Abstract The traditional recognition method of whitewash behavior of accounting statements needs to analyze a large number of special data samples. The learning rate of the algorithm is low, resulting in low recognition accuracy. To solve the aforementioned problems, this article proposes a method to identify the whitewash behavior of university accounting statements based on the FCM clustering algorithm. This article analyzes the motivation of university accounting statement whitewashing behavior, studies the common means of statement whitewashing, and establishes a fuzzy set for the identification of university accounting statement whitewashing behavior. By calculating the fuzzy partition coefficient, the membership matrix of whitewash behavior recognition is established, and the whitewash behavior is classified through the iteration of the FCM algorithm. The comparative experimental results show that the recognition method has good recognition performance, low recognition error rate, and recognition accuracy of 82%.
{"title":"An FCM clustering algorithm based on the identification of accounting statement whitewashing behavior in universities","authors":"Qihao Yang","doi":"10.1515/jisys-2022-0022","DOIUrl":"https://doi.org/10.1515/jisys-2022-0022","url":null,"abstract":"Abstract The traditional recognition method of whitewash behavior of accounting statements needs to analyze a large number of special data samples. The learning rate of the algorithm is low, resulting in low recognition accuracy. To solve the aforementioned problems, this article proposes a method to identify the whitewash behavior of university accounting statements based on the FCM clustering algorithm. This article analyzes the motivation of university accounting statement whitewashing behavior, studies the common means of statement whitewashing, and establishes a fuzzy set for the identification of university accounting statement whitewashing behavior. By calculating the fuzzy partition coefficient, the membership matrix of whitewash behavior recognition is established, and the whitewash behavior is classified through the iteration of the FCM algorithm. The comparative experimental results show that the recognition method has good recognition performance, low recognition error rate, and recognition accuracy of 82%.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"32 1","pages":"345 - 355"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76693335","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}
Abstract The increase in the size of universities has greatly increased the number of teachers, students, and courses and has also increased the difficulty of scheduling courses. This study used coevolution to improve the genetic algorithm and applied it to solve the course scheduling problem in universities. Finally, simulation experiments were conducted on the traditional and improved genetic algorithms in MATLAB software. The results showed that the improved genetic algorithm converged faster and produced better solutions than the traditional genetic algorithm under the same crossover and mutation probability. As the mutation probability in the algorithm increased, the fitness values of both genetic algorithms gradually decreased, and the computation time increased. With the increase in crossover probability in the algorithm, the fitness value of the two genetic algorithms increased first and then decreased, and the computational time decreased first and then increased.
{"title":"An optimized solution to the course scheduling problem in universities under an improved genetic algorithm","authors":"Qiang Zhang","doi":"10.1515/jisys-2022-0114","DOIUrl":"https://doi.org/10.1515/jisys-2022-0114","url":null,"abstract":"Abstract The increase in the size of universities has greatly increased the number of teachers, students, and courses and has also increased the difficulty of scheduling courses. This study used coevolution to improve the genetic algorithm and applied it to solve the course scheduling problem in universities. Finally, simulation experiments were conducted on the traditional and improved genetic algorithms in MATLAB software. The results showed that the improved genetic algorithm converged faster and produced better solutions than the traditional genetic algorithm under the same crossover and mutation probability. As the mutation probability in the algorithm increased, the fitness values of both genetic algorithms gradually decreased, and the computation time increased. With the increase in crossover probability in the algorithm, the fitness value of the two genetic algorithms increased first and then decreased, and the computational time decreased first and then increased.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"6 1","pages":"1065 - 1073"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80287301","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}
Abstract Internet marketing requires a personalized marketing strategy. In this study, the application of data mining in personalized Internet marketing was studied. Based on the mining algorithm, a personalized marketing method was designed. Through the calculation of frequent closed item sets and support counts of positive and negative samples, the interval with a high success rate for marketing was obtained. With performance analysis, it was found that the success rate of the marketing method proposed in this study improved 8% compared with the traditional marketing method and had a better performance under the smaller interval number and smaller minimum success number. After applying the designed method in telecommunication enterprise A, it was found that after adopting the marketing method of this study, the marketing success rate of enterprise A increased from 2.72 to 6.31%, which indicated the effectiveness of the method. The research results of this study verify the role of data mining algorithms in Internet marketing, which is conducive to the further application of mining algorithms in personalized marketing and innovation of business modes.
