diagnosis, but GAN networks suffer from sample generation bias when balancing samples. To solve this problem, fusion attention mechanism and global feature cross GAN networks is proposed. Firstly, the spatial channel fusion attention mechanism is added to the generator, so that the generator selectively amplifies and processes sample features from different regions, which helps the generator learn more representative features from a few categories; secondly, the global feature cross module is added to the discriminator, so that the discriminator efficiently extracts features from different samples, and improves its ability of recognizing the sample discrepancy; at the same time, in order to improve the model’s anti-noise ability, an anti-noise module is added to the discriminator to improve the efficiency of the model’s data imbalance fault diagnosis; finally, this paper’s method is validated by using two public bearing datasets and one self-constructed dataset. The experimental results prove that this method can effectively overcome the effect of data imbalance on GAN networks, and has a high accuracy rate in real industrial fault diagnosis tasks, what’s more, it proves that the method in this paper has a very good anti-noise performance and practical application value.

Identification and Validation of New DNA-PKcs Inhibitors through High-Throughput Virtual Screening and Experimental Verification.

DNA-PKcs is a crucial protein target involved in DNA repair and response pathways, with its abnormal activity closely associated with the occurrence and progression of various cancers. In this study, we employed a deep learning-based screening and molecular dynamics (MD) simulation-based pipeline, identifying eight candidates for DNA-PKcs targets. Subsequent experiments revealed the effective inhibition of DNA-PKcs-mediated cell proliferation by three small molecules (5025-0002, M769-1095, and V008-1080). These molecules exhibited anticancer activity with IC50 (inhibitory concentration at 50%) values of 152.6 μM, 30.71 μM, and 74.84 μM, respectively. Notably, V008-1080 enhanced homology-directed repair (HDR) mediated by CRISPR/Cas9 while inhibiting non-homologous end joining (NHEJ) efficiency. Further investigations into the structure-activity relationships unveiled the binding sites and critical interactions between these small molecules and DNA-PKcs. This is the first application of DeepBindGCN_RG in a real drug screening task, and the successful discovery of a novel DNA-PKcs inhibitor demonstrates its efficiency as a core component in the screening pipeline. Moreover, this study provides important insights for exploring novel anticancer therapeutics and advancing the development of gene editing techniques by targeting DNA-PKcs.

Designing learning trajectory to support preservice mathematics teachers' skills in creating and implementing realistic mathematics tasks

In mathematics teaching and learning, mathematics tasks embedded in realistic contexts are crucial for developing mathematical concepts, procedures, and the application of mathematical knowledge. Despite this, mathematics teachers often encounter challenges in designing and implementing such realistic mathematics tasks. Therefore, this study aims to construct a learning trajectory to enhance preservice mathematics teachers' abilities to create and implement Realistic Mathematics Tasks (RMTs). Employing a design research methodology, the study comprises three phases: preliminary design, teaching experiment, and retrospective analysis. The data presented in this article are from the first cycle, encompassing all three phases. The research involved four fourth-year preservice mathematics teachers from Sanata Dharma University in Yogyakarta, Indonesia, participating in a preliminary experiment conducted over three months outside their regular coursework. Data were collected through pre-course written tests, observations, interviews, journal reflections, and analyses of the preservice teachers' work. The findings indicate that the learning trajectory effectively raised awareness among preservice teachers about key characteristics of realistic mathematics tasks. This heightened awareness was evident in their reflections and work. Additionally, observations of their teaching practices revealed that preservice teachers valued student reasoning in solving RMTs. However, they faced difficulties in classroom management and guiding students to construct new knowledge actively through their experiences in solving RMTs.

Structure-based out-of-distribution (OOD) materials property prediction: a benchmark study

In real-world materials research, machine learning (ML) models are usually expected to predict and discover novel exceptional materials that deviate from the known materials. It is thus a pressing question to provide an objective evaluation of ML model performances in property prediction of out-of-distribution (OOD) materials that are different from the training set. Traditional performance evaluation of materials property prediction models through the random splitting of the dataset frequently results in artificially high-performance assessments due to the inherent redundancy of typical material datasets. Here we present a comprehensive benchmark study of structure-based graph neural networks (GNNs) for extrapolative OOD materials property prediction. We formulate five different categories of OOD ML problems for three benchmark datasets from the MatBench study. Our extensive experiments show that current state-of-the-art GNN algorithms significantly underperform for the OOD property prediction tasks on average compared to their baselines in the MatBench study, demonstrating a crucial generalization gap in realistic material prediction tasks. We further examine the latent physical spaces of these GNN models and identify the sources of CGCNN, ALIGNN, and DeeperGATGNN’s significantly more robust OOD performance than those of the current best models in the MatBench study (coGN and coNGN) as a case study for the perovskites dataset, and provide insights to improve their performance.

