Computational Ability Research Articles

BloodPatrol: Revolutionizing Blood Cancer Diagnosis - Advanced Real-Time Detection Leveraging Deep Learning & Cloud Technologies.

Cloud computing and Internet of Things (IoT) technologies are gradually becoming the technological changemakers in cancer diagnosis. Blood cancer is an aggressive disease affecting the blood, bone marrow, and lymphatic system, and its early detection is crucial for subsequent treatment. Flow cytometry has been widely studied as a commonly used method for detecting blood cancer. However, the high computation and resource consumption severely limit its practical application, especifically in regions with limited medical and computational resources. In this study, with the help of cloud computing and IoT technologies, we develop a novel blood cancer dynamic monitoring diagnostic model named BloodPatrol based on an intelligent feature weight fusion mechanism. The proposed model is capable of capturing the dual-view importance relationship between cell samples and features, greatly improving prediction accuracy and significantly surpassing previous models. Besides, benefiting from the powerful processing ability of cloud computing, BloodPatrol can run on a distributed network to efficiently process large-scale cell data, which provides immediate and scalable blood cancer diagnostic services. We have also created a cloud diagnostic platform to facilitate access to our work, the latest access link and updates are available at: https://github.com/kkkayle/BloodPatrol.

Gas Entrainment Simulation for Fast Reactors using Freesurface Lattice Boltzmann Method

In fast reactor designs, preventing gas entrainment (GE) caused by free-surface vortices is crucial. Numerical analysis is typically used to identify scenarios with and without GE and speedup the design process. However, conventional numerical methods based on incompressible Navier-Stokes solvers are often inefficient and inaccurate. To address this, we explore the Lattice Boltzmann Method (LBM) as an alternative, known for its computational efficiency and ability to simulate complex flows. In this study, we employ freesurface LBM to accelerate GE analysis, greatly reducing computational costs. Our LBM simulations show good agreement with experimental data, suggesting a promising approach for faster analysis in future fast reactor designs.

Dynamic Event-Triggered Strategy-Based Optimal Control of Modular Robot Manipulator: A Multiplayer Nonzero-Sum Game Perspective.

Due to the limited computing and processing ability of modular robot manipulator (MRM) components, such as sensors and controllers, event-triggered mechanisms are considered a crucial communication paradigm shift in resource constrained applications. Dynamic event-triggered mechanism is developing into a new technology by reason of its higher resource utilization efficiency and more flexible system design requirements than traditional event-triggered. Therefore, an optimal control scheme of multiplayer nonzero-sum game based on dynamic event-triggered is developed for MRM systems with uncertain disturbances. First, dynamic model of the MRM is established according to joint torque feedback technique and model uncertainty is estimated by data-driven-based neural network identifier. In the framework of differential game, the tracking control problem of MRM system is transformed into the optimal control problem for multiplayer nonzero-sum game with the control input of each joint module as the player. Then, the static event-triggered control problem of MRM system is studied based on adaptive dynamic programming algorithm. On this basis, the internal dynamic variable describing the previous state of the system is introduced, and the characteristics of dynamic trigger rule and its relationship with static rule are revealed theoretically. By designing an exponential attenuation signal, the minimum sampling interval of the system is always positive, so that Zeno behavior is excluded. Lyapunov theory proves that the system is asymptotically stable and the experimental results verify the validity of the proposed method.

Research on cultivating computational thinking in university computer courses - Based on PBL strategies

This study focuses on the cultivation of computational thinking in the "University Computer" course, proposing a teaching model strategy based on PBL (Problem-Based Learning). By designing authentic and challenging teaching problems, students are guided to actively explore and collaborate to solve them, thereby cultivating their computational thinking abilities. The research found that the PBL teaching model can effectively enhance students' computational thinking skills, as well as strengthen their problem-solving and innovation abilities. Additionally, this study also discusses the issues and challenges encountered during implementation, providing valuable insights for future teaching practices.

