Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083476
Igor Stupavský, V. Vranić
We can learn most of the information we need about the world by reading unstructured text on the internet. The role of artificial intelligence is to understand this text and correctly understand it, categorize it, and make the correct system response. Texts in languages like English, which is spoken by most of the world, have their own trained tools. These models are fairly good at determining things like part of speech, sentence structure, and the like. If we want to teach artificial intelligence to recognize even minority languages, we run into a problem. In our case, we chose Slovak as the minority language. In the previous article, we reported on the misidentification of word types by the available tools. In this paper, we have significantly improved the identification of words in the second place. We plan to continue the development of Slovak-optimized software.
{"title":"Analysing the controversial social media community","authors":"Igor Stupavský, V. Vranić","doi":"10.1109/Informatics57926.2022.10083476","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083476","url":null,"abstract":"We can learn most of the information we need about the world by reading unstructured text on the internet. The role of artificial intelligence is to understand this text and correctly understand it, categorize it, and make the correct system response. Texts in languages like English, which is spoken by most of the world, have their own trained tools. These models are fairly good at determining things like part of speech, sentence structure, and the like. If we want to teach artificial intelligence to recognize even minority languages, we run into a problem. In our case, we chose Slovak as the minority language. In the previous article, we reported on the misidentification of word types by the available tools. In this paper, we have significantly improved the identification of words in the second place. We plan to continue the development of Slovak-optimized software.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"249 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116458295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083486
Ophir Almagor, Ofer Avin, R. Rosipal, O. Shriki
A major problem in brain-computer interfaces (BCIs) relates to the non-stationarity of brain signals. Consequently, the performance of a classification algorithm trained for an individual subject on a certain day deteriorates during the following days. The traditional approach is to recalibrate the algorithm every session, limiting the wide use of BCIs. Here, we use an autoencoder convolutional neural network to identify a low dimensional representation of the EEG signals from the first day (or days) and show that this allows for stable decoding performance on the following days without resorting to recalibration. Furthermore, we demonstrate that the residual signals, namely the difference between the original and reconstructed EEG, can be used to accurately discriminate among different recording sessions. In line with that, the reconstructed EEG cannot be used to discriminate among recording sessions. This implies that the reconstructed EEG reflects an invariant representation of the subject's intent, whereas the residual signals reflect a non-stationary component, which differs from one session to another. The findings are demonstrated through two different datasets.
{"title":"Using Autoencoders to Denoise Cross-Session Non-Stationarity in EEG-Based Motor-Imagery Brain-Computer Interfaces","authors":"Ophir Almagor, Ofer Avin, R. Rosipal, O. Shriki","doi":"10.1109/Informatics57926.2022.10083486","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083486","url":null,"abstract":"A major problem in brain-computer interfaces (BCIs) relates to the non-stationarity of brain signals. Consequently, the performance of a classification algorithm trained for an individual subject on a certain day deteriorates during the following days. The traditional approach is to recalibrate the algorithm every session, limiting the wide use of BCIs. Here, we use an autoencoder convolutional neural network to identify a low dimensional representation of the EEG signals from the first day (or days) and show that this allows for stable decoding performance on the following days without resorting to recalibration. Furthermore, we demonstrate that the residual signals, namely the difference between the original and reconstructed EEG, can be used to accurately discriminate among different recording sessions. In line with that, the reconstructed EEG cannot be used to discriminate among recording sessions. This implies that the reconstructed EEG reflects an invariant representation of the subject's intent, whereas the residual signals reflect a non-stationary component, which differs from one session to another. The findings are demonstrated through two different datasets.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129050911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083447
William Steingartner, Darko Možnik, D. Galinec
How to distinguish truth from lies in the environment in which we are exposed to vast amounts of information from different sources and through different platforms - is a burning challenge today. Although lies and manipulations in public information space are not a novelty, the quantity and speed of spreading misinformation, especially through social networks and mobile communication applications present an unprecedented challenge. The level of pollution of the information environment and the complexity of the actors, the technique and the motivation behind its contributors have never been so present, within hybrid threats today. Novelty is, therefore, the speed of spreading misinformation through network channels, platforms and applications. In this paper, a new non-hierarchical conceptual model of hybrid threats aiming to improve awareness and systems' resilience is developed, which includes disinformation campaigns and provides an industrial example of a system for combating fake news.