Efficient Heuristic Management of Cloud Data Lakes for Enhanced Big Data Handling in Cloud Computing Environments

Efficient control of cloud data lakes is important for improving big data dealing in cloud computing settings. Cloud data lakes serve as stores for vast amounts of organized, semi-structured, and unstructured data from varied sources, offering scaling and freedom in data storage and processing. However, the sheer amount and range of data offer difficulties in improving storage, access, and analysis processes. This study suggests a heuristic method to handle these issues by improving the order and recovery of data within cloud data lakes. The heuristic method focuses on automatically handling data splitting, storage, and caching strategies based on data access trends and task factors. By leveraging heuristic principles, such as adaptable splitting and predictive caching, the suggested method aims to reduce delay and improve total query speed. Key components of the heuristic management include adaptable splitting methods that actively change data partition sizes based on access frequency and data distribution patterns. This flexible method ensures that frequently accessed data is easily available, while less accessed data is efficiently saved to improve storage usage. Furthermore, predictive storing methods expect future data access patterns using machine learning models or previous data trends. By strategically saving relevant datasets in memory or high-speed storage, the system reduces delay for repeated questions, thereby improving real-time data processing capabilities. The success of the suggested heuristic management method is tested through models and benchmarks using real big data tasks. Comparative research against standard static splitting and caching methods shows significant changes in query response times and resource utilization efficiency.

Framework for the adoption, evaluation and impact of occupational Exoskeletons at different technology readiness levels: A systematic review

Work-related Musculoskeletal Disorders (WMSDs) are the most common occupational diseases caused by the prolonged performance of strenuous work, such as manual handling of loads or long-term maintenance of incongruous postures. Different safety protocols are implemented to reduce WMSDs and optimize the working environment, but one of the most promising solutions is using occupational exoskeletons (OEs). However, to truly acknowledge the benefits of OEs and be able to introduce them into daily business use, devices must pass several development and testing stages that determine the Technology Readiness Level (TRL). This review study aims to present an up-to-date collection of the most advanced assessments of exoskeletons for upper and back support, ranging from laboratory real-task simulations to operational scenarios in industrial sites. To identify relevant studies, we conducted comprehensive searches across different electronic databases, i.e., PubMed, Scopus, and Web of Science. Different keywords were used for the literature search, e.g., occupational exoskeleton, industrial exoskeleton, etc. Studies were included if they investigated the assessment of exoskeletons in the laboratory with real tasks or an industrial environment. We identified 45 research articles that fulfilled this selection criterion. Several features are compared and discussed in detail, such as industrial environment, experimental protocol, task performed, and exoskeleton typology. These data allowed us to formulate results that report the correspondence or discrepancy between the number of papers testing exoskeletons and WMSDs in different industrial sectors, the type of assessment performed, and the impact of exoskeletons on workers and industries at different TRLs. Among the results, the incidence of WMSDs in the manufacturing industry is 21.13%, while the adoption of exoskeletons in the same field is the highest with respect to the other industrial fields, at 44.45%. Electromyography (EMG) and Questionnaires were the most evaluated typologies across all development and testing stages (with an incidence of 64% across the selected articles). Additionally, an average reduction of EMG activity was reported, with 24% for Upper Limb and 20% for Back Support. Regarding the subjective assessment reported in the questionnaires, 68% of the studies reported a positive evaluation. Based on these outcomes, this work provides a framework for an effective evaluation process for the OEs to raise TRL with recommendations for future research activities.

Addressing attentional issues in augmented reality with adaptive agents: Possibilities and challenges

Recent work on augmented reality (AR) has explored the use of adaptive agents to overcome attentional issues that negatively impact task performance. However, despite positive technical evaluations, adaptive agents have shown no significant improvements to user task performance in AR. Furthermore, previous works have primarily evaluated such agents using abstract tasks. In this paper, we develop an agent that observes user behaviour and performs appropriate actions to mitigate attentional issues in a realistic sense-making task in AR. We employ mixed methods to evaluate our agent in a between-subject experiment (N=60) to understand the agent’s effect on user task performance and behaviour. While we find no significant improvements in task performance, our analysis revealed that users’ preferences and trust in the agent affected their receptiveness of the agent’s recommendations. We discuss the pitfalls of autonomous agents and highlight the need to shift from designing better Human–AI interactions to better Human–AI collaborations.