Parallel Adaptive Stochastic Gradient Descent Algorithms for Latent Factor Analysis of High-Dimensional and Incomplete Industrial Data

Latent factor analysis (LFA) is efficient in knowledge discovery from a high-dimensional and incomplete (HDI) matrix frequently encountered in industrial big data-related applications. A stochastic gradient descent (SGD) algorithm is commonly adopted as a learning algorithm for LFA owing to its high efficiency. However, its sequential nature makes it less scalable when processing large-scale data. Although alternating SGD decouples an LFA process to achieve parallelization, its performance relies on its hyper-parameters that are highly expensive to tune. To address this issue, this paper presents three extended alternating SGD algorithms whose hyper-parameters are made adaptive through particle swarm optimization. Correspondingly, three Parallel Adaptive LFA (PAL) models are proposed and achieve highly efficient latent factor acquisition from an HDI matrix. Experiments have been conducted on four HDI matrices collected from industrial applications, and the benchmark models are LFA models based on state-of-the-art parallel SGD algorithms including the alternative SGD, Hogwild!, distributed gradient descent, and sparse matrix factorization parallelization. The results demonstrate that compared with the benchmarks, with 32 threads, the proposed PAL models achieve much speedup gain. They achieve the highest prediction accuracy for missing data on most cases. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —HDI data are commonly encountered in many industrial big data-related applications, where rich knowledge and patterns can be extracted efficiently. An SGD based-LFA model is popular in addressing HDI data due to its efficiency. Yet when dealing with large-scale HDI data, its serial nature greatly reduces its scalability. Although alternating SGD can decouple an LFA process to implement parallelization, its performance depends on its hyper-parameter whose tuning is tedious. To address this vital issue, this study proposes three extended alternating SGD algorithms whose hyper-parameters are made via through a particle swarm optimizer. Based on them, three models are realized, which are able to efficiently obtain latent factors from HDI matrices. Compared with the existing and state-of-the-art models, they enjoy their hyper-parameter-adaptive learning process, as well as highly competitive computational efficiency and representation learning ability. Hence, they provide practitioners with more scalable solutions when addressing large HDI data from industrial applications.

User-centric Heterogeneous-action Deep Reinforcement Learning for Virtual Reality in the Metaverse over Wireless Networks

The Metaverse emerging as maturing technologies are empowering the different facets. Virtual Reality (VR) technologies serve as the backbone of the virtual universe within the Metaverse to offer a highly immersive user experience. As mobility is emphasized in the Metaverse context, VR devices reduce their weights at the sacrifice of local computation abilities. In this paper, for a system consisting of a Metaverse server and multiple VR users, we consider two cases of (i) the server generating frames and transmitting them to users, and (ii) users generating frames locally and thus consuming device energy. As Metaverse emphasizes on the accessibility for all users anywhere and anytime, the users can have totally different characteristics, devices and demands. In this paper, the channel access arrangement (including the decisions on frame generation location), and transmission powers for the downlink communications from the server to the users are jointly optimized by our proposed user-centric Deep Reinforcement Learning (DRL) algorithm, namely User-centric Critic with Heterogenous Actors (UCHA). Comprehensive experiments demonstrate that our UCHA algorithm leads to remarkable results under various requirements and constraints.

Computational thinking ability becomes a predictor of mathematical critical thinking ability

This research aims to determine the relationship between computational thinking ability and mathematical critical thinking ability of seventh-grade students at SMPN 49 Jakarta. Computational thinking ability is skills or processes in solving problems effectively, while mathematical critical thinking ability is a process of analyzing problems to make an accurate decision. This research used correlation research with research subjects consisting of 14 female students and 16 male students. The instrument in this study is a test in the form of a description, which is measured using indicators. This study used decomposition, pattern recognition, abstraction, and algorithm thinking as indicators of computational thinking ability. This study uses elementary clarification, essential support, inference, advanced clarification, strategy, and tactics as indicators of mathematical critical thinking. The outcome of this research showed a positive correlation between computational thinking ability and students’ mathematical critical thinking ability. The relationship analysis shows a simultaneous and significant relationship between computational thinking and essential mathematical thinking abilities. The correlation test results using Pearson Product Moment obtained a result of 0.897 with a coefficient of determination of 80.5%. The magnitude of the correlation indicates that the two abilities have a solid relationship. If computational thinking ability is high, then mathematical critical thinking ability is also high, and if computational thinking ability is low, then mathematical critical thinking ability is also low.

