HighTech and Innovation Journal
Not a member yet
317 research outputs found
Sort by
Development of an Algorithm for Multicriteria Optimization of Deep Learning Neural Networks
Full text linkNowadays, machine learning methods are actively used to process big data. A promising direction is neural networks, in which structure optimization occurs on the principles of self-configuration. Genetic algorithms are applied to solve this nontrivial problem. Most multicriteria evolutionary algorithms use a procedure known as non-dominant sorting to rank decisions. However, the efficiency of procedures for adding points and updating rank values in non-dominated sorting (incremental non-dominated sorting) remains low. In this regard, this research improves the performance of these algorithms, including the condition of an asynchronous calculation of the fitness of individuals. The relevance of the research is determined by the fact that although many scholars and specialists have studied the self-tuning of neural networks, they have not yet proposed a comprehensive solution to this problem. In particular, algorithms for efficient non-dominated sorting under conditions of incremental and asynchronous updates when using evolutionary methods of multicriteria optimization have not been fully developed to date. To achieve this goal, a hybrid co-evolutionary algorithm was developed that significantly outperforms all algorithms included in it, including error-back propagation and genetic algorithms that operate separately. The novelty of the obtained results lies in the fact that the developed algorithms have minimal asymptotic complexity. The practical value of the developed algorithms is associated with the fact that they make it possible to solve applied problems of increased complexity in a practically acceptable time. Doi: 10.28991/HIJ-2023-04-01-011 Full Text: PD
Exploring the Flexibility and Accuracy of Sentiment Scoring Models through a Hybrid KNN-RNN-CNN Algorithm and ChatGPT
Full text linkThis study aimed to address the limitations of sentiment analysis by developing a more accurate and flexible sentiment scoring model using ChatGPT in combination with KNN, RNN, and CNN algorithms. To achieve this, primary data from ChatGPT and secondary data from Kaggle were utilized for training. The model's performance was evaluated, yielding an impressive accuracy rate of 88.17%. This research underscores ChatGPT's pivotal role in offering theoretical insights and precise data for diverse applications. The novelty of this study lies in its innovative approach of combining KNN, RNN, and CNN algorithms to create a more adaptable and accurate sentiment scoring model. Additionally, the primary data from ChatGPT greatly enhances the creation of precise and relevant training data across various topics and languages. Despite these achievements, there remains a need for further exploration of testing methods to mitigate the impact of data limitations on result generalizability. Moreover, it is acknowledged that the model's effectiveness may be diminished when applied to languages other than English. Nevertheless, this research provides a promising avenue for users seeking enhanced and precise sentiment analysis by integrating KNN, RNN, and CNN algorithms with ChatGPT. The findings of this study can serve as a solid foundation for future research endeavors in the advancement of sophisticated and effective sentiment analysis technologies. Doi: 10.28991/HIJ-2023-04-02-06 Full Text: PD
Trainable Regularization in Dense Image Matching Problems
Full text linkThis study examines the development of specialized models designed to solve image-matching problems. The purpose of this research is to develop a technique based on energy tensor aggregation for dense image matching. This task is relevant within the framework of computer systems since image comparison makes it possible to solve current problems such as reconstructing a three-dimensional model of an object, creating a panorama scene, ensuring object recognition, etc. This paper examines in detail the key features of the image matching process based on the use of binocular stereo reconstruction and the features of calculating energies during this process, and establishes the main parts of the proposed method in the form of diagrams and formulas. This research develops a machine learning model that provides solutions to image matching problems for real data using parallel programming tools. A detailed description of the architecture of the convolutional recurrent neural network that underlies this method is given. Appropriate computational experiments were conducted to compare the results obtained with the methods proposed in the scientific literature. The method discussed in this article is characterized by better efficiency, both in terms of the speed of work execution and the number of possible errors. Doi: 10.28991/HIJ-2023-04-03-011 Full Text: PD
Innovative Strategy for Selecting Industries for Program-Target Stimulation of Regional Economic Diversification
