When applied to an entire field, automation and autonomous systems are among the rare creative superpowers capable of catapulting progress at an exponential rate. The arrival of machine intelligence will give such automated machines the intelligence to perform their tasks with power of outcome, drastically reducing the need for human intervention in redundant processes. Large-scale technological progress can be traced back to responsibilities that are simplified and, as a result, more easily distinguished by means of automation. In accordance with these guidelines, we propose creating a product that eliminates or significantly reduces the need for human intervention in primary issue statements that can be automated and processed. The public safety infrastructure of today relies on surveillance cameras, but these devices are merely video recorders; they have no intelligence of their own. Automated video streams are now required for automatic event detection thanks to the massive amount of data produced by surveillance cameras. The project's main objective is to increase public safety through the mechanization of crime measurement and review using actual Closed-Circuit Television footage (CCTV). This is achieved by assigning the task of recognizing criminal behavior to a system that can do so automatically, allowing for more precise tracking. In this study, we present a model with a precision of 0.95 for assault and 0.97 for abuse.
{"title":"Deep Learning Algorithm for Detecting and Analyzing Criminal Activity","authors":"Raddam Sami Mehsen","doi":"10.47839/ijc.22.2.3095","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3095","url":null,"abstract":"When applied to an entire field, automation and autonomous systems are among the rare creative superpowers capable of catapulting progress at an exponential rate. The arrival of machine intelligence will give such automated machines the intelligence to perform their tasks with power of outcome, drastically reducing the need for human intervention in redundant processes. Large-scale technological progress can be traced back to responsibilities that are simplified and, as a result, more easily distinguished by means of automation. In accordance with these guidelines, we propose creating a product that eliminates or significantly reduces the need for human intervention in primary issue statements that can be automated and processed. The public safety infrastructure of today relies on surveillance cameras, but these devices are merely video recorders; they have no intelligence of their own. Automated video streams are now required for automatic event detection thanks to the massive amount of data produced by surveillance cameras. The project's main objective is to increase public safety through the mechanization of crime measurement and review using actual Closed-Circuit Television footage (CCTV). This is achieved by assigning the task of recognizing criminal behavior to a system that can do so automatically, allowing for more precise tracking. In this study, we present a model with a precision of 0.95 for assault and 0.97 for abuse.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90337012","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}
Prothama Sardar, Romana Rahman Ema, Sk. Shalauddin Kabir, Md. Nasim Adnan, S. Galib
In Bangladesh, most people depend on agriculture for their livelihood. The country's economy and agricultural production are hampered if plants are affected by diseases. Crop pests also disrupt agricultural production. So, this paper proposes leaf disease, disease severity stage, and pest detection strategies with suggestions for prevention strategies using five notable Convolutional Neural Network models (CNN) such as VGG16, Resnet50, AlexNet, EfficientNetB2, and EfficientNetB3. This paper uses a dataset of tomato leaves consisting of 41,763 images which cover 10 classes of tomato disease, and a dataset of pests consisting of 4,271 images which cover 8 classes of pests. Firstly, these models are used for the classification of diseases and pests. Then disease and pest prevention techniques are shown. For disease and pest detection, EfficientNetB3 gives the best accuracy for training (99.85%), (99.80%), and validation (97.85%), (97.45%) respectively. For severity stage identification, AlexNet gives the best accuracy for training (69.02%) and validation (72.49%).
