The Experts below are selected from a list of 327 Experts worldwide ranked by ideXlab platform
Yiu-ming Cheung - One of the best experts on this subject based on the ideXlab platform.
-
An Ordinal Data Clustering Algorithm with Automated Distance Learning
Proceedings of the AAAI Conference on Artificial Intelligence, 2020Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:Clustering Ordinal Data is a common task in Data mining and machine learning fields. As a major type of categorical Data, Ordinal Data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, Ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between Ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory Ordinal Data clustering results. This paper, therefore, proposes a novel Ordinal Data clustering algorithm, which iteratively learns: 1) The partition of Ordinal Dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of Ordinal Data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.
-
AAAI - An Ordinal Data Clustering Algorithm with Automated Distance Learning
2020Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:Clustering Ordinal Data is a common task in Data mining and machine learning fields. As a major type of categorical Data, Ordinal Data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, Ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between Ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory Ordinal Data clustering results. This paper, therefore, proposes a novel Ordinal Data clustering algorithm, which iteratively learns: 1) The partition of Ordinal Dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of Ordinal Data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.
-
ISMIS - Exploiting Order Information Embedded in Ordered Categories for Ordinal Data Clustering
Lecture Notes in Computer Science, 2018Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:As a major type of categorical Data, Ordinal Data are those with the attributes whose possible values (also called categories interchangeably) are naturally ordered. As far as we know, all the existing distance metrics proposed for categorical Data do not take the underlying order information into account during the distance measurement. This will make the produced distance incorrect and will further influence the results of Ordinal Data clustering. We therefore propose a specially designed distance metric, which can exploit the order information embedded in the ordered categories for distance measurement. It quantifies the distance between two Ordinal categories by accumulating the sub-entropies of all the categories ordered between them. Since the proposed distance metric takes the order information into account, distance produced by it will be more reasonable than the other metrics proposed for categorical Data. Moreover, it is parameter-free and can be easily applied to different Ordinal Data clustering tasks. Experimental results show the promising advantages of the proposed distance metric.
Christophe Biernacki - One of the best experts on this subject based on the ideXlab platform.
-
OrdinalClust: An R Package to Analyze Ordinal Data
The R Journal, 2021Co-Authors: Margot Selosse, Julien Jacques, Christophe BiernackiAbstract:Ordinal Data are used in many domains, especially when measurements are collected from people through observations, tests, or questionnaires. OrdinalClust is an innovative R package dedicated to Ordinal Data that provides tools for modeling, clustering, co-clustering and classifying such Data. Ordinal Data are modeled using the BOS distribution, which is a model with two meaningful parameters referred to as "position" and "precision". The former indicates the mode of the distribution and the latter describes how scattered the Data are around the mode: the user is able to easily interpret the distribution of their Data when given these two parameters. The package is based on the coclustering framework (when rows and columns are simultaneously clustered). The co-clustering approach uses the Latent Block Model (LBM) and the SEM-Gibbs algorithm for parameter inference. On the other hand, the clustering and the classification methods follow on from simplified versions of the SEM-Gibbs algorithm. For the classification process, two approaches are proposed. In the first one, the BOS parameters are estimated from the training Dataset in the conventional way. In the second approach, parsimony is introduced by estimating the parameters and column-clusters from the training Dataset. We empirically show that this approach can yield better results. For the clustering and co-clustering processes, the ICL-BIC criterion is used for model selection purposes. An overview of these methods is given, and the way to use them with the OrdinalClust package is described using real Datasets. The latest stable package version is available on the Comprehensive R Archive Network (CRAN).
-
OrdinalClust: An R Package to Analyze Ordinal Data
The R Journal, 2020Co-Authors: Margot Selosse, Julien Jacques, Christophe BiernackiAbstract:Ordinal Data are used in a lot of domains, especially when measurements are collected from persons by observations, testings, or questionnaires. OrdinalClust is an R package dedicated to Ordinal Data that proposes tools for modeling, clustering, co-clustering and classification. Ordinal Data are modeled by the BOS distribution, which is a meaningful model parametrized by a position and a precision parameter. On one hand, the co-clustering framework uses the Latent Block Model (LBM) and an SEM-Gibbs algorithm for the parameters inference. On the other hand, the clustering and the classification methods follow on from simplified versions of this algorithm. An overview of these methods is given, and the way of using them with the OrdinalClust package is described through real Datasets.
-
Model-Based Co-clustering for Ordinal Data
Computational Statistics & Data Analysis, 2018Co-Authors: Julien Jacques, Christophe BiernackiAbstract:A model-based co-clustering algorithm for Ordinal Data is presented. This algorithm relies on the latent block model embedding a probability distribution specific to Ordinal Data (the so-called BOS or Binary Ordinal Search distribution). Model inference relies on a Stochastic EM algorithm coupled with a Gibbs sampler, and the ICL-BIC criterion is used for selecting the number of co-clusters (or blocks). The main advantage of this Ordinal dedicated co-clustering model is its parsimony, the interpretability of the co-cluster parameters (mode, precision) and the possibility to take into account missing Data. Numerical experiments on simulated Data show the efficiency of the inference strategy, and real Data analyses illustrate the interest of the proposed procedure.
-
Model-based co-clustering for Ordinal Data
2016Co-Authors: Julien Jacques, Christophe BiernackiAbstract:A model-based coclustering algorithm for Ordinal Data is presented. This algorithm relies on the latent block model using the BOS model (Biernacki and Jacques, 2015, Stat. Comput.) for Ordinal Data and a SEM-Gibbs algorithm for inference. Nu- merical experiments on simulated Data illustrate the eciency of the inference strategy.
Yiqun Zhang - One of the best experts on this subject based on the ideXlab platform.
