Nassim Abbas, Y. Chibani, Zineb Belhadi, Mehdia Hedir
This paper presents a new combination scheme for reducing the number of focal elements to manipulate in order to reduce the complexity of the combination process in the multi-class framework. The basic idea consists in using of p sources of information involved in the global scheme providing p kinds of complementary information to feed each set of p one class support vector machine classifiers independently of each other, which are designed for detecting the outliers of the same target class, then, the outputs issued from this set of classifiers are combined through the plausible and paradoxical reasoning theory for each target class. The main objective of this approach is to render calibrated outputs even when less complementary responses are encountered. An inspired version of Appriou's model for estimating the generalized basic belief assignments is presented in this paper. The proposed methodology allows decomposing a n-class problem into a series of n-combination, while providing n-calibrated outputs into the multi-class framework. The effectiveness of the proposed combination scheme with proportional conflict redistribution algorithm is validated on digit recognition application and is compared with existing statistical, learning, and evidence theory based combination algorithms.
{"title":"A DSmT based combination scheme for multi-class classification","authors":"Nassim Abbas, Y. Chibani, Zineb Belhadi, Mehdia Hedir","doi":"10.5281/ZENODO.22915","DOIUrl":"https://doi.org/10.5281/ZENODO.22915","url":null,"abstract":"This paper presents a new combination scheme for reducing the number of focal elements to manipulate in order to reduce the complexity of the combination process in the multi-class framework. The basic idea consists in using of p sources of information involved in the global scheme providing p kinds of complementary information to feed each set of p one class support vector machine classifiers independently of each other, which are designed for detecting the outliers of the same target class, then, the outputs issued from this set of classifiers are combined through the plausible and paradoxical reasoning theory for each target class. The main objective of this approach is to render calibrated outputs even when less complementary responses are encountered. An inspired version of Appriou's model for estimating the generalized basic belief assignments is presented in this paper. The proposed methodology allows decomposing a n-class problem into a series of n-combination, while providing n-calibrated outputs into the multi-class framework. The effectiveness of the proposed combination scheme with proportional conflict redistribution algorithm is validated on digit recognition application and is compared with existing statistical, learning, and evidence theory based combination algorithms.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128238038","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}
Erik Blasch, Kathryn B. Laskey, A. Jousselme, V. Dragos, P. Costa, J. Dezert
For many operational information fusion systems, both reliability and credibility are evaluation criteria for collected information. The Uncertainty Representation and Reasoning Evaluation Framework (URREF) is a comprehensive ontology that represents measures of uncertainty. URREF supports standards such as the NATO Standardization Agreement (STANAG) 2511, which incorporates categories of reliability and credibility. Reliability has traditionally been assessed for physical machines to support failure analysis. Source reliability of a human can also be assessed. Credibility is associated with a machine process or human assessment of collected evidence for information content. Other related constructs for URREF are data relevance and completeness. In this paper, we seek to develop a mathematical relation of weight of evidence using credibility and reliability as criteria for characterizing uncertainty in information fusion systems.
{"title":"URREF reliability versus credibility in information fusion (STANAG 2511)","authors":"Erik Blasch, Kathryn B. Laskey, A. Jousselme, V. Dragos, P. Costa, J. Dezert","doi":"10.5281/ZENODO.22660","DOIUrl":"https://doi.org/10.5281/ZENODO.22660","url":null,"abstract":"For many operational information fusion systems, both reliability and credibility are evaluation criteria for collected information. The Uncertainty Representation and Reasoning Evaluation Framework (URREF) is a comprehensive ontology that represents measures of uncertainty. URREF supports standards such as the NATO Standardization Agreement (STANAG) 2511, which incorporates categories of reliability and credibility. Reliability has traditionally been assessed for physical machines to support failure analysis. Source reliability of a human can also be assessed. Credibility is associated with a machine process or human assessment of collected evidence for information content. Other related constructs for URREF are data relevance and completeness. In this paper, we seek to develop a mathematical relation of weight of evidence using credibility and reliability as criteria for characterizing uncertainty in information fusion systems.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130539828","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 : 2013-07-09DOI: 10.14201/ADCAIJ2014261527
A. Keyhanipour, B. Moshiri
Nowadays, Web spam pages are a critical challenge for Web retrieval systems which have drastic influence on the performance of such systems. Although these systems try to combat the impact of spam pages on their final results list, spammers increasingly use more sophisticated techniques to increase the number of views for their intended pages in order to have more commercial success. This paper employs the recently proposed Layered Multi-population Genetic Programming model for Web spam detection task as well application of correlation coefficient analysis for feature space reduction. Based on our tentative results, the designed classifier, which is based on a combination of easy to compute features, has a very reasonable performance in comparison with similar methods.