{"title":"Severity Stage Identification and Pest Detection of Tomato Disease Using Deep Learning","authors":"Prothama Sardar, Romana Rahman Ema, Sk. Shalauddin Kabir, Md. Nasim Adnan, S. Galib","doi":"10.47839/ijc.22.2.3088","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3088","url":null,"abstract":"In Bangladesh, most people depend on agriculture for their livelihood. The country's economy and agricultural production are hampered if plants are affected by diseases. Crop pests also disrupt agricultural production. So, this paper proposes leaf disease, disease severity stage, and pest detection strategies with suggestions for prevention strategies using five notable Convolutional Neural Network models (CNN) such as VGG16, Resnet50, AlexNet, EfficientNetB2, and EfficientNetB3. This paper uses a dataset of tomato leaves consisting of 41,763 images which cover 10 classes of tomato disease, and a dataset of pests consisting of 4,271 images which cover 8 classes of pests. Firstly, these models are used for the classification of diseases and pests. Then disease and pest prevention techniques are shown. For disease and pest detection, EfficientNetB3 gives the best accuracy for training (99.85%), (99.80%), and validation (97.85%), (97.45%) respectively. For severity stage identification, AlexNet gives the best accuracy for training (69.02%) and validation (72.49%).","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88735392","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}
Zakaria Chabou, Abdessalam Aitmadi, A. Addaim, Z. Guennoun
The launch and successful operation of the Mars Cube One (MarCO) CubeSat in May 2018 heralded a new era in solar system exploration and the setup of the first Interplanetary CubeSat Network (ICN). The success of this mission could give rise of an Interplanetary DTN–Based CubeSat network, in which the CubeSat Nanosatellite, as DTN custody node, plays the role of Data Mule to collect data from rovers on a planet such as Mars. In order to maximize the contact volume which is the amount of transmitting data from rovers to the CubeSat during its pass over their service zone, we will need to design an efficient MAC protocol. This research focuses on the simulation and evaluation of the performance of the Slotted AlohaCA MAC Protocol on the planet Mars compared to Earth taking into account the different properties between the two planets, such as radius, mass and speed of rotation of the Nanosatellite in its orbital at the same altitude. We have conducted many simulations using the NS2 simulator that takes into consideration the spatial dynamic behavior of the Nanosatellite, which is dependent on motion of the Nanosatellite in its orbit. Three appropriate performance measures are evaluated: Throughput, stability and power consumption. The obtained simulation results on the planet Mars show an improvement on performance of the Slotted AlohaCA on the planet Mars compared to Earth.
{"title":"Performance Evaluation of Enhanced Slotted AlohaCA Protocol on Planet Mars","authors":"Zakaria Chabou, Abdessalam Aitmadi, A. Addaim, Z. Guennoun","doi":"10.47839/ijc.22.2.3090","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3090","url":null,"abstract":"The launch and successful operation of the Mars Cube One (MarCO) CubeSat in May 2018 heralded a new era in solar system exploration and the setup of the first Interplanetary CubeSat Network (ICN). The success of this mission could give rise of an Interplanetary DTN–Based CubeSat network, in which the CubeSat Nanosatellite, as DTN custody node, plays the role of Data Mule to collect data from rovers on a planet such as Mars. In order to maximize the contact volume which is the amount of transmitting data from rovers to the CubeSat during its pass over their service zone, we will need to design an efficient MAC protocol. This research focuses on the simulation and evaluation of the performance of the Slotted AlohaCA MAC Protocol on the planet Mars compared to Earth taking into account the different properties between the two planets, such as radius, mass and speed of rotation of the Nanosatellite in its orbital at the same altitude. We have conducted many simulations using the NS2 simulator that takes into consideration the spatial dynamic behavior of the Nanosatellite, which is dependent on motion of the Nanosatellite in its orbit. Three appropriate performance measures are evaluated: Throughput, stability and power consumption. The obtained simulation results on the planet Mars show an improvement on performance of the Slotted AlohaCA on the planet Mars compared to Earth.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80792894","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}
Malware detection remains an urgent task today. Various means for the development of information technology and providing users with useful applications are being transformed by attackers into tools for malicious influences and manifestations. A variety of countermeasures and detection tools have been developed to detect malware, but the problem of malware distribution remains relevant. It is especially important for enterprises and organizations. Their corporate networks and resources are becoming objects of interest to intruders. To counteract and prevent the effects of malware, they have various systems in place. In order to improve the counteraction to malicious influences and manifestations, the paper proposes the use of distributed discrete systems, in the architecture of which the principles of self-organization, adaptability and partial centralization are synthesized. Such tools and their functioning will be difficult to understand for attackers and, therefore, will be difficult to circumvent. The architecture of the proposed tools will integrate the implemented methods of malware detection for a holistic counteraction to malware. Such a system will be a single sensor that will detect malicious influences and anomalies. To organize its functioning, descriptions of characteristic indicators are needed. The paper presents the developed mathematical models for determining the values of characteristic indicators. According to obtained values the system architecture was formed. In order to evaluate the sustainability of the developed distributed discrete system a set of experiments were conducted. In addition, to study the accuracy of malware detection, the developed system was tested for the possibility of worm virus detection. Experimental studies have confirmed the effectiveness of the proposed solution, which makes it possible to use the obtained solutions for the development of such systems.
