Volumen 08http://sedici.unlp.edu.ar:80/handle/10915/1346382024-09-13T11:34:36Z2024-09-13T11:34:36ZNota Editorial de Volumen 8Godoy, Daniela LisMaguitman, Ana Gabrielahttp://sedici.unlp.edu.ar:80/handle/10915/1355022022-05-03T20:03:09Z2008-06-26T00:00:00ZContribucion a revista
Electronic Journal of SADIO; vol. 8
Nota sobre el contenido del volumen 8 del Electronic Journal of SADIO.
2008-06-26T00:00:00ZNota sobre el contenido del volumen 8 del Electronic Journal of SADIO.A hybrid wrapper/filter approach for feature subset selectionPrati, Ronaldo C.Batista, Gustavo E. A. P. A.Monard, Maria Carolinahttp://sedici.unlp.edu.ar:80/handle/10915/1354492022-05-02T20:03:30Z2008-06-26T00:00:00ZArticulo
Electronic Journal of SADIO; vol. 8
This work presents a hybrid wrapper/ﬁlter algorithm for feature subset selection that can use a combination of several quality criteria measures to rank the set of features of a dataset. These ranked features are used to prune the search space of subsets of possible features such that the number of times the wrapper executes the learning algorithm for a dataset with M features is reduced to O(M) runs. Experimental results using 14 datasets show that, for most of the datasets, the AUC assessed using the reduced feature set is comparable to the AUC of the model constructed using all the features. Furthermore, the algorithm archieved a good reduction in the number of features.
2008-06-26T00:00:00ZThis work presents a hybrid wrapper/ﬁlter algorithm for feature subset selection that can use a combination of several quality criteria measures to rank the set of features of a dataset. These ranked features are used to prune the search space of subsets of possible features such that the number of times the wrapper executes the learning algorithm for a dataset with M features is reduced to O(M) runs. Experimental results using 14 datasets show that, for most of the datasets, the AUC assessed using the reduced feature set is comparable to the AUC of the model constructed using all the features. Furthermore, the algorithm archieved a good reduction in the number of features.An Immune-based Approach to Student DiagnosisWebber, CarineTavares da Silva, João Luíshttp://sedici.unlp.edu.ar:80/handle/10915/1354152022-05-02T20:03:31Z2008-06-26T00:00:00ZArticulo
Electronic Journal of SADIO; vol. 8
Biologically-inspired approaches constitute innovative problem solving techniques that have been applied to several domains including monitoring, detection and diagnosis. The human immune system (HIS) has especially motivated the development of new approaches to deal with problems where complexity and distribution are crucial constraints. Work in this paper reflects how characteristics from the HIS can be applied to conceive diagnosis systems. An application was implemented to student diagnosis to be integrated to Modal, an educational environment to the learning of basic programming skills.
2008-06-26T00:00:00ZBiologically-inspired approaches constitute innovative problem solving techniques that have been applied to several domains including monitoring, detection and diagnosis. The human immune system (HIS) has especially motivated the development of new approaches to deal with problems where complexity and distribution are crucial constraints. Work in this paper reflects how characteristics from the HIS can be applied to conceive diagnosis systems. An application was implemented to student diagnosis to be integrated to Modal, an educational environment to the learning of basic programming skills.Using Neural Networks to improve classical Operating System Fingerprinting techniquesSarraute, CarlosBurroni, Javierhttp://sedici.unlp.edu.ar:80/handle/10915/1354082022-05-02T20:03:32Z2008-06-26T00:00:00ZArticulo
Electronic Journal of SADIO; vol. 8
We present remote Operating System detection as an inference problem: given a set of observations (the target host responses to a set of tests), we want to infer the OS type which most probably generated these observations. Classical techniques used to perform this analysis present several limitations. To improve the analysis, we have developed tools using neural networks and Statistics tools. We present two working modules: one which uses DCE-RPC endpoints to distinguish Windows versions, and another which uses Nmap signatures to distinguish different version of Windows, Linux, Solaris, OpenBSD, FreeBSD and NetBSD systems. We explain the details of the topology and inner workings of the neural networks used, and the ﬁne tuning of their parameters. Finally we show positive experimental results.
2008-06-26T00:00:00ZWe present remote Operating System detection as an inference problem: given a set of observations (the target host responses to a set of tests), we want to infer the OS type which most probably generated these observations. Classical techniques used to perform this analysis present several limitations. To improve the analysis, we have developed tools using neural networks and Statistics tools. We present two working modules: one which uses DCE-RPC endpoints to distinguish Windows versions, and another which uses Nmap signatures to distinguish different version of Windows, Linux, Solaris, OpenBSD, FreeBSD and NetBSD systems. We explain the details of the topology and inner workings of the neural networks used, and the ﬁne tuning of their parameters. Finally we show positive experimental results.On Artificial Gene Regulatory NetworksCarballido, Jessica A.Ponzoni, Ignaciohttp://sedici.unlp.edu.ar:80/handle/10915/1354062022-05-02T20:03:33Z2008-06-26T00:00:00ZArticulo
Electronic Journal of SADIO; vol. 8
Gene regulatory networks (GRNs) represent dependencies between genes and their products during protein synthesis at the molecular level. At the present there exist many inference methods that infer GRNs form observed data. However, gene expression data sets have in general considerable noise that make understanding and learning even simple regulatory patterns difficult. Also, there is no well-known method to test the accuracy of inferred GRNs. Given these drawbacks, characterizing the effectiveness of different techniques to uncover gene networks remains a challenge. The development of artificial GRNs with known biological features of expression complexity, diversity and interconnectivities provides a more controlled means of investigating the appropriateness of those techniques. In this work we introduce this problem in terms of machine learning and present a review of the main formalisms that have been used
2008-06-26T00:00:00ZGene regulatory networks (GRNs) represent dependencies between genes and their products during protein synthesis at the molecular level. At the present there exist many inference methods that infer GRNs form observed data. However, gene expression data sets have in general considerable noise that make understanding and learning even simple regulatory patterns difficult. Also, there is no well-known method to test the accuracy of inferred GRNs. Given these drawbacks, characterizing the effectiveness of different techniques to uncover gene networks remains a challenge. The development of artificial GRNs with known biological features of expression complexity, diversity and interconnectivities provides a more controlled means of investigating the appropriateness of those techniques. In this work we introduce this problem in terms of machine learning and present a review of the main formalisms that have been used