The digital transformation of jurisprudence: an evaluation of ChatGPT-4’s applicability to solve cases in business law

AbstractIn the evolving landscape of legal information systems, ChatGPT-4 and other advanced conversational agents (CAs) offer the potential to disruptively transform the law industry. This study evaluates commercially available CAs within the German legal context, thereby assessing the generalizability of previous U.S.-based findings. Employing a unique corpus of 200 distinct legal tasks, ChatGPT-4 was benchmarked against Google Bard, Google Gemini, and its predecessor, ChatGPT-3.5. Human-expert and automated assessments of 4000 CA-generated responses reveal ChatGPT-4 to be the first CA to surpass the threshold of solving realistic legal tasks and passing a German business law exam. While ChatGPT-4 outperforms ChatGPT-3.5, Google Bard, and Google Gemini in both consistency and quality, the results demonstrate a considerable degree of variability, especially in complex cases with no predefined response options. Based on these findings, legal professionals should manually verify all texts produced by CAs before use. Novices must exercise caution with CA-generated legal advice, given the expertise needed for its assessment.

Transdisciplinary understanding of the fundamental and practice-oriented nature of pedagogical education

Relevance. Increasing of fundamentality and practice-oriented nature are two important strategies for development of pedagogical education at the present stage. Fundamentalization ensures the formation of deep basic knowledge system, defines a holistic scientific picture of the human world. Practice orientation is revealed through a prism of gaining experience in the course of various activities, the willingness of a person to productively increase and use his or her knowledge and methods of activity to solve problems arising in the reality, orientation towards continuous self-education and self-development. The aim of the article is scientific substantiation of transdisciplinarity as a strategy for bringing together the fundamentality and practice-oriented pedagogical education at the present stage. Methodology and methods. The methodological basis of transdisciplinarity is based on the ideas of a synergetic approach. The ideas of the praxiological approach are used in determining the essential characteristics of pedagogical actions success in terms of improving the quality of professional training for teachers. The professional training of future teachers to solve comprehensively set real tasks based on the formed knowledge system and acquired experience is focused on the concept of an activity-based approach. The results of the study. The results are presented by the scientific substantiation of the integration of the fundamentality and practice-oriented training for future teachers on the basis of a transdisciplinary strategy, the definition of conditions for the formation of transdisciplinary thinking among teachers, the development of practical skills and skills for solving complex problems through the prism of diverse knowledge and methods of activity: the inclusion of teachers in project, research, experimental activities, the organization of continuing education, strengthening the nonlinearity of the development in the educational system and ensuring disciplinary flexibility, reflexive understanding of the results for activities based on the consideration of praxiological knowledge to ensure the success of pedagogical actions. Conclusion. A teacher can carry out labor pedagogical actions effectively, efficiently, technologically and safely using the pedagogical education based on a transdisciplinary basis, in close interaction of deep fundamental knowledge acquired through direct practice of skills and experience, as well as using praxiological knowledge based on an understanding of the general theory of successful activity. The presented results can be used in the development of educational programmes for pedagogical bachelor’s degree and programmes of additional professional education based on the capabilities of the digital learning environment.

Semi-supervised prototype network based on compact-uniform-sparse representation for rotating machinery few-shot class incremental fault diagnosis

Distinct from the ideal ‘closed-set’ assumption of comprehensive fault types and arbitrary data access, the entangled challenges of labeled instance paucity and fault increasing with operation are difficult to circumvent in industrial diagnostics. However, conventional deep learning-based diagnostic models lack the capacity for dynamic and incremental acceptance of new faults under limited labeled instances. When confronted with this more industrially realistic diagnostic task, their performance is constrained by overfitting, imbalance, and catastrophic forgetting, thus diminish the feasibility of engineering deployment. In this research, we define this task as a few-shot class incremental fault diagnosis (FS-CIFD) problem and develop a semi-supervised prototype network based on compact-uniform-sparse representation (CUS-SSPN) for this purpose. Structurally, CUS-SSPN incorporates the designed compact-uniform-sparse (CUS) composite loss, optimal transport (OT)-based semi-supervised strategy, and multi-level knowledge distillation (KD) strategy into the prototype network. With the CUS composite loss and OT-based semi-supervised strategy, CUS-SSPN can assign representative prototypes for each incremental fault and form an intra-class compact and inter-class uniform and sparse representation space. This allows the model to incrementally learn new faults while effectively suppressing the overfitting triggered by instance paucity and the biased diagnosis triggered by unbalanced labeling. Combining the designed multi-level KD strategy to convey category centroid information and classification decision knowledge from both prototype and logits levels, catastrophic forgetting of known faults can be effectively mitigated. Extensive FS-CIFD tasks are constructed on both publicly available dataset and practical wind turbine dataset to validate the effectiveness of CUS-SSPN and the feasibility of engineering diagnostics.