{"title":"Distributed Discrete Malware Detection Systems Based on Partial Centralization and Self-Organization","authors":"Sergii Lysenko, B. Savenko","doi":"10.47839/ijc.22.2.3082","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3082","url":null,"abstract":"Malware detection remains an urgent task today. Various means for the development of information technology and providing users with useful applications are being transformed by attackers into tools for malicious influences and manifestations. A variety of countermeasures and detection tools have been developed to detect malware, but the problem of malware distribution remains relevant. It is especially important for enterprises and organizations. Their corporate networks and resources are becoming objects of interest to intruders. To counteract and prevent the effects of malware, they have various systems in place. In order to improve the counteraction to malicious influences and manifestations, the paper proposes the use of distributed discrete systems, in the architecture of which the principles of self-organization, adaptability and partial centralization are synthesized. Such tools and their functioning will be difficult to understand for attackers and, therefore, will be difficult to circumvent. The architecture of the proposed tools will integrate the implemented methods of malware detection for a holistic counteraction to malware. Such a system will be a single sensor that will detect malicious influences and anomalies. To organize its functioning, descriptions of characteristic indicators are needed. The paper presents the developed mathematical models for determining the values of characteristic indicators. According to obtained values the system architecture was formed. In order to evaluate the sustainability of the developed distributed discrete system a set of experiments were conducted. In addition, to study the accuracy of malware detection, the developed system was tested for the possibility of worm virus detection. Experimental studies have confirmed the effectiveness of the proposed solution, which makes it possible to use the obtained solutions for the development of such systems.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72935517","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}
Computer measurement systems play an important role on process automation and Industry 4.0 implementation strategies. They can be easily integrated on modern production systems, enabling real time test and control of multiple product and process characteristics that need to be monitored. If for one side the big data provided by these systems is an important asset for production analytics and optimization, on the other hand, the high frequency data sampling, commonly used in these systems, can lead to autocorrelated data violating, this way, statistical independence requirements for statistical process control implementation. In this paper we present a simulation model, using digital recursive filters, to properly handle and deal with these issues. The model demonstrates how to eliminate the autocorrelation from data time series, creating and ensuring the conditions for statistical process control application through the application of real time control charts. A performance comparison between Shewhart of Residuals and Exponentially Weighted Moving Average (EWMA) of Individual Observations control charts is made for autocorrelated data time series with the presence of different mean shift amplitude perturbations.
{"title":"Real Time Statistical Process Control for Autocorrelated Serial Data: A Simulation Approach","authors":"Artur M. F. Graxinha, J. M. D. Dias Pereira","doi":"10.47839/ijc.22.2.3081","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3081","url":null,"abstract":"Computer measurement systems play an important role on process automation and Industry 4.0 implementation strategies. They can be easily integrated on modern production systems, enabling real time test and control of multiple product and process characteristics that need to be monitored. If for one side the big data provided by these systems is an important asset for production analytics and optimization, on the other hand, the high frequency data sampling, commonly used in these systems, can lead to autocorrelated data violating, this way, statistical independence requirements for statistical process control implementation. In this paper we present a simulation model, using digital recursive filters, to properly handle and deal with these issues. The model demonstrates how to eliminate the autocorrelation from data time series, creating and ensuring the conditions for statistical process control application through the application of real time control charts. A performance comparison between Shewhart of Residuals and Exponentially Weighted Moving Average (EWMA) of Individual Observations control charts is made for autocorrelated data time series with the presence of different mean shift amplitude perturbations.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81799334","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}
Getaneh Awulachew’ Zimbele, Samuel Asferaw Demilew
Cryptographic techniques in cyber security can be categorized into symmetric and asymmetric. Among asymmetric cryptographic techniques, the RSA algorithm is more popular and considered as secured. Since, RSA uses the common modulus in both encryption and decryption, this modulus is openly available for the public which makes it exposed for attack. Its security is based on the assumption of large integer factorization problem, but this could leave it open to different cryptanalysis attacks: low private exponent attack, Shor’s polynomial-time quantum algorithm, quantum inverse Fourier transform and phase estimation. To address these shortcomings, this paper proposes a public-key security algorithm called Hidden Real Modulus RSA (HRM-RSA) which hides real modulus by masking it. The public mask modulus which is a pseudo random masking number is derived from real modulus. Then, this derived public mask modulus is introduced in a public key component; as a result, a real modulus is kept hidden from the public unlike the case in RSA. Encryption is done using this public mask modulus and the decryption process is done using a private hidden real modulus. For performance analysis Net bean IDE 8.2 is used, and the proposed algorithm is compared with state-of-the-art algorithms: RSA, ESRKGS, and MRSA based on security strength, time complexity, key generation time, encryption speed, and decryption speed. The performance analysis shows that HRM-RSA is less complex but highly secured than existing algorithms. It improves key generation time of ESRKGS, and MRSA by 311%, 42%; encryption time of RSA, ESRKGS, MRSA by 0.7%, 139%, 735%; decryption time of RSA, ESRKGS, MRSA by 3%, 138%, 799%, respectively.
