Pub Date : 2022-10-01DOI: 10.2478/jaiscr-2022-0018
Marcin Zalasiński, Łukasz Laskowski, Tacjana Niksa-Rynkiewicz, K. Cpałka, A. Byrski, Krzysztof Przybyszewski, Paweł Trippner, Shiquan Dong
Abstract In the verification of identity, the aim is to increase effectiveness and reduce involvement of verified users. A good compromise between these issues is ensured by dynamic signature verification. The dynamic signature is represented by signals describing the position of the stylus in time. They can be used to determine the velocity or acceleration signal. Values of these signals can be analyzed, interpreted, selected, and compared. In this paper, we propose an approach that: (a) uses an evolutionary algorithm to create signature partitions in the time and velocity domains; (b) selects the most characteristic partitions in terms of matching with reference signatures; and (c) works individually for each user, eliminating the need of using skilled forgeries. The proposed approach was tested using Biosecure DS2 database which is a part of the DeepSignDB, a database with genuine dynamic signatures. Our simulations confirmed the correctness of the adopted assumptions.
{"title":"Evolutionary Algorithm for Selecting Dynamic Signatures Partitioning Approach","authors":"Marcin Zalasiński, Łukasz Laskowski, Tacjana Niksa-Rynkiewicz, K. Cpałka, A. Byrski, Krzysztof Przybyszewski, Paweł Trippner, Shiquan Dong","doi":"10.2478/jaiscr-2022-0018","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0018","url":null,"abstract":"Abstract In the verification of identity, the aim is to increase effectiveness and reduce involvement of verified users. A good compromise between these issues is ensured by dynamic signature verification. The dynamic signature is represented by signals describing the position of the stylus in time. They can be used to determine the velocity or acceleration signal. Values of these signals can be analyzed, interpreted, selected, and compared. In this paper, we propose an approach that: (a) uses an evolutionary algorithm to create signature partitions in the time and velocity domains; (b) selects the most characteristic partitions in terms of matching with reference signatures; and (c) works individually for each user, eliminating the need of using skilled forgeries. The proposed approach was tested using Biosecure DS2 database which is a part of the DeepSignDB, a database with genuine dynamic signatures. Our simulations confirmed the correctness of the adopted assumptions.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"267 - 279"},"PeriodicalIF":2.8,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47220846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.2478/jaiscr-2022-0012
J. Bilski, Bartosz Kowalczyk, Marek Kisiel-Dorohinicki, A. Siwocha, J. Zurada
Abstract **This paper presents a novel fast algorithm for feedforward neural networks training. It is based on the Recursive Least Squares (RLS) method commonly used for designing adaptive filters. Besides, it utilizes two techniques of linear algebra, namely the orthogonal transformation method, called the Givens Rotations (GR), and the QR decomposition, creating the GQR (symbolically we write GR + QR = GQR) procedure for solving the normal equations in the weight update process. In this paper, a novel approach to the GQR algorithm is presented. The main idea revolves around reducing the computational cost of a single rotation by eliminating the square root calculation and reducing the number of multiplications. The proposed modification is based on the scaled version of the Givens rotations, denoted as SGQR. This modification is expected to bring a significant training time reduction comparing to the classic GQR algorithm. The paper begins with the introduction and the classic Givens rotation description. Then, the scaled rotation and its usage in the QR decomposition is discussed. The main section of the article presents the neural network training algorithm which utilizes scaled Givens rotations and QR decomposition in the weight update process. Next, the experiment results of the proposed algorithm are presented and discussed. The experiment utilizes several benchmarks combined with neural networks of various topologies. It is shown that the proposed algorithm outperforms several other commonly used methods, including well known Adam optimizer.