Full text linkThe purpose of this study is to substantiate the approach to the selection of industries for program-target stimulation of regional economy diversification, focusing on developing new strong industries and increasing the economic complexity of the regional economy. The research methodology is based on the application of the concept of revealed comparative advantages and an assessment of the economic complexity of industries and regions of Russia (the Udmurt Republic, Republic of Mordovia, Kaliningrad Region, and Trans-Baikal Territory) using data on tax revenues by economic sectors. The novelty of this research lies in demonstrating the effectiveness of applying the revealed comparative advantage concept, an approach to assessing economic complexity based on the use of tax revenue data by economic sectors, and a strategy for modernizing intermediate opportunities when selecting industries for program-target stimulation of regional economy diversification. The practical significance of the results is determined by the possibilities of their use in the application of program-target mechanisms to solve problems of stimulating the development of individual sectors of the regional economy. Selecting priority areas for diversification based on economic complexity methods can contribute to the improvement of budget balancing, economic growth and sustainable development, and mitigation of interregional inequality. Doi: 10.28991/HIJ-2023-04-03-013 Full Text: PD
Advancing Healthcare Security: A Cutting-Edge Zero-Trust Blockchain Solution for Protecting Electronic Health Records
Full text linkThe effective management of electronic health records (EHRs) is vital in healthcare. However, traditional systems often need help handling data inconsistently, providing limited access, and coordinating poorly across facilities. This study aims to tackle these issues using blockchain technology to improve EHR systems' data security, privacy, and interoperability. By thoroughly analyzing blockchain's applications in healthcare, we propose an innovative solution that leverages blockchain's decentralized and immutable nature, combined with advanced encryption techniques such as the Advanced Encryption Standard and Zero Knowledge Proof Protocol, to fortify EHR systems. Our research demonstrates that blockchain can effectively overcome significant EHR challenges, including fragmented data and interoperability problems, by facilitating secure and transparent data exchange, leading to enhanced coordination, care quality, and cost-efficiency across healthcare facilities. This study offers practical guidelines for implementing blockchain technology in healthcare, emphasizing a balanced approach to interoperability, privacy, and security. It represents a significant advancement over traditional EHR systems, boosting security and affording patients greater control over their health records. Doi: 10.28991/HIJ-2023-04-03-012 Full Text: PD
Hardware Engineering of Hazardous Gas and Alcoholic Substances Detector in Meat Using Microcontroller and Gas Sensor
Full text linkMeat may provide not only essential nutritional content but also possible harmful effects on human bodies. Unsafe consumption of meat potentially triggers colorectal cancer risks. Grilling is the most popular way to consume meat. However, meat grilling triggers the formation of hazardous chemical substances such as poisonous Polycyclic Aromatic Hydrocarbons (PAHs). This study conducted experiments using hardware engineering with microcontrollers and different gas sensors, aiming to identify gas substances produced by meat during grilling. The hardware prototype for the test simulation tool was assembled with integrated block systems and circuits. Evaluations were conducted on the direct grilling of three different types of meat. The data results were then utilized to analyze gas substances produced by meat during direct grilling. Based on the results, only five of the seven MQ-type gas sensors used in the research reacted to gas substances produced by all types of meat: LPG, alcohol, carbon monoxide, methane, and carbon dioxide, which were successfully detected in meat during grilling. Our research contributes to discovering a potential prevalence of increased alcoholic content in meat that has been grilled for five minutes. This finding is especially crucial for Muslims since it is highly correlated with halal certification of meat consumption. According to the results, Muslims should wait at least seven minutes or more after direct grilling to let the alcoholic content in meat thoroughly decrease so that it can be safely certified as halal to be consumed according to Islamic laws. Doi: 10.28991/HIJ-2023-04-03-01 Full Text: PD
An Innovative Mobile Application for Wellness Tourism Destination Competitiveness Assessment: The Research and Development Approach