Partial RNA design.

RNA design is a key technique to achieve new functionality in fields like synthetic biology or biotechnology. Computational tools could help to find such RNA sequences but they are often limited in their formulation of the search space. In this work, we propose partial RNA design, a novel RNA design paradigm that addresses the limitations of current RNA design formulations. Partial RNA design describes the problem of designing RNAs from arbitrary RNA sequences and structure motifs with multiple design goals. By separating the design space from the objectives, our formulation enables the design of RNAs with variable lengths and desired properties, while still allowing precise control over sequence and structure constraints at individual positions. Based on this formulation, we introduce a new algorithm, libLEARNA, capable of efficiently solving different constraint RNA design tasks. A comprehensive analysis of various problems, including a realistic riboswitch design task, reveals the outstanding performance of libLEARNA and its robustness. libLEARNA is open-source and publicly available at: https://github.com/automl/learna_tools.

Properties of Fairness Measures in the Context of Varying Class Imbalance and Protected Group Ratios

Society is increasingly relying on predictive models in fields like criminal justice, credit risk management, and hiring. To prevent such automated systems from discriminating against people belonging to certain groups, fairness measures have become a crucial component in socially relevant applications of machine learning. However, existing fairness measures have been designed to assess the bias between predictions for protected groups without considering the imbalance in the classes of the target variable. Current research on the potential effect of class imbalance on fairness focuses on practical applications rather than dataset-independent measure properties. In this article, we study the general properties of fairness measures for changing class and protected group proportions. For this purpose, we analyze the probability mass functions of six of the most popular group fairness measures. We also measure how the probability of achieving perfect fairness changes for varying class imbalance ratios. Moreover, we relate the dataset-independent properties of fairness measures described in this work to classifier fairness in real-life tasks. Our results show that measures such as Equal Opportunity and Positive Predictive Parity are more sensitive to changes in class imbalance than Accuracy Equality. These findings can help guide researchers and practitioners in choosing the most appropriate fairness measures for their classification problems.

Design and Manufacturing Using 3D Printing Technology of A 5-DOF Manipulator for Industrial Tasks

Robotic manipulators have become very necessary in industrial applications all over the world. In this paper, a 5-DOF robotic manipulator is designed and manufactured to simulate a real industrial task. The manipulator is intended to transfer an object with a weight of 30 grams from a known place to another known one, which is a pick and place task. Firstly, all parts of the manipulator are designed using SolidWorks software. During the design, all parts’ dimensions are considered. The end-effector of the manipulator is designed based on gear system. Secondly, 3D printing technology is used to manufacture these designed parts. The manufacturing process is very accurate and efficient. Servo motors are considered to do the motion of the manipulator, which are easily and directly connected to the control circuit. As, 5-DOF manipulator is manufactured, five servo motors are used: one motor for every joint. The motion of the motors is controlled by Arduino Uno unit which is a cheap and easy programming unit. Experiments are executed with the developed robot to show its effectiveness and success by preparing three boxes which the robot effectively transfers from one place to another. Eventually, the challenges during the design and manufacturing of this robot are mentioned in this paper.