-
An Ordinal Data Clustering Algorithm with Automated Distance Learning
Proceedings of the AAAI Conference on Artificial Intelligence, 2020Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:Clustering Ordinal Data is a common task in Data mining and machine learning fields. As a major type of categorical Data, Ordinal Data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, Ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between Ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory Ordinal Data clustering results. This paper, therefore, proposes a novel Ordinal Data clustering algorithm, which iteratively learns: 1) The partition of Ordinal Dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of Ordinal Data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.
-
AAAI - An Ordinal Data Clustering Algorithm with Automated Distance Learning
2020Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:Clustering Ordinal Data is a common task in Data mining and machine learning fields. As a major type of categorical Data, Ordinal Data is composed of attributes with naturally ordered possible values (also called categories interchangeably in this paper). However, due to the lack of dedicated distance metric, Ordinal categories are usually treated as nominal ones, or coded as consecutive integers and treated as numerical ones. Both these two common ways will roughly define the distances between Ordinal categories because the former way ignores the order relationship and the latter way simply assigns identical distances to different pairs of adjacent categories that may have intrinsically unequal distances. As a result, they may produce unsatisfactory Ordinal Data clustering results. This paper, therefore, proposes a novel Ordinal Data clustering algorithm, which iteratively learns: 1) The partition of Ordinal Dataset, and 2) the inter-category distances. To the best of our knowledge, this is the first attempt to dynamically adjust inter-category distances during the clustering process to search for a better partition of Ordinal Data. The proposed algorithm features superior clustering accuracy, low time complexity, fast convergence, and is parameter-free. Extensive experiments show its efficacy.
-
ISMIS - Exploiting Order Information Embedded in Ordered Categories for Ordinal Data Clustering
Lecture Notes in Computer Science, 2018Co-Authors: Yiqun Zhang, Yiu-ming CheungAbstract:As a major type of categorical Data, Ordinal Data are those with the attributes whose possible values (also called categories interchangeably) are naturally ordered. As far as we know, all the existing distance metrics proposed for categorical Data do not take the underlying order information into account during the distance measurement. This will make the produced distance incorrect and will further influence the results of Ordinal Data clustering. We therefore propose a specially designed distance metric, which can exploit the order information embedded in the ordered categories for distance measurement. It quantifies the distance between two Ordinal categories by accumulating the sub-entropies of all the categories ordered between them. Since the proposed distance metric takes the order information into account, distance produced by it will be more reasonable than the other metrics proposed for categorical Data. Moreover, it is parameter-free and can be easily applied to different Ordinal Data clustering tasks. Experimental results show the promising advantages of the proposed distance metric.
Martyna Kobus - One of the best experts on this subject based on the ideXlab platform.
-
Polarization measurement for Ordinal Data
The Journal of Economic Inequality, 2014Co-Authors: Martyna KobusAbstract:Atkinson’s Theorem (Atkinson J. Econ. Theory 2, 244–263, 1970) is a classic result in inequality measurement. It establishes Lorenz dominance as a useful criterion for comparative judgements of inequality between distributions. If distribution A Lorenz dominates distribution B, then all indices in a broad class of measures must confirm A as less unequal than B. Recent research, however, shows that standard inequality theory cannot be applied to Ordinal Data (Zheng Res. Econ. Inequal. 16, 177–188, 2008), such as self-reported health status or educational attainment. A new theory in development (Abul Naga and Yalcin J. Health Econ. 27(6), 1614–1625, 2008) measures disparity of Ordinal Data as polarization. Typically a criterion used to compare distributions is the polarization relation as proposed by Allison and Foster (J. Health Econ. 23(3), 505–524, 2004). We characterize classes of polarization measures equivalent to the AF relation analogously to Atkinson’s original approach.
-
Inequality decomposition by population subgroups for Ordinal Data
Journal of health economics, 2011Co-Authors: Martyna Kobus, Piotr MiłośAbstract:We present a class of decomposable inequality indices for Ordinal Data (e.g. self-reported health survey). It is characterized by well-known inequality axioms (e.g. scale invariance) and a decomposability axiom which states that an index can be represented as a function of inequality values in subgroups and subgroup sizes. The only decomposable indices are strictly monotonic transformations of the weighted average of frequencies in categories. Among the indices proposed in the literature only the absolute value index (Abul Naga and Yalcin, 2008; Apouey, 2007) is decomposable. As an empirical illustration we calculate regional contributions to overall health inequality in Switzerland.
Maria Iannario - One of the best experts on this subject based on the ideXlab platform.
-
Modelling Uncertainty and Overdispersion in Ordinal Data
Communications in Statistics - Theory and Methods, 2014Co-Authors: Maria IannarioAbstract:In this article we introduce a probability distribution generated by a mixture of discrete random variables to capture uncertainty, feeling, and overdispersion, possibly present in Ordinal Data surveys. The choice of the components of the new model is motivated by a study on the Data generating process. Inferential issues concerning the maximum likelihood estimates and the validation steps are presented; then, some empirical analyses are given to support the usefulness of the approach. Discussion on further extensions of the model ends the article.
-
On the identifiability of a mixture model for Ordinal Data
METRON, 2010Co-Authors: Maria IannarioAbstract:In this article we discuss the identifiability of a probability model which has been proven useful for capturing the main features of Ordinal Data generated by rating surveys. Specifically, we show that the mixture of a shifted Binomial and a Uniform discrete distribution is identifiable when the number of categories is greater than three.
-
Qualitative and quantitative models for Ordinal Data analysis
2008Co-Authors: Domenico Piccolo, Maria IannarioAbstract:En In this paper, we explore and compare classical regression and Ordinal Data models when quantitative Data are related to a qualitative assessment. Specifically, we test the approach on a Data set of graduated students and we check the relative performance and the interpretative content of the models. Some further comments end the paper.