{"title":"Application of mining algorithm in personalized Internet marketing strategy in massive data environment","authors":"Qianqian Pan, Gang Yang","doi":"10.1515/jisys-2022-0014","DOIUrl":"https://doi.org/10.1515/jisys-2022-0014","url":null,"abstract":"Abstract Internet marketing requires a personalized marketing strategy. In this study, the application of data mining in personalized Internet marketing was studied. Based on the mining algorithm, a personalized marketing method was designed. Through the calculation of frequent closed item sets and support counts of positive and negative samples, the interval with a high success rate for marketing was obtained. With performance analysis, it was found that the success rate of the marketing method proposed in this study improved 8% compared with the traditional marketing method and had a better performance under the smaller interval number and smaller minimum success number. After applying the designed method in telecommunication enterprise A, it was found that after adopting the marketing method of this study, the marketing success rate of enterprise A increased from 2.72 to 6.31%, which indicated the effectiveness of the method. The research results of this study verify the role of data mining algorithms in Internet marketing, which is conducive to the further application of mining algorithms in personalized marketing and innovation of business modes.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"126 1","pages":"237 - 244"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80390329","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}
Abstract The parking of cars is a globally recognized problem, especially at locations where there is a high demand for empty parking spaces. Drivers tend to cruise additional distances while searching for empty parking spaces during peak hours leading to problems, such as pollution, congestion, and driver frustration. Providing short-term predictions of parking availability would facilitate the driver in making informed decisions and planning their arrival to be able to choose parking locations with higher availability. Therefore, the aim of this study is to provide short-term predictions of available parking spaces with a low volume of data. The open parking lot provides parking spaces free of charge and one such parking lot, located beside a shopping center, was selected for this study. Parking availability data for 21 days were collected where 19 days were used for training, while multiple periods of the remaining 2 days were used to test and evaluate the prediction methods. The test dataset consists of data from a weekday and a weekend. Based on the reviewed literature, three prediction methods suitable for short-term prediction were selected, namely, long short-term memory (LSTM), seasonal autoregressive integrated moving average with exogenous variables (SARIMAX), and the Ensemble-based method. The LSTM method is a deep learning-based method, while SARIMAX is a regression-based method, and the Ensemble method is based on decision trees and random forest to provide predictions. The performance of the three prediction methods with a low volume of data and the use of visitor trends data as an exogenous variable was evaluated. Based on the test prediction results, the LSTM and Ensemble-based methods provided better short-term predictions at multiple times on a weekday, while the Ensemble-based method provided better predictions over the weekend. However, the use of visitor trend data did not facilitate improving the predictions of SARIMAX and the Ensemble-based method, while it improved the LSTM prediction for the weekend.
{"title":"Short-term prediction of parking availability in an open parking lot","authors":"V. Paidi","doi":"10.1515/jisys-2022-0039","DOIUrl":"https://doi.org/10.1515/jisys-2022-0039","url":null,"abstract":"Abstract The parking of cars is a globally recognized problem, especially at locations where there is a high demand for empty parking spaces. Drivers tend to cruise additional distances while searching for empty parking spaces during peak hours leading to problems, such as pollution, congestion, and driver frustration. Providing short-term predictions of parking availability would facilitate the driver in making informed decisions and planning their arrival to be able to choose parking locations with higher availability. Therefore, the aim of this study is to provide short-term predictions of available parking spaces with a low volume of data. The open parking lot provides parking spaces free of charge and one such parking lot, located beside a shopping center, was selected for this study. Parking availability data for 21 days were collected where 19 days were used for training, while multiple periods of the remaining 2 days were used to test and evaluate the prediction methods. The test dataset consists of data from a weekday and a weekend. Based on the reviewed literature, three prediction methods suitable for short-term prediction were selected, namely, long short-term memory (LSTM), seasonal autoregressive integrated moving average with exogenous variables (SARIMAX), and the Ensemble-based method. The LSTM method is a deep learning-based method, while SARIMAX is a regression-based method, and the Ensemble method is based on decision trees and random forest to provide predictions. The performance of the three prediction methods with a low volume of data and the use of visitor trends data as an exogenous variable was evaluated. Based on the test prediction results, the LSTM and Ensemble-based methods provided better short-term predictions at multiple times on a weekday, while the Ensemble-based method provided better predictions over the weekend. However, the use of visitor trend data did not facilitate improving the predictions of SARIMAX and the Ensemble-based method, while it improved the LSTM prediction for the weekend.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"25 7 1","pages":"541 - 554"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86563086","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}
Abstract With the continuous development of society, the role of the visual guidance system in animation design has also evolved and evolved in its long history, leading to the changes in the values of modern beauty. In the field of modern social and cultural design, the visual guidance system in animation design has unique regional nature and cultural influence. The visual language should correspond to the visual environment and easy to understand and be known by people. It combines animation conception and design technology to capture the cultural charm and beauty, values, and behavioral norms of people in different fields. This article studies and analyzes the visual orientation of graphic language in the design of animation visual guidance system, and injects the graphic language with orientation into its animation design, so that the animation design is more in line with the characteristics of the times. It can be more adapted to the emerging media and better convey the information transfer between the enterprise and the audience. To further understand the audience’s tendency toward elements of graphic expression, this article analyzes the subjective perceptions of the respondents on the importance of color selection, calligraphy fonts, graphic expression, and modeling meaning. The results of the study showed that the respondents aged 21–35 paid more attention to the choice of graphic colors, and the highest number was 69.
