Academic literature on the topic 'Categorical Principal Component Analysis (CATPCA)'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Categorical Principal Component Analysis (CATPCA).'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Journal articles on the topic "Categorical Principal Component Analysis (CATPCA)"
1
Saukani, Nasir, and Noor Azina Ismail. "Identifying the Components of Social Capital by Categorical Principal Component Analysis (CATPCA)." Social Indicators Research 141, no. 2 (January 31, 2018): 631–55. http://dx.doi.org/10.1007/s11205-018-1842-2.
Full text
2
Del Casale, Antonio, Stefano Ferracuti, Alessio Mosca, Leda Marina Pomes, Federica Fiaschè, Luca Bonanni, Marina Borro, Giovanna Gentile, Paolo Martelletti, and Maurizio Simmaco. "Multiple Chemical Sensitivity Syndrome: A Principal Component Analysis of Symptoms." International Journal of Environmental Research and Public Health 17, no. 18 (September 9, 2020): 6551. http://dx.doi.org/10.3390/ijerph17186551.
Full textAbstract:
Multiple Chemical Sensitivity (MCS) is a chronic and/or recurrent condition with somatic, cognitive, and affective symptoms following a contact with chemical agents whose concentrations do not correlate with toxicity in the general population. Its prevalence is not well defined; it mainly affects women between 40 and 50 years, without variations in ethnicity, education and economic status. We aimed to assess the core symptoms of this illness in a sample of Italian patients. Two physicians investigated different symptoms with a checklist compilation in 129 patients with MCS (117 women). We conducted a categorical Principal Component Analysis (CATPCA) with Varimax rotation on the checklist dataset. A typical triad was documented: hyperosmia, asthenia, and dyspnoea were the most common symptoms. Patients also frequently showed cough and headache. The CATPCA showed seven main factors: 1, neurocognitive symptoms; 2, physical (objective) symptoms; 3, gastrointestinal symptoms; 4, dermatological symptoms; 5, anxiety-depressive symptoms; 6, respiratory symptoms; 7, hyperosmia and asthenia. Patients showed higher mean prevalence of factors 7 (89.9%), 6 (71.7%), and 1 (62.13%). In conclusion, MCS patients frequently manifest hyperosmia, asthenia, and dyspnoea, which are often concomitant with other respiratory and neurocognitive symptoms. Considering the clinical association that is often made with anxiety, more studies are necessary on the psychosomatic aspects of this syndrome. Further analytical epidemiological studies are needed to support the formulation of aetiological hypotheses of MCS.
3
Jiménez, Rafael, Elena Gervilla, Albert Sesé, Juan José Montaño, Berta Cajal, and Alfonso Palmer. "Dimensionality Reduction in Data Mining Using Artificial Neural Networks." Methodology 5, no. 1 (January 2009): 26–34. http://dx.doi.org/10.1027/1614-2241.5.1.26.
Full textAbstract:
The use of classic dimension reduction techniques can be considered customary practice within the context of data mining (DM). Nevertheless, although artificial neural networks (ANNs) are one of the most important DM techniques, specific ANN architectures for dimensionality reduction, such as the principal components analysis ANN (PCA-ANN) and the linear auto-associative ANN (LA-ANN), are used on far fewer occasions. In this study, categorical principal component analysis (CATPCA) and the two ANN procedures are studied and compared searching for uniqueness in an applied context relative to personality variables and drug consumption. A sample of 7,030 adolescents completed a personality test made up of 20 dichotomous items with a hypothesized four-factor latent model. Results point out that both ANN factor solutions converge to those obtained using CATPCA. Nevertheless, possible drawbacks of the ANN techniques lie in their relatively complex application, as well as in the need to use visual graphic analysis as a support for interpreting the factorized solutions.
4
Navas González, Francisco, Jordi Jordana Vidal, Gabriela Pizarro Inostroza, Ander Arando Arbulu, and Juan Delgado Bermejo. "Can Donkey Behavior and Cognition Be Used to Trace Back, Explain, or Forecast Moon Cycle and Weather Events?" Animals 8, no. 11 (November 19, 2018): 215. http://dx.doi.org/10.3390/ani8110215.
