A prototype portable breast microwave sensing (BMS) system for early breast cancer detection has been developed in our lab. In this paper, we provide preliminary results for the response of the antenna sensor array to a point scatterer, using both simulation and experiment. The portable system uses a horn antenna to transmit frequencies from 1.5 GHz to 6 GHz and a sensor array using thirteen patch antennas. The optimal separation of each antenna in the receiver array was calculated to be 4 mm based on an envelope correlation coefficient of 0.37. The BMS system, including the horn antenna, was designed in CST Microwave Studio to mimic the experimental setup. This study compares the E-field characteristics and DC voltages for each sensor, using simulation and experimental results, for both a free air system and with an Aluminum rod placed at different positions in the scanning plane. A 13×10 array of geometric correction constants was calculated from the simulated E-field. The range of difference between the simulated and experimental results was -4% to 3% for open space conditions and ±20% when an Aluminum rod was placed at different positions in the scanning plane. The preliminary results are promising and provide some insight as to where improvements must be made to enhance the detection ability of the portable system.
{"title":"Sensitivity Analysis of a Portable Microwave Breast Cancer Detection System","authors":"Muhammad Masud Rana, Debarati Nath, S. Pistorius","doi":"10.11159/icbes20.112","DOIUrl":"https://doi.org/10.11159/icbes20.112","url":null,"abstract":"A prototype portable breast microwave sensing (BMS) system for early breast cancer detection has been developed in our lab. In this paper, we provide preliminary results for the response of the antenna sensor array to a point scatterer, using both simulation and experiment. The portable system uses a horn antenna to transmit frequencies from 1.5 GHz to 6 GHz and a sensor array using thirteen patch antennas. The optimal separation of each antenna in the receiver array was calculated to be 4 mm based on an envelope correlation coefficient of 0.37. The BMS system, including the horn antenna, was designed in CST Microwave Studio to mimic the experimental setup. This study compares the E-field characteristics and DC voltages for each sensor, using simulation and experimental results, for both a free air system and with an Aluminum rod placed at different positions in the scanning plane. A 13×10 array of geometric correction constants was calculated from the simulated E-field. The range of difference between the simulated and experimental results was -4% to 3% for open space conditions and ±20% when an Aluminum rod was placed at different positions in the scanning plane. The preliminary results are promising and provide some insight as to where improvements must be made to enhance the detection ability of the portable system.","PeriodicalId":377357,"journal":{"name":"Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129336233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we investigate predictive models to detect the spread of hate speech on Twitter based on diffusion patterns. We experiment with a dataset of 10,000 tweets manually labelled as hate speech or not and show that classification based solely on the sharing graph yields strong F1 scores for our task and high hate speech detection precision. We also highlight the vulnerability of existing textual hate speech detection methods to adversarial attacks and demonstrate that while our methods do not outperform stateof-the-art text models, graph-based models provide robust detection mechanisms and are able to detect instances of hate speech that missed by text classifiers. We find that graph convolutional networks produce the strongest hate speech F1 score of 0.58 and that kernel methods offer strong predictive potential. Finally, we also consider the effects of automated bots in the diffusion of hate speech content and conclude that their sharing behavior plays an insignificant role in our experiments.
{"title":"Classification Methods for Hate Speech Diffusion: Detecting the\u0000Spread of Hate Speech on Twitter","authors":"Matthew Beatty","doi":"10.11159/cist20.105","DOIUrl":"https://doi.org/10.11159/cist20.105","url":null,"abstract":"In this paper, we investigate predictive models to detect the spread of hate speech on Twitter based on diffusion patterns. We experiment with a dataset of 10,000 tweets manually labelled as hate speech or not and show that classification based solely on the sharing graph yields strong F1 scores for our task and high hate speech detection precision. We also highlight the vulnerability of existing textual hate speech detection methods to adversarial attacks and demonstrate that while our methods do not outperform stateof-the-art text models, graph-based models provide robust detection mechanisms and are able to detect instances of hate speech that missed by text classifiers. We find that graph convolutional networks produce the strongest hate speech F1 score of 0.58 and that kernel methods offer strong predictive potential. Finally, we also consider the effects of automated bots in the diffusion of hate speech content and conclude that their sharing behavior plays an insignificant role in our experiments.","PeriodicalId":377357,"journal":{"name":"Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116106173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
- The latest advances in Semantic Web technologies suggest an accelerating emergence of new exciting Artificial Intelligence applications that are expected to dramatically extend and improve current web services. Yet, these new technologies are outside the scope of undergraduate computer science curriculum. This paper presents our experience with introducing a hands-on project intended to teach Linked Data and Semantic Web as part of an undergraduate Artificial Intelligence course. The project is intended to achieve the following: 1.) Demonstrate the evolution of Knowledge Engineering into Ontological Engineering; 2.) Introduce students to Semantic Web technologies and tools such as ontology editor Protégé, Web Ontology Language (OWL), Semantic Web Rule Language (SWRL), and query language SPARQL; 3.) Extend the topic on reasoning into Description Logics and demonstrate the advantages of their inferencing capabilities; 4.) Use OWL and SWRL to compare descriptive and rule-based reasoning frameworks and show how their integration can improve the efficiency and the semantic adequacy of applications; 5.) Illustrate the Linked Data principles in a practical setting. Limited assessment of the pedagogical value of this project based on student learning outcomes suggests that it enhances students’ understanding of the core AI topics, boosts their engagement and interest in the course, but more importantly introduces them to the newest advances in web application development.