{"title":"Disinformation Campaigns and Resilience in Hybrid Threats Conceptual Model","authors":"William Steingartner, Darko Možnik, D. Galinec","doi":"10.1109/Informatics57926.2022.10083447","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083447","url":null,"abstract":"How to distinguish truth from lies in the environment in which we are exposed to vast amounts of information from different sources and through different platforms - is a burning challenge today. Although lies and manipulations in public information space are not a novelty, the quantity and speed of spreading misinformation, especially through social networks and mobile communication applications present an unprecedented challenge. The level of pollution of the information environment and the complexity of the actors, the technique and the motivation behind its contributors have never been so present, within hybrid threats today. Novelty is, therefore, the speed of spreading misinformation through network channels, platforms and applications. In this paper, a new non-hierarchical conceptual model of hybrid threats aiming to improve awareness and systems' resilience is developed, which includes disinformation campaigns and provides an industrial example of a system for combating fake news.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125703580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083477
Jacek Wachowicz, Andrzej Karp, Pawel Kossecki
Modern recordings are based on computer recording and computer editing. Therefore, in fact, a real, complete tool for a modern musician is, in most cases, not only an instrument but an instrument supported by powerful software tools, in which control values may be entered using virtual controllers displayed on the screen or with physical manipulators like switches, sliders or knobs, similar to those used in classical devices. The aim of this research was to find answers to the questions: “Are physical manipulators, in your opinion, better than virtual ones? Why? Does having more manipulators always mean better? Is there any optimal number of them? If so, what?” and a second one, “What methods do you use to limit the number of manipulators and shorten the program handlers?” The study was carried out using the in-depth interview method in a hybrid way, as a part of a broader interview with music software developers about HCI factors and the perceived differences between creating emulators of existing music devices and creating completely new music software, including plugins. The selection of respondents included, on the one hand, one of the most recognizable Polish plug-in companies on the world market - PSP audioware, where interviews were conducted with all owners and members of the group, which, due to their specialization, allowed for a multidisciplinary approach; and on the other hand, internationally known creators working for the D16group, Newfangled Audio and Eventide.
{"title":"On the Friendliness of Physical vs. Virtual Manipulators in Music Software - from a Perspective of Software Developers","authors":"Jacek Wachowicz, Andrzej Karp, Pawel Kossecki","doi":"10.1109/Informatics57926.2022.10083477","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083477","url":null,"abstract":"Modern recordings are based on computer recording and computer editing. Therefore, in fact, a real, complete tool for a modern musician is, in most cases, not only an instrument but an instrument supported by powerful software tools, in which control values may be entered using virtual controllers displayed on the screen or with physical manipulators like switches, sliders or knobs, similar to those used in classical devices. The aim of this research was to find answers to the questions: “Are physical manipulators, in your opinion, better than virtual ones? Why? Does having more manipulators always mean better? Is there any optimal number of them? If so, what?” and a second one, “What methods do you use to limit the number of manipulators and shorten the program handlers?” The study was carried out using the in-depth interview method in a hybrid way, as a part of a broader interview with music software developers about HCI factors and the perceived differences between creating emulators of existing music devices and creating completely new music software, including plugins. The selection of respondents included, on the one hand, one of the most recognizable Polish plug-in companies on the world market - PSP audioware, where interviews were conducted with all owners and members of the group, which, due to their specialization, allowed for a multidisciplinary approach; and on the other hand, internationally known creators working for the D16group, Newfangled Audio and Eventide.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122781513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083497
A. Dudáš, Daniel Demian, J. Škrinárová
Relational database model, as one of the most commonly used database models, is defined by the structure used for the storing of data, methods of data manipulation and so called data integrity. In this paper, we are focusing on the methods and processes of manipulating data stored in the database with the use of relational algebra, relational calculus and Structured Query Language. The paper presents the design, implementation and experimental evaluation of a software tool that can be used for automatic translation between the mentioned relational languages themselves and restricted natural language with the integrated automatic optimization of queries in relational algebra.