Full textAbstract:
Donkeys have been reported to be highly sensitive to environmental changes. Their 8900–8400-year-old evolution process made them interact with diverse environmental situations that were very distant from their harsh origins. These changing situations not only affect donkeys’ short-term behavior but may also determine their long-term cognitive skills from birth. Thus, animal behavior becomes a useful tool to obtain past, present or predict information from the environmental situation of a particular area. We performed an operant conditioning test on 300 donkeys to assess their response type, mood, response intensity, and learning capabilities, while we simultaneously registered 14 categorical environmental factors. We quantified the effect power of such environmental factors on donkey behavior and cognition. We used principal component analysis (CATPCA) to reduce the number of factors affecting each behavioral variable and built categorical regression (CATREG) equations to model for the effects of potential factor combinations. Effect power ranged from 7.9% for the birth season on learning (p < 0.05) to 38.8% for birth moon phase on mood (p < 0.001). CATPCA suggests the percentage of variance explained by a four-dimension-model (comprising the dimensions of response type, mood, response intensity and learning capabilities), is 75.9%. CATREG suggests environmental predictors explain 28.8% of the variability of response type, 37.0% of mood, and 37.5% of response intensity, and learning capabilities.
5
Kohler, Friedbert, Roger Renton, Hugh G. Dickson, John Estell, and Carol E. Connolly. "Subacute casemix classification for stroke rehabilitation in Australia. How well does AN-SNAP v2 explain variance in outcomes?" Australian Health Review 35, no. 1 (2011): 1. http://dx.doi.org/10.1071/ah09806.
Full textAbstract:
Objective. We sought the best predictors for length of stay, discharge destination and functional improvement for inpatients undergoing rehabilitation following a stroke and compared these predictors against AN-SNAP v2. Method. The Oxfordshire classification subgroup, sociodemographic data and functional data were collected for patients admitted between 1997 and 2007, with a diagnosis of recent stroke. The data were factor analysed using Principal Components Analysis for categorical data (CATPCA). Categorical regression analyses was performed to determine the best predictors of length of stay, discharge destination, and functional improvement. Results. A total of 1154 patients were included in the study. Principal components analysis indicated that the data were effectively unidimensional, with length of stay being the most important component. Regression analysis demonstrated that the best predictor was the admission motor FIM score, explaining 38.9% of variance for length of stay, 37.4%.of variance for functional improvement and 16% of variance for discharge destination. Conclusion. The best explanatory variable in our inpatient rehabilitation service is the admission motor FIM. AN- SNAP v2 classification is a less effective explanatory variable. This needs to be taken into account when using AN-SNAP v2 classification for clinical or funding purposes. What is known about the topic? AN-SNAP v2, a major classification tool for inpatient rehabilitation units has been described and used in a small number of published studies. The ability to predict variance by AN-SNAP v2 has not been previously described. What does this paper add? This paper indicates that AN-SNAP v2 is not a good predictor of outcomes in patients in medical rehabilitation units, challenging its utility as a classification tool. What are the implications for practitioners? Practitioners will have a broader understanding of the strengths and weaknesses of the AN-SNAP v2 classification.
6
VOZALIS, MANOLIS G., ANGELOS I. MARKOS, and KONSTANTINOS G. MARGARITIS. "AN OPTIMAL SCALING FRAMEWORK FOR COLLABORATIVE FILTERING RECOMMENDATION SYSTEMS." International Journal on Artificial Intelligence Tools 21, no. 06 (December 2012): 1250033. http://dx.doi.org/10.1142/s0218213012500339.