{"title":"Towards a Very Fast Feedforward Multilayer Neural Networks Training Algorithm","authors":"J. Bilski, Bartosz Kowalczyk, Marek Kisiel-Dorohinicki, A. Siwocha, J. Zurada","doi":"10.2478/jaiscr-2022-0012","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0012","url":null,"abstract":"Abstract **This paper presents a novel fast algorithm for feedforward neural networks training. It is based on the Recursive Least Squares (RLS) method commonly used for designing adaptive filters. Besides, it utilizes two techniques of linear algebra, namely the orthogonal transformation method, called the Givens Rotations (GR), and the QR decomposition, creating the GQR (symbolically we write GR + QR = GQR) procedure for solving the normal equations in the weight update process. In this paper, a novel approach to the GQR algorithm is presented. The main idea revolves around reducing the computational cost of a single rotation by eliminating the square root calculation and reducing the number of multiplications. The proposed modification is based on the scaled version of the Givens rotations, denoted as SGQR. This modification is expected to bring a significant training time reduction comparing to the classic GQR algorithm. The paper begins with the introduction and the classic Givens rotation description. Then, the scaled rotation and its usage in the QR decomposition is discussed. The main section of the article presents the neural network training algorithm which utilizes scaled Givens rotations and QR decomposition in the weight update process. Next, the experiment results of the proposed algorithm are presented and discussed. The experiment utilizes several benchmarks combined with neural networks of various topologies. It is shown that the proposed algorithm outperforms several other commonly used methods, including well known Adam optimizer.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"181 - 195"},"PeriodicalIF":2.8,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42816278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.2478/jaiscr-2022-0013
Janusz T. Starczewski, K. Przybyszewski, A. Byrski, E. Szmidt, Christian Napoli
Abstract Fuzzy logic systems, unlike black-box models, are known as transparent artificial intelligence systems that have explainable rules of reasoning. Type 2 fuzzy systems extend the field of application to tasks that require the introduction of uncertainty in the rules, e.g. for handling corrupted data. Most practical implementations use interval type-2 sets and process interval membership grades. The key role in the design of type-2 interval fuzzy logic systems is played by the type-2 inference defuzzification method. In type-2 systems this generally takes place in two steps: type-reduction first, then standard defuzzification. The only precise type-reduction method is the iterative method known as Karnik-Mendel (KM) algorithm with its enhancement modifications. The known non-iterative methods deliver only an approximation of the boundaries of a type-reduced set and, in special cases, they diminish the profits that result from the use of type-2 fuzzy logic systems. In this paper, we propose a novel type-reduction method based on a smooth approximation of maximum/minimum, and we call this method a smooth type-reduction. Replacing the iterative KM algorithm by the smooth type-reduction, we obtain a structure of an adaptive interval type-2 fuzzy logic which is non-iterative and as close to an approximation of the KM algorithm as we like.
{"title":"A Novel Approach to Type-Reduction and Design of Interval Type-2 Fuzzy Logic Systems","authors":"Janusz T. Starczewski, K. Przybyszewski, A. Byrski, E. Szmidt, Christian Napoli","doi":"10.2478/jaiscr-2022-0013","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0013","url":null,"abstract":"Abstract Fuzzy logic systems, unlike black-box models, are known as transparent artificial intelligence systems that have explainable rules of reasoning. Type 2 fuzzy systems extend the field of application to tasks that require the introduction of uncertainty in the rules, e.g. for handling corrupted data. Most practical implementations use interval type-2 sets and process interval membership grades. The key role in the design of type-2 interval fuzzy logic systems is played by the type-2 inference defuzzification method. In type-2 systems this generally takes place in two steps: type-reduction first, then standard defuzzification. The only precise type-reduction method is the iterative method known as Karnik-Mendel (KM) algorithm with its enhancement modifications. The known non-iterative methods deliver only an approximation of the boundaries of a type-reduced set and, in special cases, they diminish the profits that result from the use of type-2 fuzzy logic systems. In this paper, we propose a novel type-reduction method based on a smooth approximation of maximum/minimum, and we call this method a smooth type-reduction. Replacing the iterative KM algorithm by the smooth type-reduction, we obtain a structure of an adaptive interval type-2 fuzzy logic which is non-iterative and as close to an approximation of the KM algorithm as we like.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"197 - 206"},"PeriodicalIF":2.8,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45365530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.2478/jaiscr-2022-0015
Pingping Liu, Jinde Cao, Yiping Luo, J. Guo, Wei Huang
Abstract This paper constructs and settles a charging facility location problem with the link capacity constraint over a mixed traffic network. The reason for studying this problem is that link capacity constraint is mostly insufficient or missing in the studies of traditional user equilibrium models, thereby resulting in the ambiguous of the definition of road traffic network status. Adding capacity constraints to the road network is a compromise to enhance the reality of the traditional equilibrium model. In this paper, we provide a two-layer model for evaluating the efficiency of the charging facilities under the condition of considering the link capacity constraint. The upper level model in the proposed bi-level model is a nonlinear integer programming formulation, which aims to maximize the captured link flows of the battery electric vehicles. Moreover, the lower level model is a typical traffic equilibrium assignment model except that it contains the link capacity constraint and driving distance constraint of the electric vehicles over the mixed road network. Based on the Frank-Wolfe algorithm, a modified algorithm framework is adopted for solving the constructed problem, and finally, a numerical example is presented to verify the proposed model and solution algorithm.