Visual Image Captioning through Transformer

Abstract: The convergence of computer vision and natural language processing in Artificial Intelligence has sparked significant interest in recent years, largely propelled by the advancements in deep learning. One notable application born from this synergy is the automatic description of images in English. Image captioning involves the computer's ability to interpret visual information from an image and translate it into one or more descriptive phrases. Generating meaningful descriptions requires understanding the state, properties, and relationships between the depicted objects, demanding a grasp of high-level picture semantics. Automatically captioning images is a complex task that intertwines image analysis with text generation. Central to this process is the concept of attention, determining what to describe and in what sequence. While transformer architectures have shown success in text analysis and translation, adapting them for image captioning presents unique challenges due to structural differences between semantic units in images (usually identified regions from object detection models) and sentences (composed of individual words). Little effort has been devoted to tailoring transformer architectures to suit images' structural characteristics. In this study, we introduce the Image Transformer, a novel architecture comprising a modified encoding transformer and an implicit decoding transformer. Our approach involves expanding the inner architecture of the original transformer layer to better accommodate the structural nuances of images. By utilizing only region features as inputs, our model achieves state-of-the-art performance on the MSCOCO Dataset. This research employing CNN-Transformer architectural models for image captioning aims to detect objects within images and convey information through textual messages. The envisioned application of this method extends to aiding individuals with visual impairments, using text-to-speech messages to facilitate their access to information and nurture their cognitive abilities. This paper meticulously explores fundamental concepts in image captioning and its standardized procedures, introducing a generative CNN-Transformer model as a significant advancement in this field.

Prediction of thermal infrared radiation using an artificial neural network applied to the projection and design of processes in renewable energies

This work aims to predict thermal infrared radiation in geographical areas where the necessary measurement devices are not available, through the design of an artificial neural network (RNA). The RNA uses the following variables as input data: specific humidity, relative humidity, ambient temperature, wind speed, and atmospheric pressure, it is important to mention that the sample of space of time is from, (1990 - 2019), they are data from Mexico City, as it is a metropolis with an extensive air quality database, which are obtained from two online tools developed by the National Aeronautics and Space Administration (NASA). In addition, thermal infrared radiation data from NASA are included, to validate the prediction made by the algorithm. Matlab was used to implement RNA, a multiplatform software that offers an integrated development environment with its own programming language. It is recognized for its computational ability and is considered a suitable tool for this purpose.

Analysis of the Principal, Fabrication and Application of AI Chips

As a matter of fact, artificial intelligence (AI) technology has developed by leaps and bounds with the rapid development computation ability and improvements of machine learning scenarios (e.g., deep learning neural network). On this basis, its applications have appeared in various fields in recent years. In order to support complex artificial intelligence computing, AI chips have gradually appeared in people's vision, as well as have played an essential role in the future development of life. With this in mind, based on the development of AI chips in recent years, this paper discusses the definition and classification, principles and design, preparation, and application of AI chips. According to the analysis, the current limitations in terms of the state-of-art facilities will be demonstrated. At the same time, this study predicts AI chips’ future development prospects, so that readers can have a more comprehensive understanding of AI chips and the field of artificial intelligence.

Automated Passive Income from Stock Market Using Machine Learning and Big Data Analytics with Security Aspects