{"title":"The application of graphic language in animation visual guidance system under intelligent environment","authors":"Luning Zhao","doi":"10.1515/jisys-2022-0074","DOIUrl":"https://doi.org/10.1515/jisys-2022-0074","url":null,"abstract":"Abstract With the continuous development of society, the role of the visual guidance system in animation design has also evolved and evolved in its long history, leading to the changes in the values of modern beauty. In the field of modern social and cultural design, the visual guidance system in animation design has unique regional nature and cultural influence. The visual language should correspond to the visual environment and easy to understand and be known by people. It combines animation conception and design technology to capture the cultural charm and beauty, values, and behavioral norms of people in different fields. This article studies and analyzes the visual orientation of graphic language in the design of animation visual guidance system, and injects the graphic language with orientation into its animation design, so that the animation design is more in line with the characteristics of the times. It can be more adapted to the emerging media and better convey the information transfer between the enterprise and the audience. To further understand the audience’s tendency toward elements of graphic expression, this article analyzes the subjective perceptions of the respondents on the importance of color selection, calligraphy fonts, graphic expression, and modeling meaning. The results of the study showed that the respondents aged 21–35 paid more attention to the choice of graphic colors, and the highest number was 69.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"47 2 1","pages":"1037 - 1054"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89189169","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}
M. Gaata, M. T. Younis, Jamal N. Hasoon, S. Mostafa
Abstract Data hiding and watermarking are considered one of the most important topics in cyber security. This article proposes an optimized method for embedding a watermark image in a cover medium (color image). First, the color of the image is separated into three components (RGB). Consequently, the discrete wavelet transform is applied to each component to obtain four bands (high–high, high–low, low–high, and low–low), resulting in 12 bands in total. By omitting the low–low band from each component, a new square matrix is formed from the rest bands to be used for the hiding process after adding keys to it. These keys are generated using a hybrid approach, combining two chaotic functions, namely Gaussian and exponential maps. The embedding matrix is divided into square blocks with a specific length, each of which is converted using Hessenberg transform into two matrices, P and H. For each block, a certain location within the H-matrix is used for embedding a secret value; the updated blocks are assembled, and the reverse process is performed. An optimization method is applied, through the application of the firework algorithm, on the set of the initial values that generate keys. Using an optimization procedure to obtain keys requires performing lowest possible change rate in an image and maintain the quality of the image. To analyze and test the efficiency of the proposed method, mean-square error (MSE) and peak signal-to-noise ratio (PSNR) measurements are calculated. Furthermore, the robustness of the watermark is computed by applying several attacks. The experimental results show that the value of the MSE is reduced by about 0.01 while the value of the PSNR is increased by about 1.25 on average. Moreover, the proposed method achieved a high-retrieval rate in comparison with the non-optimization approach.