{"title":"Flow-Capture Location Model with Link Capacity Constraint Over a Mixed Traffic Network","authors":"Pingping Liu, Jinde Cao, Yiping Luo, J. Guo, Wei Huang","doi":"10.2478/jaiscr-2022-0015","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0015","url":null,"abstract":"Abstract This paper constructs and settles a charging facility location problem with the link capacity constraint over a mixed traffic network. The reason for studying this problem is that link capacity constraint is mostly insufficient or missing in the studies of traditional user equilibrium models, thereby resulting in the ambiguous of the definition of road traffic network status. Adding capacity constraints to the road network is a compromise to enhance the reality of the traditional equilibrium model. In this paper, we provide a two-layer model for evaluating the efficiency of the charging facilities under the condition of considering the link capacity constraint. The upper level model in the proposed bi-level model is a nonlinear integer programming formulation, which aims to maximize the captured link flows of the battery electric vehicles. Moreover, the lower level model is a typical traffic equilibrium assignment model except that it contains the link capacity constraint and driving distance constraint of the electric vehicles over the mixed road network. Based on the Frank-Wolfe algorithm, a modified algorithm framework is adopted for solving the constructed problem, and finally, a numerical example is presented to verify the proposed model and solution algorithm.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"223 - 234"},"PeriodicalIF":2.8,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44729704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.2478/jaiscr-2022-0014
Yang Cao, K. Maheswari, S. Dharani, K. Sivaranjani
Abstract This paper investigates the event-based state estimation for discrete-time recurrent delayed semi-Markovian neural networks. An event-triggering protocol is introduced to find measurement output with a specific triggering condition so as to lower the burden of the data communication. A novel summation inequality is established for the existence of asymptotic stability of the estimation error system. The problem addressed here is to construct an H∞ state estimation that guarantees the asymptotic stability with the novel summation inequality, characterized by event-triggered transmission. By the Lyapunov functional technique, the explicit expressions for the gain are established. Finally, two examples are exploited numerically to illustrate the usefulness of the new methodology.
{"title":"New Event Based H∞ State Estimation for Discrete-Time Recurrent Delayed Semi-Markov Jump Neural Networks Via a Novel Summation Inequality","authors":"Yang Cao, K. Maheswari, S. Dharani, K. Sivaranjani","doi":"10.2478/jaiscr-2022-0014","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0014","url":null,"abstract":"Abstract This paper investigates the event-based state estimation for discrete-time recurrent delayed semi-Markovian neural networks. An event-triggering protocol is introduced to find measurement output with a specific triggering condition so as to lower the burden of the data communication. A novel summation inequality is established for the existence of asymptotic stability of the estimation error system. The problem addressed here is to construct an H∞ state estimation that guarantees the asymptotic stability with the novel summation inequality, characterized by event-triggered transmission. By the Lyapunov functional technique, the explicit expressions for the gain are established. Finally, two examples are exploited numerically to illustrate the usefulness of the new methodology.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"207 - 221"},"PeriodicalIF":2.8,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44990952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.2478/jaiscr-2022-0011
Guangyi Chen, A. Krzyżak, P. Duda, A. Cader
Abstract Recognizing faces under various lighting conditions is a challenging problem in artificial intelligence and applications. In this paper we describe a new face recognition algorithm which is invariant to illumination. We first convert image files to the logarithm domain and then we implement them using the dual-tree complex wavelet transform (DTCWT) which yields images approximately invariant to changes in illumination change. We classify the images by the collaborative representation-based classifier (CRC). We also perform the following sub-band transformations: (i) we set the approximation sub-band to zero if the noise standard deviation is greater than 5; (ii) we then threshold the two highest frequency wavelet sub-bands using bivariate wavelet shrinkage. (iii) otherwise, we set these two highest frequency wavelet sub-bands to zero. On obtained images we perform the inverse DTCWT which results in illumination invariant face images. The proposed method is strongly robust to Gaussian white noise. Experimental results show that our proposed algorithm outperforms several existing methods on the Extended Yale Face Database B and the CMU-PIE face database.