网络安全中的密码技术可分为对称密码和非对称密码。在非对称加密技术中，RSA算法比较流行，被认为是安全的。由于RSA在加密和解密中都使用公共模量，因此该模量对公众是公开可用的，这使得它暴露在攻击之下。它的安全性是基于大整数分解问题的假设，但这可能使它容易受到不同的密码分析攻击:低私有指数攻击、肖尔多项式时间量子算法、量子傅立叶反变换和相位估计。为了解决这些缺点，本文提出了一种通过屏蔽实模来隐藏实模的公钥安全算法——隐藏实模RSA (HRM-RSA)。公掩码模是由实模导出的伪随机掩码数。然后，将导出的公共掩码模引入到公钥组件中;因此，实模对公众是隐藏的，这与RSA中的情况不同。加密使用这个公共掩码模完成，解密过程使用一个私有的隐藏实模完成。为了进行性能分析，使用了Net bean IDE 8.2，并根据安全强度、时间复杂度、密钥生成时间、加密速度和解密速度，将所提出的算法与最先进的算法:RSA、ESRKGS和MRSA进行了比较。性能分析表明，与现有算法相比，HRM-RSA算法复杂度低，安全性高。将ESRKGS、MRSA的关键生成时间分别提高了31.1%、42%;RSA、ESRKGS、MRSA的加密时间分别提高0.7%、139%、735%;RSA、ESRKGS、MRSA的解密时间分别缩短了3%、138%、799%。
{"title":"Hidden Real Modulus RSA Cryptosystem","authors":"Getaneh Awulachew’ Zimbele, Samuel Asferaw Demilew","doi":"10.47839/ijc.22.2.3094","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3094","url":null,"abstract":"Cryptographic techniques in cyber security can be categorized into symmetric and asymmetric. Among asymmetric cryptographic techniques, the RSA algorithm is more popular and considered as secured. Since, RSA uses the common modulus in both encryption and decryption, this modulus is openly available for the public which makes it exposed for attack. Its security is based on the assumption of large integer factorization problem, but this could leave it open to different cryptanalysis attacks: low private exponent attack, Shor’s polynomial-time quantum algorithm, quantum inverse Fourier transform and phase estimation. To address these shortcomings, this paper proposes a public-key security algorithm called Hidden Real Modulus RSA (HRM-RSA) which hides real modulus by masking it. The public mask modulus which is a pseudo random masking number is derived from real modulus. Then, this derived public mask modulus is introduced in a public key component; as a result, a real modulus is kept hidden from the public unlike the case in RSA. Encryption is done using this public mask modulus and the decryption process is done using a private hidden real modulus. For performance analysis Net bean IDE 8.2 is used, and the proposed algorithm is compared with state-of-the-art algorithms: RSA, ESRKGS, and MRSA based on security strength, time complexity, key generation time, encryption speed, and decryption speed. The performance analysis shows that HRM-RSA is less complex but highly secured than existing algorithms. It improves key generation time of ESRKGS, and MRSA by 311%, 42%; encryption time of RSA, ESRKGS, MRSA by 0.7%, 139%, 735%; decryption time of RSA, ESRKGS, MRSA by 3%, 138%, 799%, respectively.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83575885","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}
In today's reality, the pace of people's lives is much higher than it was 30 years ago and it is still growing. At the same time, the amount of information is also growing. This information should be processed constantly, daily, as soon as it is received. Production volumes are not also standing still. Such a lively pace of life requires process consistency and continuity and these processes must be provided by a man. This article describes the system of watering which should automate the process of growing plants. Also, the analysis of a new branch, that is cyber-physical systems, is carried out. The analysis of modern systems of autonomous irrigation, principles of their construction and organization of their work is conducted. A method of implementing a system that provides the possibility of constant monitoring of the growing environment and provides an opportunity to influence it is suggested. The choice of components for system construction is made. The algorithm of the system operation is described. An analysis of the relationship between system components and the user's relationship with the system is performed.