Full textAbstract:
Collaborative Filtering (CF) is a popular technique employed by Recommender Systems, a term used to describe intelligent methods that generate personalized recommendations. Some of the most efficient approaches to CF are based on latent factor models and nearest neighbor methods, and have received considerable attention in recent literature. Latent factor models can tackle some fundamental challenges of CF, such as data sparsity and scalability. In this work, we present an optimal scaling framework to address these problems using Categorical Principal Component Analysis (CatPCA) for the low-rank approximation of the user-item ratings matrix, followed by a neighborhood formation step. CatPCA is a versatile technique that utilizes an optimal scaling process where original data are transformed so that their overall variance is maximized. We considered both smooth and non-smooth transformations for the observed variables (items), such as numeric, (spline) ordinal, (spline) nominal and multiple nominal. The method was extended to handle missing data and incorporate differential weighting for items. Experiments were executed on three data sets of different sparsity and size, MovieLens 100k, 1M and Jester, aiming to evaluate the aforementioned options in terms of accuracy. A combined approach with a multiple nominal transformation and a "passive" missing data strategy clearly outperformed the other tested options for all three data sets. The results are comparable with those reported for single methods in the CF literature.
7
Pizarro Inostroza, María Gabriela, Vincenzo Landi, Francisco Javier Navas González, Jose Manuel León Jurado, Juan Vicente Delgado Bermejo, Javier Fernández Álvarez, and María del Amparo Martínez Martínez. "Integrating Casein Complex SNPs Additive, Dominance and Epistatic Effects on Genetic Parameters and Breeding Values Estimation for Murciano-Granadina Goat Milk Yield and Components." Genes 11, no. 3 (March 14, 2020): 309. http://dx.doi.org/10.3390/genes11030309.
Full textAbstract:
Assessing dominance and additive effects of casein complex single-nucleotide polymorphisms (SNPs) (αS1, αS2, β, and κ casein), and their epistatic relationships may maximize our knowledge on the genetic regulation of profitable traits. Contextually, new genomic selection perspectives may translate this higher efficiency into higher accuracies for milk yield and components’ genetic parameters and breeding values. A total of 2594 lactation records were collected from 159 Murciano-Granadina goats (2005–2018), genotyped for 48 casein loci-located SNPs. Bonferroni-corrected nonparametric tests, categorical principal component analysis (CATPCA), and nonlinear canonical correlations were performed to quantify additive, dominance, and interSNP epistatic effects and evaluate the outcomes of their inclusion in quantitative and qualitative milk production traits’ genetic models (yield, protein, fat, solids, and lactose contents and somatic cells count). Milk yield, lactose, and somatic cell count heritabilities increased considerably when the model including genetic effects was considered (0.46, 0.30, 0.43, respectively). Components standard prediction errors decreased, and accuracies and reliabilities increased when genetic effects were considered. Conclusively, including genetic effects and relationships among these heritable biomarkers may improve model efficiency, genetic parameters, and breeding values for milk yield and composition, optimizing selection practices profitability for components whose technological application may be especially relevant for the cheese-making dairy sector.
8
Ciucurel, Constantin, and Elena Ioana Iconaru. "An Epidemiological Study on the Prevalence of the Clinical Features of SARS-CoV-2 Infection in Romanian People." International Journal of Environmental Research and Public Health 17, no. 14 (July 14, 2020): 5082. http://dx.doi.org/10.3390/ijerph17145082.
Full textAbstract:
The aim of this study was to investigate the prevalence of the clinical features of the SARS-CoV-2 infection in Romanian population through a novel online survey. The survey included categorical socio-demographic and health-related variables. A total of 1830 participants were selected for statistical data processing (a response rate of 90.9%). We determined reasonable reliability of the survey section for clinical features of SARS-CoV-2 infection (Cronbach’s Alpha 0.671). Two meaningful dimensions were identified through CATPCA (Categorical Principal Component Analysis) for the survey’s items. We separated two significant clusters of items, each measuring a distinct factor: the sociodemographic characteristics linked to social distancing and the relevant clinical features of SARS-CoV-2 infection. Next, a two-step cluster analysis helped to classify the sample group taking into consideration the similarity of subjects. The clustering revealed a three-cluster solution, with significant differences between clusters and allowed the cluster detection of a group of individuals, possibly more affected by the infection with the SARS-CoV-2 virus. Through binomial logistic regression analysis, we identified a statistically significant prediction model for the presumptive diagnostic of some relevant clinical features of SARS-CoV-2 infection. Our study validated a cost-effective model for rapid assessment of the health status of subjects, adapted to the context of SARS-CoV-2 pandemic.