{"title":"Hessenberg factorization and firework algorithms for optimized data hiding in digital images","authors":"M. Gaata, M. T. Younis, Jamal N. Hasoon, S. Mostafa","doi":"10.1515/jisys-2022-0029","DOIUrl":"https://doi.org/10.1515/jisys-2022-0029","url":null,"abstract":"Abstract Data hiding and watermarking are considered one of the most important topics in cyber security. This article proposes an optimized method for embedding a watermark image in a cover medium (color image). First, the color of the image is separated into three components (RGB). Consequently, the discrete wavelet transform is applied to each component to obtain four bands (high–high, high–low, low–high, and low–low), resulting in 12 bands in total. By omitting the low–low band from each component, a new square matrix is formed from the rest bands to be used for the hiding process after adding keys to it. These keys are generated using a hybrid approach, combining two chaotic functions, namely Gaussian and exponential maps. The embedding matrix is divided into square blocks with a specific length, each of which is converted using Hessenberg transform into two matrices, P and H. For each block, a certain location within the H-matrix is used for embedding a secret value; the updated blocks are assembled, and the reverse process is performed. An optimization method is applied, through the application of the firework algorithm, on the set of the initial values that generate keys. Using an optimization procedure to obtain keys requires performing lowest possible change rate in an image and maintain the quality of the image. To analyze and test the efficiency of the proposed method, mean-square error (MSE) and peak signal-to-noise ratio (PSNR) measurements are calculated. Furthermore, the robustness of the watermark is computed by applying several attacks. The experimental results show that the value of the MSE is reduced by about 0.01 while the value of the PSNR is increased by about 1.25 on average. Moreover, the proposed method achieved a high-retrieval rate in comparison with the non-optimization approach.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"31 1","pages":"440 - 453"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87154330","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}
Y. Zhang, Arshpreet Kaur, Vishal Jagota, Rahul Neware
Abstract In recent years, the network has become more complex, and the attacker’s ability to attack is gradually increasing. How to properly understand the network security situation and improve network security has become a very important issue. In order to study the method of extracting information about the security situation of the network based on cloud computing, we recommend the technology of knowledge of the network security situation based on the data extraction technology. It converts each received cyber security event into a standard format that can be defined as multiple brochures, creating a general framework for the cyber security situation. According to the large nature of network security situation data, the Hadoop platform is used to extract aggregation rules, and perform model extraction, pattern analysis, and learning on a network security event dataset to complete network security situation rule mining, and establish a framework for assessing the state of network security. According to the results of the federal rule extraction, the level of network node security risk is obtained in combination with signal reliability, signal severity, resource impact, node protection level, and signal recovery factor. A simulation test is performed to obtain the intrusion index according to the source address of the network security alarm. Through the relevant experiments and analysis of the results, the attack characteristics obtained in this study were obtained after manually reducing the network security event in the 295 h window. The results show that after the security event is canceled, the corresponding window attack index decreases to 0, indicating that this method can effectively implement a network security situation awareness. The proposed technique allows you to accurately sense changes in network security conditions.
{"title":"Study on data mining method of network security situation perception based on cloud computing","authors":"Y. Zhang, Arshpreet Kaur, Vishal Jagota, Rahul Neware","doi":"10.1515/jisys-2021-0264","DOIUrl":"https://doi.org/10.1515/jisys-2021-0264","url":null,"abstract":"Abstract In recent years, the network has become more complex, and the attacker’s ability to attack is gradually increasing. How to properly understand the network security situation and improve network security has become a very important issue. In order to study the method of extracting information about the security situation of the network based on cloud computing, we recommend the technology of knowledge of the network security situation based on the data extraction technology. It converts each received cyber security event into a standard format that can be defined as multiple brochures, creating a general framework for the cyber security situation. According to the large nature of network security situation data, the Hadoop platform is used to extract aggregation rules, and perform model extraction, pattern analysis, and learning on a network security event dataset to complete network security situation rule mining, and establish a framework for assessing the state of network security. According to the results of the federal rule extraction, the level of network node security risk is obtained in combination with signal reliability, signal severity, resource impact, node protection level, and signal recovery factor. A simulation test is performed to obtain the intrusion index according to the source address of the network security alarm. Through the relevant experiments and analysis of the results, the attack characteristics obtained in this study were obtained after manually reducing the network security event in the 295 h window. The results show that after the security event is canceled, the corresponding window attack index decreases to 0, indicating that this method can effectively implement a network security situation awareness. The proposed technique allows you to accurately sense changes in network security conditions.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"141 1","pages":"1074 - 1084"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84789477","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}
Abstract An improved variant of the Jaya optimization algorithm, called Jaya2, is proposed to enhance the performance of the original Jaya sacrificing its algorithmic design. The proposed approach arranges the solutions in a ring topology to reduce the likelihood of premature convergence. In addition, the population size reduction is used to automatically adjust the population size during the optimization process. Moreover, the translation dependency problem of the original Jaya is discussed, and an alternative solution update operation is proposed. To test Jaya2, we compare it with nine different optimization methods on the CEC 2020 benchmark functions and the CEC 2011 real-world optimization problems. The results show that Jaya2 is highly competitive on the tested problems where it generally outperforms most approaches. Having an easy-to-implement approach with little parameter tuning is highly desirable since researchers from different disciplines with basic programming skills can use it to solve their optimization problems.
