Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029471
Krisztián Bálint
The latest, modern security camera systems record numerous data at once. With the utilization of artificial intelligence, these systems can even compose an online attendance register of students present during the lectures. Data is primarily recorded on the hard disk of the NVR (Network Video Recorder), and in the long term, it is recommended to save the data in the blockchain. The purpose of the research is to demonstrate how university security cameras can be securely connected to the blockchain. This would be important for universities as this is sensitive student data that needs to be protected from unauthorized access. In my research, as part of the practical implementation, I therefore also use encryption methods and data fragmentation, which are saved at the nodes of the blockchain. Thus, even a DDoS (Distributed Denial of Service) type attack may be easily repelled, as data is not concentrated on a single, central server. To further increase security, it is useful to constitute a blockchain capable of its own data storage at the faculty itself, rather than renting data storage space, so we, ourselves may regulate the conditions of operation, and the policy of data protection. As a practical part of my research, therefore, I created a blockchain called UEDSC (Universities Data Storage Chain) where I saved the student's data.
最新的现代安全摄像系统一次记录大量数据。利用人工智能，这些系统甚至可以在课堂上制作学生的在线出勤记录。数据主要记录在NVR (Network Video Recorder)的硬盘上，从长远来看，建议将数据保存在区块链上。这项研究的目的是展示如何将大学安全摄像头安全地连接到区块链。这对大学来说很重要，因为这是需要保护的敏感学生数据，防止未经授权的访问。因此，在我的研究中，作为实际实施的一部分，我也使用了加密方法和数据分片，这些方法保存在区块链的节点上。因此，即使是DDoS(分布式拒绝服务)类型的攻击也很容易被击退，因为数据没有集中在单个中央服务器上。为了进一步提高安全性，构建一个能够在学院本身存储自己数据的区块链是有用的，而不是租用数据存储空间，因此我们可以自己规范操作条件和数据保护政策。因此，作为我研究的实际部分，我创建了一个名为UEDSC(大学数据存储链)的区块链，在那里我保存了学生的数据。
{"title":"Data Security Structure of a Students’ Attendance Register Based on Security Cameras and Blockchain Technology","authors":"Krisztián Bálint","doi":"10.1109/CINTI-MACRo57952.2022.10029471","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029471","url":null,"abstract":"The latest, modern security camera systems record numerous data at once. With the utilization of artificial intelligence, these systems can even compose an online attendance register of students present during the lectures. Data is primarily recorded on the hard disk of the NVR (Network Video Recorder), and in the long term, it is recommended to save the data in the blockchain. The purpose of the research is to demonstrate how university security cameras can be securely connected to the blockchain. This would be important for universities as this is sensitive student data that needs to be protected from unauthorized access. In my research, as part of the practical implementation, I therefore also use encryption methods and data fragmentation, which are saved at the nodes of the blockchain. Thus, even a DDoS (Distributed Denial of Service) type attack may be easily repelled, as data is not concentrated on a single, central server. To further increase security, it is useful to constitute a blockchain capable of its own data storage at the faculty itself, rather than renting data storage space, so we, ourselves may regulate the conditions of operation, and the policy of data protection. As a practical part of my research, therefore, I created a blockchain called UEDSC (Universities Data Storage Chain) where I saved the student's data.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"55 1","pages":"000185-000190"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89086026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029550
Klaudia Kaszala, György Fekete
The study reviews the movement analysis of a selection of boys’ adolescent gymnasts in knee flexion and knee flexion variation. The analysis allows coaches to design personalized training plans and achieve better results. Gait testing and analysis of knee joint angles have proven to be an appropriate way to test junior gymnasts.
{"title":"Examination of Movement in Young Gymnasts with Special Reference to Knee Angles","authors":"Klaudia Kaszala, György Fekete","doi":"10.1109/CINTI-MACRo57952.2022.10029550","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029550","url":null,"abstract":"The study reviews the movement analysis of a selection of boys’ adolescent gymnasts in knee flexion and knee flexion variation. The analysis allows coaches to design personalized training plans and achieve better results. Gait testing and analysis of knee joint angles have proven to be an appropriate way to test junior gymnasts.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"55 46 1","pages":"000325-000328"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80726259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029474
Behnaz Motamedi, B. Villányi
Agricultural irrigation management has necessitated the incorporation of wireless communication solutions, which includes GSM as the main controller device, ZigBee as a data transmission technology, and MQTT as a messaging protocol. The purpose of this study is to simplify the global and local communication networks to improve flexibility and categorize sensor data. Given its cost-effectiveness and water management, this research could be significant to the agricultural community. Furthermore, we will compare global, local wireless communication, and messaging communication protocols utilized in smart greenhouse irrigation which it will propose a new strategy in the MQTT messaging protocol to enable tracking and supervision irrigation system. In the conclusion of this comparison, the most appropriate, efficient, and beneficial wireless technologies and message protocols for smart greenhouse requirements are GSM, ZigBee, and MQTT.