Towards Controllable Generative Design: A Conceptual Design Generation Approach Leveraging the FBS Ontology and Large Language Models

Abstract Recent research in the field of design engineering is primarily focusing on using AI technologies such as Large Language Models (LLMs) to assist early-stage design. The engineer or designer can use LLMs to explore, validate and compare thousands of generated conceptual stimuli and make final choices. This was seen as a significant stride in advancing the status of the generative approach in computer-aided design. However, it is often difficult to instruct LLMs to obtain novel conceptual solutions and requirement-compliant in real design tasks, due to the lack of transparency and insufficient controllability of LLMs. This study presents an approach to leverage LLMs to infer Function-Behavior-Structure (FBS) ontology for high-quality design concepts. Prompting design based on the FBS model decomposes the design task into three sub-tasks including functional, behavioral, and structural reasoning. In each sub-task, prompting templates and specification signifiers are specified to guide the LLMs to generate concepts. User can determine the selected concepts by judging and evaluating the generated function-structure pairs. A comparative experiment has been conducted to evaluate the concept generation approach. According to the concept evaluation results, our approach achieves the highest scores in concept evaluation, and the generated concepts are more novel, useful, functional, and low-cost compared to the baseline.

Human-like Learning in Temporally Structured Environments

Natural environments have correlations at a wide range of timescales. Human cognition is tuned to this temporal structure, as seen by power laws of learning and memory, and by spacing effects whereby the intervals between repeated training data affect how long knowledge is retained. Machine learning is instead dominated by batch iid training or else relatively simple nonstationarity assumptions such as random walks or discrete task sequences. The main contributions of our work are: (1) We develop a Bayesian model formalizing the brain's inductive bias for temporal structure and show our model accounts for key features of human learning and memory. (2) We translate the model into a new gradient-based optimization technique for neural networks that endows them with human-like temporal inductive bias and improves their performance in realistic nonstationary tasks. Our technical approach is founded on Bayesian inference over 1/f noise, a statistical signature of many natural environments with long-range, power law correlations. We derive a new closed-form solution to this problem by treating the state of the environment as a sum of processes on different timescales and applying an extended Kalman filter to learn all timescales jointly. We then derive a variational approximation of this model for training neural networks, which can be used as a drop-in replacement for standard optimizers in arbitrary architectures. Our optimizer decomposes each weight in the network as a sum of subweights with different learning and decay rates and tracks their joint uncertainty. Thus knowledge becomes distributed across timescales, enabling rapid adaptation to task changes while retaining long-term knowledge and avoiding catastrophic interference. Simulations show improved performance in environments with realistic multiscale nonstationarity. Finally, we present simulations showing our model gives essentially parameter-free fits of learning, forgetting, and spacing effects in human data. We then explore the analogue of human spacing effects in a deep net trained in a structured environment where tasks recur at different rates and compare the model's behavioral properties to those of people.

The effect of VR on fine motor performance by older adults: a comparison between real and virtual tasks

Virtual Reality (VR) technology has the potential to support the aging population and improve testing of daily abilities to detect functional decline. In multiple research studies, VR performance of participants has been assessed by measuring time to complete test, but the effect of learning how to use the VR system and differences between real and virtual environments have been understudied, especially for fine motor tasks. In this study, 20 older adults ages 65–84 performed a task that required fine motor skills in real-life and then in a VR replica of the same task. All participants completed the task in each setting with no difficulties. A clear learning effect was observed in VR, which was attributed to learning how to use the device itself. Still, participants could not reach the same level of performance (time) in VR as in real-life. Participants rated the VR task more mentally and physically demanding than in real-life, as well as more stressful, but with an overall low cognitive demand. In an exploratory cluster analysis, participants with an average age of 69 years old had more technological devices, found the VR system more usable and realistic than participants in the group with an average of 76 years old. This study demonstrated that VR influences time to complete a fine motor task, and that learning effects related to the system could be confounded with actual task performance if not properly considered in VR studies with older adults.

Matching UAV-Based Simulated High-Resolution SAR and Real SAR Images

Abstract. Our experiment focuses on generating simulated high-resolution (HR) Synthetic Aperture Radar (SAR) images from the photogrammetric 3D model derived from Unmanned Aerial Vehicle (UAV) imagery. The goal is to match these simulated images with real SAR data of complex scenes containing vegetation, buildings, roads, and the construction of receiving antennas. Simulated SAR reflectivity map was based on ray tracing techniques.Two feature matching algorithms, namely Scale-Invariant Feature Transform (SIFT) and Oriented FAST and Rotated BRIEF (ORB) were employed for co-registration of the simulated to the real SAR images. The effectiveness of these algorithms is evaluated by their matching rates. The experiment results of the matching process highlight the superior performance of ORB over SIFT, attributed to its robustness against rotation variations and noise. ORB emerges as a versatile choice for high-resolution real and simulated SAR registration tasks.