{"title":"On Relational Language Translation","authors":"A. Dudáš, Daniel Demian, J. Škrinárová","doi":"10.1109/Informatics57926.2022.10083497","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083497","url":null,"abstract":"Relational database model, as one of the most commonly used database models, is defined by the structure used for the storing of data, methods of data manipulation and so called data integrity. In this paper, we are focusing on the methods and processes of manipulating data stored in the database with the use of relational algebra, relational calculus and Structured Query Language. The paper presents the design, implementation and experimental evaluation of a software tool that can be used for automatic translation between the mentioned relational languages themselves and restricted natural language with the integrated automatic optimization of queries in relational algebra.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124773241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083491
Stan Zakrzewski, Bartlomiej Stasiak, A. Wojciechowski
This paper presents an EEG-based Brain-Computer Interface (BCI) designed for classification of motor imagery tasks. Apart from the imagined right-hand / left-hand movements, a third class - the resting (idle) state - is also included. The classification is based on Linear Discriminant Analysis (LDA) and Common Spatial Patterns (CSP) filtering of the band-limited EEG signal acquired from 10 electrodes placed over the motor cortex area. The system, planned for integration with a virtual reality (VR) environment and designed for future neurorehabilitation applications is tested on an experimental database comprising 52 subjects. The observed variability be-tween individual participants and selected subgroups is further analysed with statistical tools, revealing significant differences with respect to gender, age and individual motor imagery task classes.
{"title":"EEG-based left-hand/right-hand/rest motor imagery task classification","authors":"Stan Zakrzewski, Bartlomiej Stasiak, A. Wojciechowski","doi":"10.1109/Informatics57926.2022.10083491","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083491","url":null,"abstract":"This paper presents an EEG-based Brain-Computer Interface (BCI) designed for classification of motor imagery tasks. Apart from the imagined right-hand / left-hand movements, a third class - the resting (idle) state - is also included. The classification is based on Linear Discriminant Analysis (LDA) and Common Spatial Patterns (CSP) filtering of the band-limited EEG signal acquired from 10 electrodes placed over the motor cortex area. The system, planned for integration with a virtual reality (VR) environment and designed for future neurorehabilitation applications is tested on an experimental database comprising 52 subjects. The observed variability be-tween individual participants and selected subgroups is further analysed with statistical tools, revealing significant differences with respect to gender, age and individual motor imagery task classes.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125838499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083435
Petr Šaloun, M. Kolářová, David Andresic, M. Malčík, Milan Klement
In this paper we summarize results of our internet survey which took place in Spring 2021 having over 3400 respondents and focused on adolescent sexting presence on the Instagram social network. We were interested in the frequency of this phenomena, awareness and general experience of adolescents with sexting and in general in Instagram social network security regarding adolescents. Results in terms of the danger of this phenomena are clear and demonstrable: a significant amount of the teenagers came across with it. The linkage and necessity of awareness in terms of secure behaviour on the internet and relevance with computational thinking is obvious but out of scope of this work.
{"title":"Teen-sexting on Social Network “Instagram” and it's Computational Thinking Consequences","authors":"Petr Šaloun, M. Kolářová, David Andresic, M. Malčík, Milan Klement","doi":"10.1109/Informatics57926.2022.10083435","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083435","url":null,"abstract":"In this paper we summarize results of our internet survey which took place in Spring 2021 having over 3400 respondents and focused on adolescent sexting presence on the Instagram social network. We were interested in the frequency of this phenomena, awareness and general experience of adolescents with sexting and in general in Instagram social network security regarding adolescents. Results in terms of the danger of this phenomena are clear and demonstrable: a significant amount of the teenagers came across with it. The linkage and necessity of awareness in terms of secure behaviour on the internet and relevance with computational thinking is obvious but out of scope of this work.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129864591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083397
T. Vyčítal
In a simulation model of a railway system it is important, besides other crucial algorithms, to have correct behaviour of train overtaking in stochastic conditions. This problem is being addressed in many simulation tools focused on railway traffic, however these are not very human-realistic. The goal of this paper is to create a more human-realistic overtaking decision support system for the use in railway traffic simulations. A fuzzy system has been chosen for this task as fuzzy systems are well-suited for human-like decision making. The fuzzy system designed takes into account timetables, train positions, delays and buffer times as inputs and provides an instruction to overtake or not overtake.