Full text linkObjectives: This research developed and evaluated the effectiveness of an innovative mobile application for wellness tourism destination competitiveness and also studied the adoption effectiveness of this application. Methods/Analysis: A mixed-methods research and development approach was applied to construct a wellness tourism destination competitiveness evaluation model for qualitative research using in-depth interviews, followed by quantitative research using a questionnaire. Weighted scores of criteria and indicators for wellness tourism destination competitiveness were evaluated by the DEMATEL method. The cut-off points for classifying the competitiveness level were set by K-means cluster analysis, while the internal and external accuracy of the model were validated by the confusion matrix technique and the Kruskal-Wallis test. The innovative mobile application was developed using a linear waterfall conceptual design consisting of five software development phases: requirement, design, implementation, verification, and maintenance. A questionnaire was also used to assess the adoption and commercialization of the innovative mobile application. Findings: Results showed that 1) the model gave high accuracy with the confusion matrix technique at 85.42% and the Kruskal-Wallis test classified destination competitiveness at a significance level of 0.0001; and 2) the level of adoption of the innovative mobile application was high. Target users were interested in purchasing a license as the commercial mode of the program. Novelty/Improvement:This research provides a tool to assess the overall competitiveness of wellness tourism destinations. Results can be used to support decision-making and provide practical suggestions for wellness tourism cluster users to adapt when conducting their own competitiveness assessment. The competitiveness assessment results were accurate and in line with the research objectives. Doi: 10.28991/HIJ-2023-04-03-010 Full Text: PD
Competency Model: A Study on the Cultivation of College Students' Innovation and Entrepreneurship Ability
Full text linkObjectives: This study was designed to analyze entrepreneurial competency and enhance college students' abilities in innovation and entrepreneurship. Methods: Ten relevant factors were summarized based on the interview records. Relevant data were collected through questionnaires and tested for reliability and validity. The effectiveness of the ten factors on entrepreneurial competency was tested using the regression analysis method. Then, an analytic hierarchy process model of entrepreneurial competency was established to calculate the relevant weights. Findings: The data collected from the survey questionnaire had sufficient reliability and validity. The ten relevant factors were effective in developing entrepreneurial competence. The weight distribution in the analytic hierarchy model indicated that entrepreneurial knowledge was most important, followed by entrepreneurial ability, and intrinsic potential was least significant. Novelty:The novelty of this article lies in not only verifying the effectiveness of relevant factors through regression analysis but also further analyzing the weight of these factors through an analytic hierarchy process. Doi: 10.28991/HIJ-2023-04-04-011 Full Text: PD
Enterprise Architecture: Enabling Digital Transformation for Operational Business Process during COVID-19
Full text linkThe SARS-CoV-2 pandemic and the global response to contain its spread and deaths have been unprecedented, according to UNICEF research on COVID-19 released in 2021. Many steps had been taken by countries worldwide, particularly those in South Asia. As of May 17th, 2020, Indonesia reported a total of 17,514 daily positive cases. It has been confirmed that the majority of cases throughout the archipelago occur primarily on Java, particularly in the Greater Jakarta, Greater Bandung, Semarang, Solo, and Greater Surabaya areas. The research object of this paper is a system integrator company located in, Central Jakarta. The company's business is badly impacted by this pandemic. The company provides nearly all ICT solutions, yet improving their internal systems is an issue that has never been brought up. Due to physical distance regulations, leading workers to work from home. To keep the business running, the company began using email as their only tool to run the whole system, which is not effective and causing a crisis for the company. The purpose of this paper is to propose a digital transformation plan as a solution and to support business continuity by utilizing TOGAF ADM. Doi: 10.28991/HIJ-2023-04-01-01 Full Text: PD
A Study of Dance Movement Capture and Posture Recognition Method Based on Vision Sensors
Full text linkWith the development of technology, posture recognition methods have been applied in more and more fields. However, there is relatively little research on posture recognition in dance. Therefore, this paper studied the capture and posture recognition of dance movements to understand the usability of the proposed method in dance posture recognition. Firstly, the Kinect V2 visual sensor was used to capture dance movements and obtain human skeletal joint data. Then, a three-dimensional convolutional neural network (3D CNN) model was designed by fusing joint coordinate features with joint velocity features as general features for recognizing different dance postures. Through experiments on NTU60 and self-built dance datasets, it was found that the 3D CNN performed best with a dropout rate of 0.4, a ReLU activation function, and fusion features. Compared to other posture recognition methods, the recognition rates of the 3D CNN on CS and CV in NTU60 were 88.8% and 95.3%, respectively, while the average recognition rate on the dance dataset reached 98.72%, which was higher than others. The experimental results demonstrate the effectiveness of our proposed method for dance posture recognition, providing a new approach for posture recognition research and making contributions to the inheritance of folk dances. Doi: 10.28991/HIJ-2023-04-02-03 Full Text: PD