{"title":"Noise Robust Illumination Invariant Face Recognition Via Bivariate Wavelet Shrinkage in Logarithm Domain","authors":"Guangyi Chen, A. Krzyżak, P. Duda, A. Cader","doi":"10.2478/jaiscr-2022-0011","DOIUrl":"https://doi.org/10.2478/jaiscr-2022-0011","url":null,"abstract":"Abstract Recognizing faces under various lighting conditions is a challenging problem in artificial intelligence and applications. In this paper we describe a new face recognition algorithm which is invariant to illumination. We first convert image files to the logarithm domain and then we implement them using the dual-tree complex wavelet transform (DTCWT) which yields images approximately invariant to changes in illumination change. We classify the images by the collaborative representation-based classifier (CRC). We also perform the following sub-band transformations: (i) we set the approximation sub-band to zero if the noise standard deviation is greater than 5; (ii) we then threshold the two highest frequency wavelet sub-bands using bivariate wavelet shrinkage. (iii) otherwise, we set these two highest frequency wavelet sub-bands to zero. On obtained images we perform the inverse DTCWT which results in illumination invariant face images. The proposed method is strongly robust to Gaussian white noise. Experimental results show that our proposed algorithm outperforms several existing methods on the Extended Yale Face Database B and the CMU-PIE face database.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"12 1","pages":"169 - 180"},"PeriodicalIF":2.8,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49654176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Due to the nature of hospital emergency services, density cannot be easily estimated. It is one of the important issues that should be planned for emergency service managers to have sufficient resources continuously in services that develop suddenly, and emergency interventions are made for human life. Effective and efficient management and planning of limited resources are important not only for hospital administrators but also for people who will receive service from emergency services. In this situation, estimating the number of people who will request service in the emergency service with the least error is of great importance in terms of resource management and the operations carried out in the emergency services. The density of patients coming to the emergency department may vary according to the season, special dates, and even time zones during the day. The aim of the study is to show that more successful results will be obtained because of processing the time series by considering the country and area-specific features instead of the traditional approach. In this paper, the patient admission dataset of the public hospital emergency service in Turkey was used. Data cleaning and arranging operations were carried out by considering the official and religious special days of Turkey and the time periods during the day. The data set is first handled holistically, and its performances are measured by making predictions with the LSTM (Long Short Term Memory) model. Then, to examine the effect of time zones, performance values were calculated separately by dividing each day into 3 equal time zones. Finally, to investigate the effect of triage areas on the total density, the model performance was measured by dividing the data forming each time zone into 3 different triage areas in 3 equal time periods. Three stages were applied both on the raw data set and on the data created by extracting the official, religious holidays, and weekend data specific to Turkey. According to the MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error) results, more successful results are obtained thanks to the cleaning and editing processes. Thanks to the study, it is thought that the data sets used for demand forecasting studies in the health sector will produce results closer to reality by determining and standardizing the purification criteria in this way.
{"title":"Time Series Cleaning Methods for Hospital Emergency Admissions","authors":"Yigit Alisan, Olcay Tosun","doi":"10.55195/jscai.1126611","DOIUrl":"https://doi.org/10.55195/jscai.1126611","url":null,"abstract":"Due to the nature of hospital emergency services, density cannot be easily estimated. It is one of the important issues that should be planned for emergency service managers to have sufficient resources continuously in services that develop suddenly, and emergency interventions are made for human life. Effective and efficient management and planning of limited resources are important not only for hospital administrators but also for people who will receive service from emergency services. In this situation, estimating the number of people who will request service in the emergency service with the least error is of great importance in terms of resource management and the operations carried out in the emergency services. The density of patients coming to the emergency department may vary according to the season, special dates, and even time zones during the day. The aim of the study is to show that more successful results will be obtained because of processing the time series by considering the country and area-specific features instead of the traditional approach. In this paper, the patient admission dataset of the public hospital emergency service in Turkey was used. Data cleaning and arranging operations were carried out by considering the official and religious special days of Turkey and the time periods during the day. The data set is first handled holistically, and its performances are measured by making predictions with the LSTM (Long Short Term Memory) model. Then, to examine the effect of time zones, performance values were calculated separately by dividing each day into 3 equal time zones. Finally, to investigate the effect of triage areas on the total density, the model performance was measured by dividing the data forming each time zone into 3 different triage areas in 3 equal time periods. Three stages were applied both on the raw data set and on the data created by extracting the official, religious holidays, and weekend data specific to Turkey. According to the MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error) results, more successful results are obtained thanks to the cleaning and editing processes. Thanks to the study, it is thought that the data sets used for demand forecasting studies in the health sector will produce results closer to reality by determining and standardizing the purification criteria in this way.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"25 1","pages":""},"PeriodicalIF":2.8,"publicationDate":"2022-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77308316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper aims to investigate the performances of a semi-active tuned mass damper (STMD) used to reduce the vibrations of buildings under different seismic excitation by the real-time hybrid simulation (RTHS) method. MR damper is used as a semi-active control element in the STMD. The RTHS method is an alternative to experimentally studying the semi-active tuned mass dampers (STMD) system. MR damper, which is critically significant for the system, is experimentally installed, while the other parts are designed in numerical simulation and tested simultaneously. The Groundhook control algorithm determines the voltage transmitted to the MR damper. The results show that the control method applied to MR damper-controlled STMD effectively suppresses structural vibrations.
