Pub Date : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.63
Fabricio A. Breve, Liang Zhao
Concept drift, which refers to non stationary learning problems over time, has increasing importance in machine learning and data mining. Many concept drift applications require fast response, which means an algorithm must always be (re)trained with the latest available data. But the process of data labeling is usually expensive and/or time consuming when compared to acquisition of unlabeled data, thus usually only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are based on assumptions that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenging task in machine learning. Recently, a particle competition and cooperation approach has been developed to realize graph-based semi-supervised learning from static data. We have extend that approach to handle data streams and concept drift. The result is a passive algorithm which uses a single classifier approach, naturally adapted to concept changes without any explicit drift detection mechanism. It has built-in mechanisms that provide a natural way of learning from new data, gradually "forgetting" older knowledge as older data items are no longer useful for the classification of newer data items. The proposed algorithm is applied to the KDD Cup 1999 Data of network intrusion, showing its effectiveness.
概念漂移是指随着时间推移的非平稳学习问题，在机器学习和数据挖掘中越来越重要。许多概念漂移应用需要快速响应，这意味着算法必须始终使用最新可用数据进行(重新)训练。但是，与获取未标记的数据相比，数据标记的过程通常是昂贵的和/或耗时的，因此通常只有一小部分传入数据可以有效地标记。在这种情况下，半监督学习方法可能会有所帮助，因为它们在训练过程中同时使用标记和未标记的数据。然而，它们中的大多数都是基于数据是静态的假设。因此，带有概念漂移的半监督学习仍然是机器学习中一个开放的具有挑战性的任务。近年来，人们提出了一种粒子竞争与合作的方法来实现基于图的静态数据半监督学习。我们已经扩展了这种方法来处理数据流和概念漂移。结果是使用单一分类器方法的被动算法，自然地适应概念变化，没有任何显式的漂移检测机制。它具有内置的机制，提供了一种从新数据中学习的自然方式，随着旧数据项对新数据项的分类不再有用，逐渐“忘记”旧知识。将该算法应用于KDD Cup 1999网络入侵数据，验证了算法的有效性。
{"title":"Semi-supervised Learning with Concept Drift Using Particle Dynamics Applied to Network Intrusion Detection Data","authors":"Fabricio A. Breve, Liang Zhao","doi":"10.1109/BRICS-CCI-CBIC.2013.63","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.63","url":null,"abstract":"Concept drift, which refers to non stationary learning problems over time, has increasing importance in machine learning and data mining. Many concept drift applications require fast response, which means an algorithm must always be (re)trained with the latest available data. But the process of data labeling is usually expensive and/or time consuming when compared to acquisition of unlabeled data, thus usually only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are based on assumptions that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenging task in machine learning. Recently, a particle competition and cooperation approach has been developed to realize graph-based semi-supervised learning from static data. We have extend that approach to handle data streams and concept drift. The result is a passive algorithm which uses a single classifier approach, naturally adapted to concept changes without any explicit drift detection mechanism. It has built-in mechanisms that provide a natural way of learning from new data, gradually \"forgetting\" older knowledge as older data items are no longer useful for the classification of newer data items. The proposed algorithm is applied to the KDD Cup 1999 Data of network intrusion, showing its effectiveness.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125368968","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.84
Erik Alexandre Pucci, Aurora Trinidad Ramirez Pozo, E. Spinosa
Estimation of Distribution Algorithms (EDA) are stochastic population based search algorithms that use a distribution model of the population to create new candidate solutions. One problem that directly affects the EDAs' ability to find the best solutions is the premature convergence to some local optimum due to diversity loss. Inspired by the Random Immigrants technique, this paper presents the Bayesian Optimization Algorithm with Random Immigration (BOARI). The algorithm generates and migrates random individuals as a way to improve the performance of the Bayesian Optimization Algorithm (BOA) by maintaining the genetic diversity of the population along the generations. The proposed approach has been evaluated and compared to BOA using benchmark functions. Results indicate that, with appropriate settings, the algorithm is able to achieve better solutions than the standard BOA for these functions.