{"title":"An improved Jaya optimization algorithm with ring topology and population size reduction","authors":"Mahamed G. H. Omran, Giovanni Iacca","doi":"10.1515/jisys-2022-0200","DOIUrl":"https://doi.org/10.1515/jisys-2022-0200","url":null,"abstract":"Abstract An improved variant of the Jaya optimization algorithm, called Jaya2, is proposed to enhance the performance of the original Jaya sacrificing its algorithmic design. The proposed approach arranges the solutions in a ring topology to reduce the likelihood of premature convergence. In addition, the population size reduction is used to automatically adjust the population size during the optimization process. Moreover, the translation dependency problem of the original Jaya is discussed, and an alternative solution update operation is proposed. To test Jaya2, we compare it with nine different optimization methods on the CEC 2020 benchmark functions and the CEC 2011 real-world optimization problems. The results show that Jaya2 is highly competitive on the tested problems where it generally outperforms most approaches. Having an easy-to-implement approach with little parameter tuning is highly desirable since researchers from different disciplines with basic programming skills can use it to solve their optimization problems.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"9 1","pages":"1178 - 1210"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81691131","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}
Abstract With the development of communication and computer technology, the application of big data technology has become increasingly widespread. Reasonable, effective, and fast retrieval methods for querying information from massive data have become an important content of current research. This article provides an image retrieval method based on the weighted nearest neighbor label prediction for the problem of automatic image annotation and keyword image retrieval. In order to improve the performance of the test method, scientific experimental verification was implemented. The nearest neighbor weights are determined by maximizing the training image annotation, and experiments are carried out from multiple angles based on the Mahalanobis metric learning integration model. The experimental results show that the proposed tag correlation prediction propagation model has obvious improvements in accuracy, recall rate, break-even point, and overall average accuracy performance compared with other widely used algorithm models.
{"title":"Image retrieval based on weighted nearest neighbor tag prediction","authors":"Qizhuo Yao, Dayang Jiang, Xiancheng Ding","doi":"10.1515/jisys-2022-0045","DOIUrl":"https://doi.org/10.1515/jisys-2022-0045","url":null,"abstract":"Abstract With the development of communication and computer technology, the application of big data technology has become increasingly widespread. Reasonable, effective, and fast retrieval methods for querying information from massive data have become an important content of current research. This article provides an image retrieval method based on the weighted nearest neighbor label prediction for the problem of automatic image annotation and keyword image retrieval. In order to improve the performance of the test method, scientific experimental verification was implemented. The nearest neighbor weights are determined by maximizing the training image annotation, and experiments are carried out from multiple angles based on the Mahalanobis metric learning integration model. The experimental results show that the proposed tag correlation prediction propagation model has obvious improvements in accuracy, recall rate, break-even point, and overall average accuracy performance compared with other widely used algorithm models.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"129 1","pages":"589 - 600"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80343115","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}
Abstract With the expansion of people’s needs, the translation performance of traditional models is increasingly unable to meet current demands. This article mainly studied the Transformer model. First, the structure and principle of the Transformer model were briefly introduced. Then, the model was improved by a generative adversarial network (GAN) to improve the translation effect of the model. Finally, experiments were carried out on the linguistic data consortium (LDC) dataset. It was found that the average Bilingual Evaluation Understudy (BLEU) value of the improved Transformer model improved by 0.49, and the average perplexity value reduced by 10.06 compared with the Transformer model, but the computation speed was not greatly affected. The translation results of the two example sentences showed that the translation of the improved Transformer model was closer to the results of human translation. The experimental results verify that the improved Transformer model can improve the translation quality and be further promoted and applied in practice to further improve the English translation and meet application needs in real life.
{"title":"Research on an English translation method based on an improved transformer model","authors":"Hongxia Li, Xin Tuo","doi":"10.1515/jisys-2022-0038","DOIUrl":"https://doi.org/10.1515/jisys-2022-0038","url":null,"abstract":"Abstract With the expansion of people’s needs, the translation performance of traditional models is increasingly unable to meet current demands. This article mainly studied the Transformer model. First, the structure and principle of the Transformer model were briefly introduced. Then, the model was improved by a generative adversarial network (GAN) to improve the translation effect of the model. Finally, experiments were carried out on the linguistic data consortium (LDC) dataset. It was found that the average Bilingual Evaluation Understudy (BLEU) value of the improved Transformer model improved by 0.49, and the average perplexity value reduced by 10.06 compared with the Transformer model, but the computation speed was not greatly affected. The translation results of the two example sentences showed that the translation of the improved Transformer model was closer to the results of human translation. The experimental results verify that the improved Transformer model can improve the translation quality and be further promoted and applied in practice to further improve the English translation and meet application needs in real life.","PeriodicalId":46139,"journal":{"name":"Journal of Intelligent Systems","volume":"2 1","pages":"532 - 540"},"PeriodicalIF":3.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90127226","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：