{"title":"Hybrid experimental investigation of MR damper controlled tuned mass damper used for structures under earthquakes","authors":"Hüseyin Aggümüş, R. Güçlü","doi":"10.55195/jscai.1122514","DOIUrl":"https://doi.org/10.55195/jscai.1122514","url":null,"abstract":"This paper aims to investigate the performances of a semi-active tuned mass damper (STMD) used to reduce the vibrations of buildings under different seismic excitation by the real-time hybrid simulation (RTHS) method. MR damper is used as a semi-active control element in the STMD. The RTHS method is an alternative to experimentally studying the semi-active tuned mass dampers (STMD) system. MR damper, which is critically significant for the system, is experimentally installed, while the other parts are designed in numerical simulation and tested simultaneously. The Groundhook control algorithm determines the voltage transmitted to the MR damper. The results show that the control method applied to MR damper-controlled STMD effectively suppresses structural vibrations.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"31 1","pages":""},"PeriodicalIF":2.8,"publicationDate":"2022-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75635518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Coronavirus (COVID-19) epidemic emerged in China and has caused many problems such as loss of life, and deterioration of social and economic structure. Thus, understanding and predicting the course of the epidemic is very important. In this study, SEIR model and machine learning methods LSTM and SVM were used to predict the values of Susceptible, Exposed, Infected, and Recovered for COVID-19. For this purpose, COVID-19 data of Egypt and South Korea provided by John Hopkins University were used. The results of the methods were compared by using MAPE. Total 79% of MAPE were between 0-10. The comparisons show that although LSTM provided the better results, the results of all three methods were successful in predicting the number of cases, the number of patients who died, the peaks and dimensions of the epidemic.
{"title":"Applying Machine Learning Prediction Methods to COVID-19 Data","authors":"Faruk Serin, Adnan Kece, Yigit Alisan","doi":"10.55195/jscai.1108528","DOIUrl":"https://doi.org/10.55195/jscai.1108528","url":null,"abstract":"The Coronavirus (COVID-19) epidemic emerged in China and has caused many problems such as loss of life, and deterioration of social and economic structure. Thus, understanding and predicting the course of the epidemic is very important. In this study, SEIR model and machine learning methods LSTM and SVM were used to predict the values of Susceptible, Exposed, Infected, and Recovered for COVID-19. For this purpose, COVID-19 data of Egypt and South Korea provided by John Hopkins University were used. The results of the methods were compared by using MAPE. Total 79% of MAPE were between 0-10. The comparisons show that although LSTM provided the better results, the results of all three methods were successful in predicting the number of cases, the number of patients who died, the peaks and dimensions of the epidemic.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"76 1","pages":""},"PeriodicalIF":2.8,"publicationDate":"2022-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83310531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, studies on the elements used in producing electronic device components and the interaction of their different compounds with radiation have been emphasized. In developing this situation, giving importance to space studies and different searches in energy production have been very effective. In the light of these developments, the interaction of tin, which is widely used in the production of electronic device components and different industrial areas, with radiation has been investigated. For this purpose, the variation of albedo factor values in some compounds of the tin element was investigated and presented. Am-241 radioactive sources were used as incident radiation in determining the albedo factor values mentioned in the study. The albedo factor values obtained by examining the Compton and coherent scattering peaks were used to determine the albedo factor values.
{"title":"Investigation of Albedo Factor Parameters in Some Selected Sn Compounds","authors":"A. Turşucu","doi":"10.55195/jscai.1118684","DOIUrl":"https://doi.org/10.55195/jscai.1118684","url":null,"abstract":"In recent years, studies on the elements used in producing electronic device components and the interaction of their different compounds with radiation have been emphasized. In developing this situation, giving importance to space studies and different searches in energy production have been very effective. In the light of these developments, the interaction of tin, which is widely used in the production of electronic device components and different industrial areas, with radiation has been investigated. For this purpose, the variation of albedo factor values in some compounds of the tin element was investigated and presented. Am-241 radioactive sources were used as incident radiation in determining the albedo factor values mentioned in the study. The albedo factor values obtained by examining the Compton and coherent scattering peaks were used to determine the albedo factor values.","PeriodicalId":48494,"journal":{"name":"Journal of Artificial Intelligence and Soft Computing Research","volume":"33 1","pages":""},"PeriodicalIF":2.8,"publicationDate":"2022-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72657244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","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：