{"title":"Bayesian Optimization Algorithm with Random Immigration","authors":"Erik Alexandre Pucci, Aurora Trinidad Ramirez Pozo, E. Spinosa","doi":"10.1109/BRICS-CCI-CBIC.2013.84","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.84","url":null,"abstract":"Estimation of Distribution Algorithms (EDA) are stochastic population based search algorithms that use a distribution model of the population to create new candidate solutions. One problem that directly affects the EDAs' ability to find the best solutions is the premature convergence to some local optimum due to diversity loss. Inspired by the Random Immigrants technique, this paper presents the Bayesian Optimization Algorithm with Random Immigration (BOARI). The algorithm generates and migrates random individuals as a way to improve the performance of the Bayesian Optimization Algorithm (BOA) by maintaining the genetic diversity of the population along the generations. The proposed approach has been evaluated and compared to BOA using benchmark functions. Results indicate that, with appropriate settings, the algorithm is able to achieve better solutions than the standard BOA for these functions.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123071764","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.117
Vinicius da F Vieira, C. R. Xavier, Alexandre Evsukoff
One of the most important features of a network is its division into communities, groups of nodes with many internal and few external connections. Furthermore, the community structure of a network can be organized hierarchically, which reflects a natural behavior of real life phenomena. It is a difficult task to detect and understand the community structure of a network and it becomes even more challenging as data availability (and networks sizes) increases. This work presents a efficient implementation for community detection in networks aiming on modularity maximization based on the Newman's spectral method with a fine tuning(FT) stage. This work presents a modification on the FT which substantially reduces the execution time, while preserving the division quality. A high performance implementation of the method enables their application to large real world networks. The Newman's spectral method can be applied to networks with more than 1 million nodes in a personal computer.
{"title":"Efficient Community Detection in Large Scale Networks","authors":"Vinicius da F Vieira, C. R. Xavier, Alexandre Evsukoff","doi":"10.1109/BRICS-CCI-CBIC.2013.117","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.117","url":null,"abstract":"One of the most important features of a network is its division into communities, groups of nodes with many internal and few external connections. Furthermore, the community structure of a network can be organized hierarchically, which reflects a natural behavior of real life phenomena. It is a difficult task to detect and understand the community structure of a network and it becomes even more challenging as data availability (and networks sizes) increases. This work presents a efficient implementation for community detection in networks aiming on modularity maximization based on the Newman's spectral method with a fine tuning(FT) stage. This work presents a modification on the FT which substantially reduces the execution time, while preserving the division quality. A high performance implementation of the method enables their application to large real world networks. The Newman's spectral method can be applied to networks with more than 1 million nodes in a personal computer.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"329 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116123383","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.87
Rafael Xavier Valente, Antonio Braga, W. Pedrycz
This paper presents a new validity index for fuzzy partitions generated by the fuzzy c-means algorithm. The proposed validity index is based on the calculation of factors from the proximity matrix generated from the membership matrix generated by a fuzzy clustering partition algorithm, such as FCM. The experimental results show that the proposed approach is consistent with other well-known metrics and with the dataset structure as observed from Proximity Matrices.
{"title":"A New Fuzzy Clustering Validity Index Based on Fuzzy Proximity Matrices","authors":"Rafael Xavier Valente, Antonio Braga, W. Pedrycz","doi":"10.1109/BRICS-CCI-CBIC.2013.87","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.87","url":null,"abstract":"This paper presents a new validity index for fuzzy partitions generated by the fuzzy c-means algorithm. The proposed validity index is based on the calculation of factors from the proximity matrix generated from the membership matrix generated by a fuzzy clustering partition algorithm, such as FCM. The experimental results show that the proposed approach is consistent with other well-known metrics and with the dataset structure as observed from Proximity Matrices.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122458175","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.52
Jobson Renan, J. Silva, S. Galdino
This paper introduces two new approaches to fit univariate resistant linear regression models on interval-valued data. Linear regressions on interval-valued data gives point predictions. The prediction of the lower and upper bounds from interval-valued data of dependent variable are estimated from the fitted range resistant linear regression model. The new proposed methods should be used in presence of outliers.