{"title":"Drip Irrigation Cyber-physical System with Remote Control","authors":"Y. Klushyn","doi":"10.47839/ijc.22.2.3096","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3096","url":null,"abstract":"In today's reality, the pace of people's lives is much higher than it was 30 years ago and it is still growing. At the same time, the amount of information is also growing. This information should be processed constantly, daily, as soon as it is received. Production volumes are not also standing still. Such a lively pace of life requires process consistency and continuity and these processes must be provided by a man. \u0000This article describes the system of watering which should automate the process of growing plants. Also, the analysis of a new branch, that is cyber-physical systems, is carried out. The analysis of modern systems of autonomous irrigation, principles of their construction and organization of their work is conducted. A method of implementing a system that provides the possibility of constant monitoring of the growing environment and provides an opportunity to influence it is suggested. The choice of components for system construction is made. The algorithm of the system operation is described. An analysis of the relationship between system components and the user's relationship with the system is performed.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79605921","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}
A smart wearable ankle sprain prevention device would require an intelligent monitoring system that can classify data from the sensors as sprain or non-sprain motion. This paper aims to explore Deep Neural Network method, specifically the Long Short Term Memory Fully Convolutional Network (LSTM-FCN) for classifying sprain and non-sprain motion. A study is conducted on 11 participants to record sprain and non-sprain motions, which are used to train and test the LSTM-FCN model and previously used Support Vector Machine (SVM) model. It has been demonstrated that the LSTM-FCN model is more accurate at classifying sprain and non-sprain motion. The LSTM-FCN also proved to be more useful as its architecture allows for the Class Activation Mapping (CAM) method to be employed. The CAM method allows for the identification of temporal regions of the time series that contribute most or least to the classification decision of the LSTMFCN. Visualizing the regions of high or low contribution makes it easy to see patterns in the data correlation with sprain motion and better understand why certain non-sprain data can be misclassified as sprain motion. Overall, LSTM-FCN is found to be a viable method for the classification of sprain and non-sprain motion.
{"title":"Classification of Sprain and Non-sprain Motion using Deep Learning Neural Networks for Ankle Sprain Prevention","authors":"Natrisha Francis, Suhaimi Suhaimi, E. Abas","doi":"10.47839/ijc.22.2.3085","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3085","url":null,"abstract":"A smart wearable ankle sprain prevention device would require an intelligent monitoring system that can classify data from the sensors as sprain or non-sprain motion. This paper aims to explore Deep Neural Network method, specifically the Long Short Term Memory Fully Convolutional Network (LSTM-FCN) for classifying sprain and non-sprain motion. A study is conducted on 11 participants to record sprain and non-sprain motions, which are used to train and test the LSTM-FCN model and previously used Support Vector Machine (SVM) model. It has been demonstrated that the LSTM-FCN model is more accurate at classifying sprain and non-sprain motion. The LSTM-FCN also proved to be more useful as its architecture allows for the Class Activation Mapping (CAM) method to be employed. The CAM method allows for the identification of temporal regions of the time series that contribute most or least to the classification decision of the LSTMFCN. Visualizing the regions of high or low contribution makes it easy to see patterns in the data correlation with sprain motion and better understand why certain non-sprain data can be misclassified as sprain motion. Overall, LSTM-FCN is found to be a viable method for the classification of sprain and non-sprain motion.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83916330","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}
S. Abdulateef, A. N. Ismael, Mohanad Dawood Salman
The diagnosis of Parkinson has become easier with the existence of machine learning. It includes using existing features from the biometric dataset generated by the person to identify whether he has Parkinson or not. The features differ in their discrimination capability and they suffer from redundancy. Hence, researchers have recommended using feature selection for Parkinson's identification. The feature selection aims at finding the most important and relevant features to produce an efficient and effective model. In this article, we present entropy-based Parkinson classification. The goal is to select only 50% of the most relevant features for Parkinson prediction. Two variants of neural networks are used for evaluation, the first one is a feed-forward Extreme Learning Machine ELM and the second one is Fast Learning Machine FLN. Also, the K-Nearest Neighbor KNN algorithm is used for evaluation. The results show the superiority of ELM and FLN when the model of feature selection is used with an accuracy of 80% compared with only 78% when the model is not used.