{"title":"Designing a web spam classifier based on feature fusion in the Layered Multi-population Genetic Programming framework","authors":"A. Keyhanipour, B. Moshiri","doi":"10.14201/ADCAIJ2014261527","DOIUrl":"https://doi.org/10.14201/ADCAIJ2014261527","url":null,"abstract":"Nowadays, Web spam pages are a critical challenge for Web retrieval systems which have drastic influence on the performance of such systems. Although these systems try to combat the impact of spam pages on their final results list, spammers increasingly use more sophisticated techniques to increase the number of views for their intended pages in order to have more commercial success. This paper employs the recently proposed Layered Multi-population Genetic Programming model for Web spam detection task as well application of correlation coefficient analysis for feature space reduction. Based on our tentative results, the designed classifier, which is based on a combination of easy to compute features, has a very reasonable performance in comparison with similar methods.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121721575","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}
This paper presents new absolute and relative conditioning rules as possible solution of multi-level conditioning in threat assessment problem. An example of application of these rules with respect to target observation threat model has been provided. The paper also presents useful directions in order to manage the implemented multiple rules of conditioning in the real system.
{"title":"Application of new absolute and relative conditioning rules in threat assessment","authors":"K. Krenc, F. Smarandache","doi":"10.5281/ZENODO.30125","DOIUrl":"https://doi.org/10.5281/ZENODO.30125","url":null,"abstract":"This paper presents new absolute and relative conditioning rules as possible solution of multi-level conditioning in threat assessment problem. An example of application of these rules with respect to target observation threat model has been provided. The paper also presents useful directions in order to manage the implemented multiple rules of conditioning in the real system.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125451686","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 : 2013-07-09DOI: 10.1007/978-3-319-22527-2_2
S. Shapiro, D. Schlegel
{"title":"Natural language understanding for soft information fusion","authors":"S. Shapiro, D. Schlegel","doi":"10.1007/978-3-319-22527-2_2","DOIUrl":"https://doi.org/10.1007/978-3-319-22527-2_2","url":null,"abstract":"","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128652109","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}
Total Electron Content (TEC) is one of the most important parameters for observing and monitoring Space Weather. GPS Sensor Networks provide a cost-effective means for estimation of TEC. Due to sparsity of TEC measurements and possible outages of data, TEC may have gaps in temporal and spatial domain. In this study, an automatic spatial interpolation algorithm is developed for TEC mapping. The algorithm utilizes isotropic Universal Kriging with linear trend as its main technique for spatial interpolation. The theoretical semivariogram function that represents the spatial correlation structure of TEC is chosen to be from the Matern Family. The trend model parameters are obtained from TEC samples in closed form and the Matern parameters are estimated using Particle Swarm Optimization. The automatic structure enables TEC maps to be obtained with any desired time interval. The developed algorithm is applied to a midlatitude GPS Network and high spatial resolution TEC maps are obtained between May 2009 and May 2012 with 5 minute temporal update period. The algorithm is applicable to any midlatitude GPS network for analysis of temporal and spatial variability of ionosphere.