{"title":"Design of a Smart Irrigation using wireless communication protocols in Greenhouse","authors":"Behnaz Motamedi, B. Villányi","doi":"10.1109/CINTI-MACRo57952.2022.10029474","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029474","url":null,"abstract":"Agricultural irrigation management has necessitated the incorporation of wireless communication solutions, which includes GSM as the main controller device, ZigBee as a data transmission technology, and MQTT as a messaging protocol. The purpose of this study is to simplify the global and local communication networks to improve flexibility and categorize sensor data. Given its cost-effectiveness and water management, this research could be significant to the agricultural community. Furthermore, we will compare global, local wireless communication, and messaging communication protocols utilized in smart greenhouse irrigation which it will propose a new strategy in the MQTT messaging protocol to enable tracking and supervision irrigation system. In the conclusion of this comparison, the most appropriate, efficient, and beneficial wireless technologies and message protocols for smart greenhouse requirements are GSM, ZigBee, and MQTT.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"48 1","pages":"000179-000184"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87262993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029515
T. Obo, Noyuki Kubota
This paper presents an immersive virtual reality (VR) system for assessment of unilateral spatial neglect (USN). USN is defined by the inability for a person to perceive stimuli on one side of the body or environment that is not due to a lack of sensation. In clinical practice, paper-and-pencil tests are used as the quick assessments. However, the sensitivity of these tests is not sufficient for USN assessment in activities of daily living. Thus, VR systems and multimodal sensing systems have been applied to more sensitive diagnosis for USN. In this study, we aim to propose a method to extract specific features of USN patients in behavioral and cognitive levels.
{"title":"Topoplogical Structured Learning for Assessment of Unilateral Spatial Neglect","authors":"T. Obo, Noyuki Kubota","doi":"10.1109/CINTI-MACRo57952.2022.10029515","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029515","url":null,"abstract":"This paper presents an immersive virtual reality (VR) system for assessment of unilateral spatial neglect (USN). USN is defined by the inability for a person to perceive stimuli on one side of the body or environment that is not due to a lack of sensation. In clinical practice, paper-and-pencil tests are used as the quick assessments. However, the sensitivity of these tests is not sufficient for USN assessment in activities of daily living. Thus, VR systems and multimodal sensing systems have been applied to more sensitive diagnosis for USN. In this study, we aim to propose a method to extract specific features of USN patients in behavioral and cognitive levels.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"22 1","pages":"000137-000142"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75274145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029508
Awudu Atinga, J. Bitó, J. Tar
In model-based control of dynamical systems the order of the necessary controller is that of the time-derivative of the generalized coordinate of the system that immediately can be set by the controller. It is determined by the “physics” of the system to be controlled. In finite time-resolution approach the digital controllers model the derivatives by finite differences, and the simplest numerical integration via Euler’s method also uses such differences. Whenever some fractional order derivative of the tracking error is fed back, this long memory term allows the realization of such control strategies for various integer order systems by appropriately masking” and weighting certain elements of longer sequences. On this mathematical basis the idea of controlling a higher order system by the application of a lower order strategy can be formulated, too. The main benefit may be achieving monotonic decrease in the components of the tracking error. In this paper the first order control of a popular second order system often used for benchmark applications, the van der Pol oscillator is considered. The realization of a simple proportional feedback first order controller in two variants, and a PID-type Computed Torque Control are considered in the fixed point iteration-based adaptive framework to tackle the problems caused by modeling imprecisions. The applicability and the limits of the investigated approaches are concluded.