{"title":"Feature Weighting for Parkinson's Identification using Single Hidden Layer Neural Network","authors":"S. Abdulateef, A. N. Ismael, Mohanad Dawood Salman","doi":"10.47839/ijc.22.2.3092","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3092","url":null,"abstract":"The diagnosis of Parkinson has become easier with the existence of machine learning. It includes using existing features from the biometric dataset generated by the person to identify whether he has Parkinson or not. The features differ in their discrimination capability and they suffer from redundancy. Hence, researchers have recommended using feature selection for Parkinson's identification. The feature selection aims at finding the most important and relevant features to produce an efficient and effective model. In this article, we present entropy-based Parkinson classification. The goal is to select only 50% of the most relevant features for Parkinson prediction. Two variants of neural networks are used for evaluation, the first one is a feed-forward Extreme Learning Machine ELM and the second one is Fast Learning Machine FLN. Also, the K-Nearest Neighbor KNN algorithm is used for evaluation. The results show the superiority of ELM and FLN when the model of feature selection is used with an accuracy of 80% compared with only 78% when the model is not used.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82214193","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}
The paper considers two forms of models: seasonal and non-seasonal analogues of oscillations. Additive models belong to the first form, which reflects a relatively constant seasonal wave, as well as a wave that dynamically changes depending on the trend. The second ones are multiplicative models. The paper analyzes the basic adaptive models: Brown, Holt and autoregression models. The parameters of adaptation and layout are considered by the method of numerical estimation of parameters. The mechanism of reflection of oscillatory (seasonal or cyclic) development of the studied process through reproduction of the scheme of moving average and the scheme of autoregression is analyzed. The paper determines the optimal value of the smoothing coefficient through adaptive polynomial models of the first and second order. Prediction using the Winters model (exponential smoothing with multiplicative seasonality and linear growth) is proposed. The application of the Winters model allows us to determine the calculated values and forecast using the model of exponential smoothing with multiplicative seasonality and linear growth. The results are calculated according to the model of exponential smoothing and with the multiplicative seasonality of Winters. The best model is determined, which allows improving the forecast results through the correct selection of the optimal value of α. The paper also forecasts the production volume according to the Tayle-Vage model, i.e., the analysis of exponential smoothing with additive seasonality and linear growth is given to determine the calculated values α. The paper proves that the additive model makes it possible to build a model with multiplicative seasonality and exponential tendency. The paper proves statements that allow one to choose the right method for better modeling and forecasting of data.
{"title":"Application of Adaptive and Multiplicative Models for Analysis and Forecasting of Time Series","authors":"N. Boyko","doi":"10.47839/ijc.22.2.3089","DOIUrl":"https://doi.org/10.47839/ijc.22.2.3089","url":null,"abstract":"The paper considers two forms of models: seasonal and non-seasonal analogues of oscillations. Additive models belong to the first form, which reflects a relatively constant seasonal wave, as well as a wave that dynamically changes depending on the trend. The second ones are multiplicative models. The paper analyzes the basic adaptive models: Brown, Holt and autoregression models. The parameters of adaptation and layout are considered by the method of numerical estimation of parameters. The mechanism of reflection of oscillatory (seasonal or cyclic) development of the studied process through reproduction of the scheme of moving average and the scheme of autoregression is analyzed. The paper determines the optimal value of the smoothing coefficient through adaptive polynomial models of the first and second order. Prediction using the Winters model (exponential smoothing with multiplicative seasonality and linear growth) is proposed. The application of the Winters model allows us to determine the calculated values and forecast using the model of exponential smoothing with multiplicative seasonality and linear growth. The results are calculated according to the model of exponential smoothing and with the multiplicative seasonality of Winters. The best model is determined, which allows improving the forecast results through the correct selection of the optimal value of α. The paper also forecasts the production volume according to the Tayle-Vage model, i.e., the analysis of exponential smoothing with additive seasonality and linear growth is given to determine the calculated values α. The paper proves that the additive model makes it possible to build a model with multiplicative seasonality and exponential tendency. The paper proves statements that allow one to choose the right method for better modeling and forecasting of data.","PeriodicalId":37669,"journal":{"name":"International Journal of Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90392249","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：