{"title":"Automatic regional mapping of Total Electron Content using a GPS sensor network and isotropic Universal Kriging","authors":"M. N. Deviren, F. Arıkan, O. Arikan","doi":"10.5072/ZENODO.25872","DOIUrl":"https://doi.org/10.5072/ZENODO.25872","url":null,"abstract":"Total Electron Content (TEC) is one of the most important parameters for observing and monitoring Space Weather. GPS Sensor Networks provide a cost-effective means for estimation of TEC. Due to sparsity of TEC measurements and possible outages of data, TEC may have gaps in temporal and spatial domain. In this study, an automatic spatial interpolation algorithm is developed for TEC mapping. The algorithm utilizes isotropic Universal Kriging with linear trend as its main technique for spatial interpolation. The theoretical semivariogram function that represents the spatial correlation structure of TEC is chosen to be from the Matern Family. The trend model parameters are obtained from TEC samples in closed form and the Matern parameters are estimated using Particle Swarm Optimization. The automatic structure enables TEC maps to be obtained with any desired time interval. The developed algorithm is applied to a midlatitude GPS Network and high spatial resolution TEC maps are obtained between May 2009 and May 2012 with 5 minute temporal update period. The algorithm is applicable to any midlatitude GPS network for analysis of temporal and spatial variability of ionosphere.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115048104","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}
Ionosphere plays an important role in High Frequency (HF) communications. In this paper, Global Ionospheric Maps (GIM) of Total Electron Content (TEC), maximum ionization height (hmF2) and critical frequency (foF2) of F2-layer are presented. These maps are obtained from the Nonlinear Least Squares (NLSQ) optimization of the International Reference Ionosphere (IRI) model extended to the Plasmasphere (IRI-Plas) using the one-hour resolution GIM-TEC (UHR) maps provided by the Polytechnical University of Catalonia (UPC) as the reference input.
{"title":"IRI-Plas optimization based ionospheric tomography","authors":"Onur Cilibas, U. Sezen, F. Arıkan, T. Gulyaeva","doi":"10.5072/ZENODO.25030","DOIUrl":"https://doi.org/10.5072/ZENODO.25030","url":null,"abstract":"Ionosphere plays an important role in High Frequency (HF) communications. In this paper, Global Ionospheric Maps (GIM) of Total Electron Content (TEC), maximum ionization height (hmF2) and critical frequency (foF2) of F2-layer are presented. These maps are obtained from the Nonlinear Least Squares (NLSQ) optimization of the International Reference Ionosphere (IRI) model extended to the Plasmasphere (IRI-Plas) using the one-hour resolution GIM-TEC (UHR) maps provided by the Polytechnical University of Catalonia (UPC) as the reference input.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124294319","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 analyze Bayes fusion rule in details from a fusion standpoint, as well as the emblematic Dempster's rule of combination introduced by Shafer in his Mathematical Theory of evidence based on belief functions. We propose a new interesting formulation of Bayes rule and point out some of its properties. A deep analysis of the compatibility of Dempster's fusion rule with Bayes fusion rule is done. We show that Dempster's rule is compatible with Bayes fusion rule only in the very particular case where the basic belief assignments (bba's) to combine are Bayesian, and when the prior information is modeled either by a uniform probability measure, or by a vacuous bba. We show clearly that Dempster's rule becomes incompatible with Bayes rule in the more general case where the prior is truly informative (not uniform, nor vacuous). Consequently, this paper proves that Dempster's rule is not a generalization of Bayes fusion rule.