{"title":"On the Simulation of Lower Order Control Strategies for Higher Order Systems","authors":"Awudu Atinga, J. Bitó, J. Tar","doi":"10.1109/CINTI-MACRo57952.2022.10029508","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029508","url":null,"abstract":"In model-based control of dynamical systems the order of the necessary controller is that of the time-derivative of the generalized coordinate of the system that immediately can be set by the controller. It is determined by the “physics” of the system to be controlled. In finite time-resolution approach the digital controllers model the derivatives by finite differences, and the simplest numerical integration via Euler’s method also uses such differences. Whenever some fractional order derivative of the tracking error is fed back, this long memory term allows the realization of such control strategies for various integer order systems by appropriately masking” and weighting certain elements of longer sequences. On this mathematical basis the idea of controlling a higher order system by the application of a lower order strategy can be formulated, too. The main benefit may be achieving monotonic decrease in the components of the tracking error. In this paper the first order control of a popular second order system often used for benchmark applications, the van der Pol oscillator is considered. The realization of a simple proportional feedback first order controller in two variants, and a PID-type Computed Torque Control are considered in the fixed point iteration-based adaptive framework to tackle the problems caused by modeling imprecisions. The applicability and the limits of the investigated approaches are concluded.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"28 1","pages":"000119-000124"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73183427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029648
A. Pózna
In this paper, a special team orienteering problem is presented where a set of control points should be distributed among the members of the team while a course with fixed order of controls is need to be completed too. The aim of the team is to minimize the maximum route length of the members. In this paper, a simple genetic algorithm is proposed to solve this special team orienteering problem. The performance of the algorithm is evaluated on real competition examples and the results are compared to the solutions given by the real winners of the race.
{"title":"A Genetic Algorithm Based Solution of Hungarian Team Orienteering Competitions","authors":"A. Pózna","doi":"10.1109/CINTI-MACRo57952.2022.10029648","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029648","url":null,"abstract":"In this paper, a special team orienteering problem is presented where a set of control points should be distributed among the members of the team while a course with fixed order of controls is need to be completed too. The aim of the team is to minimize the maximum route length of the members. In this paper, a simple genetic algorithm is proposed to solve this special team orienteering problem. The performance of the algorithm is evaluated on real competition examples and the results are compared to the solutions given by the real winners of the race.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"4 1","pages":"000399-000404"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73281700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029600
Balázs Varga
Single rope ascending technique is used in industrial alpinism, forestry, or various leisure activities. This paper presents a multi-body model of this technique involving an actuated 3D model of a humanoid, the climbing gear, and the rope, modeled as a finite-element object. This model serves as a training ground for reinforcement learning agents trying to mimic humans in rope climbing. To demonstrate the environment, an agent with a state-of-the-art reinforcement learning algorithm (Soft Actor-Critic) was trained. Results suggest that the agent can learn how to ascend the rope with speed comparable to real humans. However, the learned technique is not human-like: the artificial agent uses its arms excessively to climb, which would be too tiring for a human. That is because the environment only rewards ascension and does not penalize the energy used. The presented learning environment is developed for humanoid robots in mind that can perform complex tasks while on the rope and can carry much heavier payloads compared to climbing robots in the literature.
{"title":"Learn to climb: teaching a reinforcement learning agent the single rope ascending technique","authors":"Balázs Varga","doi":"10.1109/CINTI-MACRo57952.2022.10029600","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029600","url":null,"abstract":"Single rope ascending technique is used in industrial alpinism, forestry, or various leisure activities. This paper presents a multi-body model of this technique involving an actuated 3D model of a humanoid, the climbing gear, and the rope, modeled as a finite-element object. This model serves as a training ground for reinforcement learning agents trying to mimic humans in rope climbing. To demonstrate the environment, an agent with a state-of-the-art reinforcement learning algorithm (Soft Actor-Critic) was trained. Results suggest that the agent can learn how to ascend the rope with speed comparable to real humans. However, the learned technique is not human-like: the artificial agent uses its arms excessively to climb, which would be too tiring for a human. That is because the environment only rewards ascension and does not penalize the energy used. The presented learning environment is developed for humanoid robots in mind that can perform complex tasks while on the rope and can carry much heavier payloads compared to climbing robots in the literature.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"14 1","pages":"000209-000214"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84724809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029518
S. Brassai, A. Németh, A. Hammas, Szabolcs Laszlo GABor
Drone imagery based object supervising has become more and more widespread. In the paper the Single shot Alignment Network is used to classify and localize the objects. The images were acquired by using two types of drones, DJI Tello and Zll SG906 Pro 2 in about thirty classes, and about nineteen were processed and detailed in the paper. The objects labeling was realized in CVAT labeling tool. For neural network management the mmdetection framework was used, and the obtained results were detailed on s2a-net. The paper focuses on the preparation of the neural network system to be used for agricultural machine detection. The network was trained on a PC with reduced processing capabilities. The dataset was cut in smaller tasks. An architecture is proposed to be used in future for dataset management during the training process.