{"title":"Resistant Regression for Interval-Valued Data","authors":"Jobson Renan, J. Silva, S. Galdino","doi":"10.1109/BRICS-CCI-CBIC.2013.52","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.52","url":null,"abstract":"This paper introduces two new approaches to fit univariate resistant linear regression models on interval-valued data. Linear regressions on interval-valued data gives point predictions. The prediction of the lower and upper bounds from interval-valued data of dependent variable are estimated from the fitted range resistant linear regression model. The new proposed methods should be used in presence of outliers.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"219 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122838965","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.27
D. A. Augusto, H. Bernardino, H. Barbosa
Since their early development, genetic programming-based algorithms have been showing to be successful at challenging problems, attaining several human-competitive results and other awards. This paper will present another achievement of such algorithms by describing how our team has won an international machine-learning competition. We have solved, by means of grammar-based genetic programming techniques, a real-world problem of meritocracy in jobs by evolving classifiers that were both accurate and human-readable.
{"title":"Predicting the Performance of Job Applicants by Means of Genetic Programming","authors":"D. A. Augusto, H. Bernardino, H. Barbosa","doi":"10.1109/BRICS-CCI-CBIC.2013.27","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.27","url":null,"abstract":"Since their early development, genetic programming-based algorithms have been showing to be successful at challenging problems, attaining several human-competitive results and other awards. This paper will present another achievement of such algorithms by describing how our team has won an international machine-learning competition. We have solved, by means of grammar-based genetic programming techniques, a real-world problem of meritocracy in jobs by evolving classifiers that were both accurate and human-readable.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132916784","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.54
T. C. Silva, Liang Zhao
Traditional data classification considers only physical features (e.g., geometrical or statistical features) of the input data. Here, it is referred to low level classification. In contrast, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is here called high level classification. In this paper, we present an alternative technique which combines both low and high level data classification techniques. The low level term can be implemented by any classification technique, while the high level term is realized by means of the extraction of the underlying network's features (graph) constructed from the input data, which measures the compliance of the test instances with the pattern formation of the training data. Out of various high level perspectives that can be utilized to capture semantical meaning, we utilize the dynamical features that are generated from a tourist walker in a networked environment. Specifically, a weighted combination of transient and cycle lengths are employed for that end. Furthermore, we show computer simulations with synthetic and widely accepted real-world data sets from the machine learning literature. Interestingly, our study shows that the proposed technique is able to further improve the already optimized performance of traditional classification techniques.
{"title":"Pattern-Based Classification via a High Level Approach Using Tourist Walks in Networks","authors":"T. C. Silva, Liang Zhao","doi":"10.1109/BRICS-CCI-CBIC.2013.54","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.54","url":null,"abstract":"Traditional data classification considers only physical features (e.g., geometrical or statistical features) of the input data. Here, it is referred to low level classification. In contrast, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is here called high level classification. In this paper, we present an alternative technique which combines both low and high level data classification techniques. The low level term can be implemented by any classification technique, while the high level term is realized by means of the extraction of the underlying network's features (graph) constructed from the input data, which measures the compliance of the test instances with the pattern formation of the training data. Out of various high level perspectives that can be utilized to capture semantical meaning, we utilize the dynamical features that are generated from a tourist walker in a networked environment. Specifically, a weighted combination of transient and cycle lengths are employed for that end. Furthermore, we show computer simulations with synthetic and widely accepted real-world data sets from the machine learning literature. Interestingly, our study shows that the proposed technique is able to further improve the already optimized performance of traditional classification techniques.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134514578","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.17
Tiago F. B. de Sousa, Marcelo A. C. Fernandes
Optical fibers are commonly used in communications today, mainly because that the data transmission rates of those systems are faster than those in any other digital communication system. Despite this great advantage, some problems prevent the full use of optical connection: by increasing transmission rates over longer distances, the data is affected by non-linear inter-symbol interference caused by the dispersion phenomena in the fiber. Adaptive equalizers can be used to compensate for the effects caused by channel non-linear responses, restoring the originally transmitted signal. The present study discusses a proposal based on artificial neural networks, a neural equalizer. The proposal is validated through a simulated optic channel and the comparison with other adaptive equalization techniques.