With the evolution of Machine Learning, the Internet of Things (IoT), and Big Data Technologies, digital data has increased exponentially. To handle and process such a large bunch of data, high-performance computers are used across various fields. One of the fields has been the financial capital market of stocks, bonds, commodities, foreign exchange, and cryptocurrencies, where supercomputers essentially trade securities with high computational ability and intelligent algorithms. Large financial institutions with significant capital spend a lot of money on programmers and data analysts to develop the best accuracy trading algorithms to drive the overall market. A novice trader or investor with less to no experience in financial markets feels it difficult to search for good trades or stocks to invest their hard-earned money on a short to long-term time horizon. They rely on expert advice for stock recommendations and sometimes end up making significant losses. This paper focuses on developing a universal trend trading indicator that can analyze and predict the overall future trend of any stock, bond, commodity, forex, or cryptocurrency with the highest possible profitability. The historically traded extensive dataset of stock prices and investment reports of large financial institutions worldwide are gathered. Various machine learning and decision-making models are employed to perform technical and fundamental analysis across multiple securities. The output of the trend trading indicator is displayed on charting platforms, which can provide entry-exit levels at which even novice investors can decide where to invest their money. Multi timeframe analysis is deployed to predict short-term, mediumterm, and long-term overall trends, thus increasing the output accuracy. The indicator is helpful for all kinds of retail traders and investors worldwide who struggle to earn profits from financial markets. Our proposed system was able to achieve a profitability of 86.28% annual returns. The entire system, along with the trading orders, is automated so that anyone can earn extra passive income every month from the stock market.

Chaos inspired invasive weed optimization algorithm for parameter estimation of solar PV models

High performance solar photovoltaic models require precise knowledge of solar PV cell parameters. Numerous methods based on both deterministic and meta-heuristics have been developed for identifying solar cell parameters. However, the presented methods in the literature have a heavy computational load and limited ability to extract crucial parameters due to nonlinear dynamics of solar PV systems. In addition, because they rely on approximations to determine the objective function, the preceding state-of-the-art parameter estimation techniques do not provide accurate results. Thus, a novel chaos-inspired invasive weed optimization (CIIWO) has been developed for accurate solar PV system parameter estimation. Adding a chaotic map to IWO improves the performance of suggested method by expanding the search space globally. Moreover, to cope with the inadequacy in state-of-art objective functions, Newton Raphson approach has been combined with proposed CIIWO algorithm. The suggested approach for solar cell parametric identification has been tested on one-diode, two-diode, and three-diode models. By contrasting the outcomes with nine contemporary optimization strategies for parameter estimation, the superiority of the suggested algorithm has been demonstrated. Commercial PV cell RTC France has been used for the experimental validation. Comprehensive study of experimental data validates the efficacy and stability of the suggested algorithm.

فعل الأمر في ضوء اللسانيات الحاسوبية

The present study aims to employ computational linguistic as a mold for reconstructing Arabic grammar generally, and imperative verb specifically, in an attempt to exceed the traditional description of the cognitive human awareness to a specification which is close enough to language concept.&#x0D; The study regards "Imperative verb" as a model through which a new perspective of Arabic grammar is presented, taking computational linguistics into account, and going beyond description to specification.&#x0D; The principal objective of the present study is to rim language with computational frames, demonstrate and describe the ambiguities and complexities facing these frames, and base data that are automatically generated about the nature of the speaker’s awareness of the language. This step starts with defining the parts of the words up to identifying the imperative verb-the model investigated in the present study. Also, it goes beyond the computational description that considers linguistic issues that converge with the model under investigation, morphologically, phonologically, syntactically, and semantically.&#x0D; Objectives:&#x0D; &#x0D; Introducing computational linguistics: description and specification&#x0D; Specifying imperative verb in terms of unaugmentation and addition through exploring: trilateral unaugmented imperative verb both strong and weak.&#x0D; Specifying the structure of imperative verb computationally.&#x0D; Complexities of clarifying the imperative verb&#x0D; &#x0D; Results/ Conclusion&#x0D; The present study has reached to a group of results that can be summarized as follows:&#x0D; &#x0D; Considering computational linguistics and computerizing the Arabic language in a way that assists in promoting its contribution in knowledge community and specifying research projects in the field.&#x0D; Presenting a model that suits to be the basis of specifying other areas of grammar.&#x0D; Establishing a mathematical database with the air of giving computer linguistic intuition that is as close as possible to that of the native speaker. The goal is to reach a computational ability that distinguishes right and wrong, permissible, or un-permissible.&#x0D; &#x0D;

Patterns of Computational Thinking Skills for Elementary Prospectives Teacher in Science Learning: Gender Analysis Studies