{"title":"Why Dempster's fusion rule is not a generalization of Bayes fusion rule","authors":"J. Dezert, A. Tchamova, Deqiang Han, J. Tacnet","doi":"10.5281/ZENODO.22676","DOIUrl":"https://doi.org/10.5281/ZENODO.22676","url":null,"abstract":"In this paper, we analyze Bayes fusion rule in details from a fusion standpoint, as well as the emblematic Dempster's rule of combination introduced by Shafer in his Mathematical Theory of evidence based on belief functions. We propose a new interesting formulation of Bayes rule and point out some of its properties. A deep analysis of the compatibility of Dempster's fusion rule with Bayes fusion rule is done. We show that Dempster's rule is compatible with Bayes fusion rule only in the very particular case where the basic belief assignments (bba's) to combine are Bayesian, and when the prior information is modeled either by a uniform probability measure, or by a vacuous bba. We show clearly that Dempster's rule becomes incompatible with Bayes rule in the more general case where the prior is truly informative (not uniform, nor vacuous). Consequently, this paper proves that Dempster's rule is not a generalization of Bayes fusion rule.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125904382","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. Sensoy, Geeth de Mel, Lance M. Kaplan, T. Pham, T. Norman
In recent years, the number of information sources available to support decision-making has increased dramatically. However, more information sources do not always mean higher precision in the fused information. This is partially due to the fact that some of these sources may be erroneous or malicious. Therefore, it is critical to asses the trust in information before performing fusion. To estimate trust in information, existing approaches use trustworthiness of its source as a proxy. We argue that conflicts between information may also serve as evidence to reduce trust in information. In this paper, we use subjective opinions to represent information from diverse sources. We propose to exploit conflicts between opinions to revise their trustworthiness. For this purpose, we formalise trust revision as a constraint optimisation problem. Through extensive empirical studies, we show that our approach significantly outperform existing ones in the face of malicious information sources.
{"title":"TRIBE: Trust revision for information based on evidence","authors":"M. Sensoy, Geeth de Mel, Lance M. Kaplan, T. Pham, T. Norman","doi":"10.5072/ZENODO.21406","DOIUrl":"https://doi.org/10.5072/ZENODO.21406","url":null,"abstract":"In recent years, the number of information sources available to support decision-making has increased dramatically. However, more information sources do not always mean higher precision in the fused information. This is partially due to the fact that some of these sources may be erroneous or malicious. Therefore, it is critical to asses the trust in information before performing fusion. To estimate trust in information, existing approaches use trustworthiness of its source as a proxy. We argue that conflicts between information may also serve as evidence to reduce trust in information. In this paper, we use subjective opinions to represent information from diverse sources. We propose to exploit conflicts between opinions to revise their trustworthiness. For this purpose, we formalise trust revision as a constraint optimisation problem. Through extensive empirical studies, we show that our approach significantly outperform existing ones in the face of malicious information sources.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124268418","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}
We derive a new algorithm for particle flow with non-zero diffusion corresponding to Bayes' rule, and we report the results of Monte Carlo simulations which show that the new filter is an order of magnitude more accurate than the extended Kalman filter for a difficult nonlinear filter problem. Our new algorithm is simple and fast to compute, and it has an especially nice intuitive formula, which is the same as Newton's method to solve the maximum likelihood estimation (MLE) problem (but for each particle rather than only the MLE), and it is also the same as the extended Kalman filter for the special case of Gaussian densities (but for each particle rather than just the point estimate). All of these particle flows apply to arbitrary multimodal densities with smooth nowhere vanishing non-Gaussian densities.
{"title":"Particle flow for nonlinear filters, Bayesian decisions and transport","authors":"F. Daum, Jim Huang","doi":"10.1117/12.2044122","DOIUrl":"https://doi.org/10.1117/12.2044122","url":null,"abstract":"We derive a new algorithm for particle flow with non-zero diffusion corresponding to Bayes' rule, and we report the results of Monte Carlo simulations which show that the new filter is an order of magnitude more accurate than the extended Kalman filter for a difficult nonlinear filter problem. Our new algorithm is simple and fast to compute, and it has an especially nice intuitive formula, which is the same as Newton's method to solve the maximum likelihood estimation (MLE) problem (but for each particle rather than only the MLE), and it is also the same as the extended Kalman filter for the special case of Gaussian densities (but for each particle rather than just the point estimate). All of these particle flows apply to arbitrary multimodal densities with smooth nowhere vanishing non-Gaussian densities.","PeriodicalId":117803,"journal":{"name":"Proceedings of the 16th International Conference on Information Fusion","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126706943","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：