{"title":"Bi-dimensional Neural Equalizer Applied to Optical Receiver","authors":"Tiago F. B. de Sousa, Marcelo A. C. Fernandes","doi":"10.1109/BRICS-CCI-CBIC.2013.17","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.17","url":null,"abstract":"Optical fibers are commonly used in communications today, mainly because that the data transmission rates of those systems are faster than those in any other digital communication system. Despite this great advantage, some problems prevent the full use of optical connection: by increasing transmission rates over longer distances, the data is affected by non-linear inter-symbol interference caused by the dispersion phenomena in the fiber. Adaptive equalizers can be used to compensate for the effects caused by channel non-linear responses, restoring the originally transmitted signal. The present study discusses a proposal based on artificial neural networks, a neural equalizer. The proposal is validated through a simulated optic channel and the comparison with other adaptive equalization techniques.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114533609","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.121
E. Christo, M. Ferreira, K. C. Alonso
The forecasting demand is the basis of strategic planning for production, sales and finances of any company. They are used for planning and control of production for planning productive system (long term) and the using (short term) of this system. With the increasing of the competition in the automobile market, there are, consequently, the increasing of concerning about establishing a balance between offering and demand of vehicles. Then come the need to calculate statistical predictions of future demands, which are translated into a real approximation of future events of the company in question. Thus, this work is divided in two stages: first - find out the best forecasting model (lower mean percentage of error between the actual and predicted) for the vehicle demand, second - analyze the residuals control charts of the best forecasting model so that to observe and draw the points that may be outside the control limits. The main goal is to plan the production of vehicle sales within a stipulated period and create scenarios for future periods.
{"title":"Use of Statistical Control for Improved Demand Forecasting","authors":"E. Christo, M. Ferreira, K. C. Alonso","doi":"10.1109/BRICS-CCI-CBIC.2013.121","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.121","url":null,"abstract":"The forecasting demand is the basis of strategic planning for production, sales and finances of any company. They are used for planning and control of production for planning productive system (long term) and the using (short term) of this system. With the increasing of the competition in the automobile market, there are, consequently, the increasing of concerning about establishing a balance between offering and demand of vehicles. Then come the need to calculate statistical predictions of future demands, which are translated into a real approximation of future events of the company in question. Thus, this work is divided in two stages: first - find out the best forecasting model (lower mean percentage of error between the actual and predicted) for the vehicle demand, second - analyze the residuals control charts of the best forecasting model so that to observe and draw the points that may be outside the control limits. The main goal is to plan the production of vehicle sales within a stipulated period and create scenarios for future periods.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122713527","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 : 2013-09-08DOI: 10.1109/BRICS-CCI-CBIC.2013.76
F. Corona, Zhanxing Zhu, Amauri H. Souza Junior, M. Mulas, G. Barreto, R. Baratti
In this work, we discuss a recently proposed approach for supervised dimensionality reduction, the Supervised Distance Preserving Projection (SDPP) and, we investigate its applicability to monitoring material's properties from spectroscopic observations using Local Linear Regression (LLR). An experimental evaluation is conducted to show the performance of the SDPP and LLR and compare it with a number of state-of-the-art approaches for unsupervised and supervised dimensionality reduction. For the task, the results obtained on a benchmark problem consisting of a set of NIR spectra of diesel fuels and six different chemico-physical properties of those fuels are discussed. Based on the experimental results, the SDPP leads to accurate and parsimonious projections that can be effectively used in the design of estimation models based on local linear regression.
{"title":"Monitoring Diesel Fuels with Supervised Distance Preserving Projections and Local Linear Regression","authors":"F. Corona, Zhanxing Zhu, Amauri H. Souza Junior, M. Mulas, G. Barreto, R. Baratti","doi":"10.1109/BRICS-CCI-CBIC.2013.76","DOIUrl":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.76","url":null,"abstract":"In this work, we discuss a recently proposed approach for supervised dimensionality reduction, the Supervised Distance Preserving Projection (SDPP) and, we investigate its applicability to monitoring material's properties from spectroscopic observations using Local Linear Regression (LLR). An experimental evaluation is conducted to show the performance of the SDPP and LLR and compare it with a number of state-of-the-art approaches for unsupervised and supervised dimensionality reduction. For the task, the results obtained on a benchmark problem consisting of a set of NIR spectra of diesel fuels and six different chemico-physical properties of those fuels are discussed. Based on the experimental results, the SDPP leads to accurate and parsimonious projections that can be effectively used in the design of estimation models based on local linear regression.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125805621","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：