The PISA data results show that computational thinking abilities are still lacking. Computational thinking ability is influenced by gender. This research aims to analyze patterns of computational thinking skills of prospective elementary school teachers based on gender at 8 universities in Indonesia. In this research, the components of computational thinking skills analyzed are abstraction, algorithmic, decomposition, and pattern recognition. This research is a mix method research with research subjects as many as 234 prospective elementary school teachers at 8 higher educational institutions. The instruments used were test and interviews. The data analysis technique used is a quantitative data analysis technique using SEM PLS and for qualitative data analysis using miles and Huberman. The research results show that computational thinking skills are still low on the decomposition and pattern recognition components. Based on the SEM PLS test results, it shows that computational thinking abilities are related to gender. In general, the computational thinking ability of female students is slightly higher in all sub-indicators than men and there are differences in the pattern of computational thinking ability between male and female elementary school teacher prospective. The ability of prospective female elementary school teachers to answer in more detail and more structured, while the answers of male prospective teachers are shorter and less comprehensive. Recommendations for developing computational thinking skills by developing problem-based learning, contextual project-based learning and STEAM based learning.

Application Of Deep Learning For Image Deepfake Detector Using Convolutional Neural Network Algorithm

Social media has long been used by the public in general as a means of exchanging information. Behind this commonly exchange of information, hide the malicious intent of those who are not responsible for spreading false information or hoaxes. This false information, which can come in various forms such as images, sounds, or videos, can actually be useful when used as stock photos or simply used as caricatures and satire. Unfortunately, false information often used on famous people instead to make them look like they said or did something that never happened. This certainly needs to be controlled, one of which is by using deepfake detector that aims to recognize false information pattern. Deepfake detector utilizes the computer's ability to self-learn to recognize that invisible patterns in images using one of deep learning algorithms, namely Convolutional Neural Network, which converts images into a collection of arrays containing numbers and then performs mathematical operations repeatedly on each layer. The result of the mathematical operation can then be used as a reference to determine whether an image is real or hoax. Author’s deepfake detector application using Convolutional Neural Network, specifically using the Resnet-50 model on hoax images created using AI with the ProGAN model, appears to be able to detect hoax images with the same model, with an accuracy of 85%, precision of 100%, and recall of 65%, but appears to experience decrease in accuracy when used in deepfakes with other models such as StyleGAN and BigGAN.

Design and Implementation of a Model Predictive Formation Tracking Control System for Underwater Multiple Small Spherical Robots

Due to the characteristics of good concealment ability and strong mobility, multiple, small spherical underwater robot formations play an important role in near coast defense missions, such as cruising, reconnaissance, surveillance, and sensitive target capturing. Referring to the formation problem for underwater small spherical robots with limited energy, perception, and computation abilities, a trajectory tracking-based formation strategy that transforms the complex formation tracking problem into a simple trajectory tracking problem of a single robot is provided. Two layers are designed in the formation tracking strategy. The upper layer is a virtual structure-based formation algorithm. The bottom layer is a tracking controller based on model predictive control (MPC). The formation algorithm is in charge of calculating reference trajectory for each robot in the formation according to the global formation path. The MPC-based dynamic controller for each robot is designed to track the self reference trajectory. Compared with the model predictive control method used for the traditional trajectory tracking problem of a single robot, this paper additionally considers the formation constraints and the internal collision avoidance. In addition, the extended state observer (ESO) is utilized to estimate the lumped disturbance composed of environment disturbance and the inaccurate dynamic model of a small spherical robot. Not only are the numerical simulations based on MATLAB v.2015a, but physical simulations based on self-building multi-spherical robot formation platform are also carried out. Furthermore, through using two small spherical robots, a formation tracing experiment is conducted. All of the results prove that the proposed formation method is feasible and practical for small spherical robots.

Memristor-Based Artificial Chips.