基于无人机图像的目标监控越来越广泛。本文采用单镜头对准网络对目标进行分类和定位。图像是通过使用两种类型的无人机，DJI Tello和Zll SG906 Pro 2在大约30个类中获得的，并且在论文中对大约19个进行了处理和详细说明。在CVAT标注工具中实现对象标注。对于神经网络的管理，采用了mmdetection框架，并在s2a-net上详细介绍了得到的结果。本文重点研究了用于农业机械检测的神经网络系统的研制。该网络是在一台处理能力较差的个人电脑上进行训练的。数据集被分成更小的任务。提出了一种用于训练过程中数据集管理的体系结构。
{"title":"Deep learning based object detection for agricultural machinery","authors":"S. Brassai, A. Németh, A. Hammas, Szabolcs Laszlo GABor","doi":"10.1109/CINTI-MACRo57952.2022.10029518","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029518","url":null,"abstract":"Drone imagery based object supervising has become more and more widespread. In the paper the Single shot Alignment Network is used to classify and localize the objects. The images were acquired by using two types of drones, DJI Tello and Zll SG906 Pro 2 in about thirty classes, and about nineteen were processed and detailed in the paper. The objects labeling was realized in CVAT labeling tool. For neural network management the mmdetection framework was used, and the obtained results were detailed on s2a-net. The paper focuses on the preparation of the neural network system to be used for agricultural machine detection. The network was trained on a PC with reduced processing capabilities. The dataset was cut in smaller tasks. An architecture is proposed to be used in future for dataset management during the training process.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"20 1","pages":"000089-000094"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81547694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029496
E. Nagy, D. Drexler
Cyber-medical systems provides lots of possibilities that help doctors plan more effective treatments. A reliable mathematical model that can be customized is essential for therapy optimization. We deal with a mathematical model that we use to optimize chemotherapy. We have parameter sets that we use to create virtual patients and create therapy with random doses. Then we use a non-linear function optimization procedure with different initial values, Whose task is to fit the unknown parameters. Our goal is to examine the extent to which the procedure is able to find the real parameters of the virtual patients in the neighborhood of the original parameters. We found that there are parameters in the model where the parameter can not be found if the initial estimation is far from the real value.
{"title":"The effect of the choice of initial estimation for a tumor model parameter estimation problem","authors":"E. Nagy, D. Drexler","doi":"10.1109/CINTI-MACRo57952.2022.10029496","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029496","url":null,"abstract":"Cyber-medical systems provides lots of possibilities that help doctors plan more effective treatments. A reliable mathematical model that can be customized is essential for therapy optimization. We deal with a mathematical model that we use to optimize chemotherapy. We have parameter sets that we use to create virtual patients and create therapy with random doses. Then we use a non-linear function optimization procedure with different initial values, Whose task is to fit the unknown parameters. Our goal is to examine the extent to which the procedure is able to find the real parameters of the virtual patients in the neighborhood of the original parameters. We found that there are parameters in the model where the parameter can not be found if the initial estimation is far from the real value.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"10 1","pages":"000227-000232"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89818225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-21DOI: 10.1109/CINTI-MACRo57952.2022.10029559
Walelign Tewabe Sewunetie, L. Kovács
Automatic question generation is a technique that generates a question from various sources like structured or unstructured content. Even if, each method has its own shortcoming and strength, we didn’t find any study that did an experimental comparison of the current automatic question generation techniques using a similar dataset. In this study, we analyzed the different state-of-the-art research works, and we identified significant challenges of each model. In addition, we have used BLEU, METEOR, and ROUGE automatic evaluation metrics, to evaluate the experimental test result. From our evaluation result, we observed that most of the tested techniques score below 0.5 in all automated evaluation metrics. Out of tested techniques T5-transformer-based, scores the maximum result. This result point out this research area still needs further investigation and preparing standardized training.
{"title":"Comparison of Automatic Question Generation Techniques","authors":"Walelign Tewabe Sewunetie, L. Kovács","doi":"10.1109/CINTI-MACRo57952.2022.10029559","DOIUrl":"https://doi.org/10.1109/CINTI-MACRo57952.2022.10029559","url":null,"abstract":"Automatic question generation is a technique that generates a question from various sources like structured or unstructured content. Even if, each method has its own shortcoming and strength, we didn’t find any study that did an experimental comparison of the current automatic question generation techniques using a similar dataset. In this study, we analyzed the different state-of-the-art research works, and we identified significant challenges of each model. In addition, we have used BLEU, METEOR, and ROUGE automatic evaluation metrics, to evaluate the experimental test result. From our evaluation result, we observed that most of the tested techniques score below 0.5 in all automated evaluation metrics. Out of tested techniques T5-transformer-based, scores the maximum result. This result point out this research area still needs further investigation and preparing standardized training.","PeriodicalId":18535,"journal":{"name":"Micro","volume":"9 1","pages":"000025-000030"},"PeriodicalIF":0.0,"publicationDate":"2022-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87876686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：