{"title":"Fuzzy decision support system for human-realistic overtaking in railway traffic simulations","authors":"T. Vyčítal","doi":"10.1109/Informatics57926.2022.10083397","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083397","url":null,"abstract":"In a simulation model of a railway system it is important, besides other crucial algorithms, to have correct behaviour of train overtaking in stochastic conditions. This problem is being addressed in many simulation tools focused on railway traffic, however these are not very human-realistic. The goal of this paper is to create a more human-realistic overtaking decision support system for the use in railway traffic simulations. A fuzzy system has been chosen for this task as fuzzy systems are well-suited for human-like decision making. The fuzzy system designed takes into account timetables, train positions, delays and buffer times as inputs and provides an instruction to overtake or not overtake.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121131950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083406
Zuzana Káčereková
Radial basis functions (RBFs) are commonly used in solving partial differential equations, interpolation and ap-proximation of scalar and vector data, image reconstruction, geographic information systems, and machine learning. In this contribution, the properties of interpolations using standard and normalized RBFs are compared, and the impact of an added polynomial term is analyzed using various samplings of input functions. Experimental results across a range of tested kernel functions and shape parameters are provided.
{"title":"Comparing Interpolations Using Standard and Normalized Radial Basis Functions","authors":"Zuzana Káčereková","doi":"10.1109/Informatics57926.2022.10083406","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083406","url":null,"abstract":"Radial basis functions (RBFs) are commonly used in solving partial differential equations, interpolation and ap-proximation of scalar and vector data, image reconstruction, geographic information systems, and machine learning. In this contribution, the properties of interpolations using standard and normalized RBFs are compared, and the impact of an added polynomial term is analyzed using various samplings of input functions. Experimental results across a range of tested kernel functions and shape parameters are provided.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126566315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-23DOI: 10.1109/Informatics57926.2022.10083422
M. Ahmadzai, Giang Nguyen
Internet and digital devices have caused a flood of information in the fourth industrial revolution era. Health-care, government, banks, military, and the banking sector, have embraced machine learning as a convenient way to recognize patterns in data. For a conventional machine learning model, the information from the data owner must be uploaded in a centralized location for training, which causes data owners to worry about the lack of their private information assurance. On the other hand, a massive amount of computing power is available to train intelligent models. In this way, federated learning becomes more and more popular. It preserves privacy of data owners and allows models to be trained over distributed clients. In this work, the federated learning approach is implemented using a neural network to model several Afghan-People opinion investigations such as the question about country's overall situation or people's security and safety in a distributed computing environment with measurable results. Further, the performance of federated learning using the same dataset is compared with centralized machine learning.
{"title":"An opinion mining with federated learning on the Afghan-People survey dataset","authors":"M. Ahmadzai, Giang Nguyen","doi":"10.1109/Informatics57926.2022.10083422","DOIUrl":"https://doi.org/10.1109/Informatics57926.2022.10083422","url":null,"abstract":"Internet and digital devices have caused a flood of information in the fourth industrial revolution era. Health-care, government, banks, military, and the banking sector, have embraced machine learning as a convenient way to recognize patterns in data. For a conventional machine learning model, the information from the data owner must be uploaded in a centralized location for training, which causes data owners to worry about the lack of their private information assurance. On the other hand, a massive amount of computing power is available to train intelligent models. In this way, federated learning becomes more and more popular. It preserves privacy of data owners and allows models to be trained over distributed clients. In this work, the federated learning approach is implemented using a neural network to model several Afghan-People opinion investigations such as the question about country's overall situation or people's security and safety in a distributed computing environment with measurable results. Further, the performance of federated learning using the same dataset is compared with centralized machine learning.","PeriodicalId":101488,"journal":{"name":"2022 IEEE 16th International Scientific Conference on Informatics (Informatics)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124070585","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：