Memristors, promising nanoelectronic devices with in-memory resistive switching behavior that is assembled with a physically integrated core processing unit (CPU) and memory unit and even possesses highly possible multistate electrical behavior, could avoid the von Neumann bottleneck of traditional computing devices and show a highly efficient ability of parallel computation and high information storage. These advantages position them as potential candidates for future data-centric computing requirements and add remarkable vigor to the research of next-generation artificial intelligence (AI) systems, particularly those that involve brain-like intelligence applications. This work provides an overview of the evolution of memristor-based devices, from their initial use in creating artificial synapses and neural networks to their application in developing advanced AI systems and brain-like chips. It offers a broad perspective of the key device primitives enabling their special applications from the view of materials, nanostructure, and mechanism models. We highlight these demonstrations of memristor-based nanoelectronic devices that have potential for use in the field of brain-like AI, point out the existing challenges of memristor-based nanodevices toward brain-like chips, and propose the guiding principle and promising outlook for future device promotion and system optimization in the biomedical AI field.

Application of Big Data Analysis in Sales Forecasting

As a matter of fact, with the rapid development of the computation ability and data delivering, the big data analysis becomes a common tool to realize the predictions. With this in mind, this paper focuses on the impact of big data analysis on sales forecasting, starting with an introduction to the present-day context and significance of sales forecasting and some of the forecasting scenarios implemented by previous generations based on machine learning. After introducing these backgrounds, we start to introduce big data analysis and its impact on sales forecasting. According to the analysis and the summaty of the previous experience, we introduce the commonly used models, the parameters of the models, and the evaluation of the models as well as the application scenarios, and the specific application results. Finally, we demonstrate the limitations as well as future perspectives of big data analysis and sales forecasting. Overall, these results shed light on guiding further exploratio of bigdata analysis in sales prediction.

Ability to Solve Complex Social Problems of Prospective Teachers according to Gender and Computational Thinking

Entering the 21st century, computational thinking has become a basic skill that all students must have. This research aims to determine gender differences in learning outcomes to solve social problems, differences in students' levels of computational thinking by gender, and the influence of students' levels of computational thinking on learning outcomes in solving social problems. This research uses a descriptive verification method with a quantitative analysis approach. The number of research subjects was 256 students, who came from the Faculty of Teacher Training and Education, Sebelas Maret University. The quantitative data analysis used is based on the results of computational thinking ability tests using the Wilcoxon Test. Further analysis of this research uses K-Means for clustering, while analysis of the relationship between variables uses Spearman's rho. This research shows that there is an influence of gender on learning outcomes for solving social problems, there is no significant relationship between students' level of computational thinking and gender, and there is a significant influence of students' level of computational thinking on learning outcomes for solving social problems. The results of this research show that the two factors above play a large role in influencing student learning outcomes. This capability works in synergy with the computational level of thinking. With the right efforts, growing computational thinking skills can improve students' ability to solve various learning problems.

Improving Product Design and Injection Molding for Children’s Toy Plastic with the Help of CAE Software

CAE is the abbreviation for Computer-Aided Engineering. Taking advantage of the computer's ability to analyze and calculate accurately and quickly, to understand the theoretical model of the system (Theoretical Model), and at the same time combine the computer graphics function (Computer Graphics), helping users obtain rapid analysis results, and use the results to modify design and injection molding parameter optimization.&#x0D; CAE combines graphics, computer-aided design (Computer-Aided Design/Draft, CAD) and computer-aided manufacturing (CAM). Due to the reliability of CAE results, it is possible to point out potential problems in the injection molding process and design, suggest design modifications and directions to overcome obstacles and feasible solutions, which can avoid blind spots.&#x0D; CAE at the design stage can be performed on computers to evaluate and optimize design modification options, contribute to reducing time and cost, mold testing, actual mold repair, shortening the actual trial and error cycle, shortening product development time (Product Development Time) and time to market (Time-to-Market), reducing waste and time; time and money in the stages. It can help injection molders predict and understand injection molding parameters that affect product quality, find processing windows and optimize injection molding parameters. It can also indicate the main factors affecting injection molding quality, thereby providing design modification parameters, injection molding parameters and quantitative indicators. It can open the "black box" of injection molding, with a vivid and specific method of displaying machining and design parameters for status sequences and product quality effects, which can help users get quickly accumulated experience in design and injection molding, with relatively large help in employee training.