{"title":"A Hands-on Project for Teaching Semantic Web Technologies in an Undergraduate AI Course","authors":"N. Zlatareva","doi":"10.11159/cist20.101","DOIUrl":"https://doi.org/10.11159/cist20.101","url":null,"abstract":"- The latest advances in Semantic Web technologies suggest an accelerating emergence of new exciting Artificial Intelligence applications that are expected to dramatically extend and improve current web services. Yet, these new technologies are outside the scope of undergraduate computer science curriculum. This paper presents our experience with introducing a hands-on project intended to teach Linked Data and Semantic Web as part of an undergraduate Artificial Intelligence course. The project is intended to achieve the following: 1.) Demonstrate the evolution of Knowledge Engineering into Ontological Engineering; 2.) Introduce students to Semantic Web technologies and tools such as ontology editor Protégé, Web Ontology Language (OWL), Semantic Web Rule Language (SWRL), and query language SPARQL; 3.) Extend the topic on reasoning into Description Logics and demonstrate the advantages of their inferencing capabilities; 4.) Use OWL and SWRL to compare descriptive and rule-based reasoning frameworks and show how their integration can improve the efficiency and the semantic adequacy of applications; 5.) Illustrate the Linked Data principles in a practical setting. Limited assessment of the pedagogical value of this project based on student learning outcomes suggests that it enhances students’ understanding of the core AI topics, boosts their engagement and interest in the course, but more importantly introduces them to the newest advances in web application development.","PeriodicalId":377357,"journal":{"name":"Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121059588","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}
Antonio Leonti, J. Fonseca, I. Valova, R. Beemer, Devin Cannistraro, C. Pilskaln, Dylan DeFlorio, Grayson Kelly
There is much to be gained from analysing and studying calcareous sediment, with applications ranging from the study of climate change, rock dating, and even building offshore oil rigs and wind farms. One way of performing this analysis is to obtain a µCT scan of the sediment, allowing scientists and engineers to automate much of their analysis using software. Many existing and prospective analysis techniques require handling individual grains. Thus, fast and effective segmentation is an essential first step for any such analysis. Segmentation is non-trivial; these scans hold a lot of information, exhibit ambiguous boundaries between objects, and many objects are hollow, making it even more difficult to apply traditional watershed segmentation. Addressing these issues, in this paper we propose an optimized 3D segmentation (O3DS) algorithm based on watersheds. We make use of branch recursion, partition the image by height prior to segmentation, artificially reducing the size of the largest connected objects. These and additional changes are extremely effective in optimizing performance; O3DS reduces the time to segment a 659x925x932 scan of sediment by 95.4% and produces better or comparable results when compared to similar implementation by our co-author.
{"title":"Optimized 3D Segmentation Algorithm for Shelly Sand Images","authors":"Antonio Leonti, J. Fonseca, I. Valova, R. Beemer, Devin Cannistraro, C. Pilskaln, Dylan DeFlorio, Grayson Kelly","doi":"10.11159/cist20.107","DOIUrl":"https://doi.org/10.11159/cist20.107","url":null,"abstract":"There is much to be gained from analysing and studying calcareous sediment, with applications ranging from the study of climate change, rock dating, and even building offshore oil rigs and wind farms. One way of performing this analysis is to obtain a µCT scan of the sediment, allowing scientists and engineers to automate much of their analysis using software. Many existing and prospective analysis techniques require handling individual grains. Thus, fast and effective segmentation is an essential first step for any such analysis. Segmentation is non-trivial; these scans hold a lot of information, exhibit ambiguous boundaries between objects, and many objects are hollow, making it even more difficult to apply traditional watershed segmentation. Addressing these issues, in this paper we propose an optimized 3D segmentation (O3DS) algorithm based on watersheds. We make use of branch recursion, partition the image by height prior to segmentation, artificially reducing the size of the largest connected objects. These and additional changes are extremely effective in optimizing performance; O3DS reduces the time to segment a 659x925x932 scan of sediment by 95.4% and produces better or comparable results when compared to similar implementation by our co-author.","PeriodicalId":377357,"journal":{"name":"Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133148648","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}
Spatial information in broadband array signals is embedded in the relative delay with which sources illuminate different sensors. Therefore, second order statistics, on which cost functions such as the mean square rest, must include such delays. Typically, a space-time covariance matrix therefore arises, which can be represented as a Laurent polynomial matrix. The optimisation of a cost function then requires extending the utility of the eigenvalue decomposition from narrowband covariance matrices to the broadband case of operating in a space-time covariance matrix. This overview paper summarises efforts in performing such factorisations, and demonstrated via the exemplar application of a broadband beamformer how thus well-known narrowband solutions can be extended to the broadband case using polynomial matrices and their factorisations.
{"title":"Mathematical Tools for Processing Broadband Multi-Sensor Signals","authors":"Stephan Weiss","doi":"10.11159/mhci20.104","DOIUrl":"https://doi.org/10.11159/mhci20.104","url":null,"abstract":"Spatial information in broadband array signals is embedded in the relative delay with which sources illuminate different sensors. Therefore, second order statistics, on which cost functions such as the mean square rest, must include such delays. Typically, a space-time covariance matrix therefore arises, which can be represented as a Laurent polynomial matrix. The optimisation of a cost function then requires extending the utility of the eigenvalue decomposition from narrowband covariance matrices to the broadband case of operating in a space-time covariance matrix. This overview paper summarises efforts in performing such factorisations, and demonstrated via the exemplar application of a broadband beamformer how thus well-known narrowband solutions can be extended to the broadband case using polynomial matrices and their factorisations.","PeriodicalId":377357,"journal":{"name":"Proceedings of the 6th World Congress on Electrical Engineering and Computer Systems and Science","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115390906","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：