9
Dzurec, Laura, Aryn Karpinski, Monica Kennison, Randy Rair, and Shawn Fitzgerald. "Assessing Family-Like Dynamics in the Workplace as Possible Precursors of Workplace Bullying: Psychometric Analysis of a Modified Instrument." Journal of Nursing Measurement 27, no. 2 (August 1, 2019): 297–312. http://dx.doi.org/10.1891/1061-3749.27.2.297.
Full textAbstract:
Background and PurposeFamily-like dynamics in workplaces may serve as antecedents to workplace bullying. This study addressed the psychometric properties of an instrument modified to assess family-like dynamics in the workplace.DesignThe investigators used categorical principal components analysis (CATPCA) to investigate the psychometric properties of an instrument modified to measure coworker perceptions of family-like dynamics in the workplace.MethodsPrimarily White (95%) study participants (N = 273) completed a brief, demographic form and the modified Family Relationships Index (FRI) of the Family Assessment Scale (FES) (Moos & Moos, 1981, 1986). Demographic data were analyzed using descriptive statistics and perceptions of family-like dynamics in the workplace were analyzed using CATPCA.ResultsThe modified FRI served as a reasonable model for capturing coworker perceptions of family-like dynamics in the workplace.ConclusionsFurther research is indicated to determine the overall utility of the modified FRI and to ascertain whether family-like dynamics actually are precursors to workplace bullying victimization.
10
Hamidi, M. Luthfi, and Andrew C. Worthington. "Islamic banking sustainability: theory and evidence using a novel quadruple bottom line framework." International Journal of Bank Marketing 39, no. 5 (March 4, 2021): 751–67. http://dx.doi.org/10.1108/ijbm-06-2020-0345.
Full textAbstract:
PurposeThe study aims to extend the conventional triple bottom line (TBL) framework (prosperity, people and planet) to the quadruple bottom line (QBL) by newly adding a “prophet” dimension for Islamic banks seeking compliance with Islamic law in their pursuit of sustainability.Design/methodology/approachEmploy Chapra's corollaries of maqasid al-shari'ah (the goals of Islamic law) to develop constructs for a survey of 504 Islamic bank stakeholders from five Indonesian provinces to gather primary data to quantitatively verify the dimensions and items in the proposed QBL framework. Categorical principal component analysis (CATPCA) then identifies the sustainability of ten Islamic banks from ten countries as a trial application of the resulting QBL index.FindingsUsing the dimensions and items identified using CATPCA, the authors develop a QBL index to assess the sustainability of the ten Islamic banks. The findings suggest that half of the banks are sufficiently sustainable, with three being proactive (doing more than is required) and two being accommodative (doing all that is required). The remaining five banks are unsustainable, with two banks being defensive (doing the least that is required) and three being reactive (doing less than is required). Most of the banks perform relatively poorly according to the “planet” (38%) and “people” (41%) dimensions and perform better on the “prosperity” (53%) and “prophet” (63%) dimensions. Nonetheless, there is ample room for improvement across all dimensions of sustainability.Research limitations/implicationsThe generalizability of the findings is limited by the small-scale single-country survey used in the CATPCA part of the analysis. Only ten Islamic banks were included in the QBL scoring and ranking exercisesPractical implicationsIslamic banks can improve their sustainability by increasing green financing and reaching out to rural areas and disadvantaged populations. In countries with Islamic banking systems, regulators can support this through training, guidance and incentives.Originality/valuePioneering exploration of TBL from maqasid al-shari'ah perspective. First, we develop a QBL index to assess the sustainability of Islamic banks in line with actual stakeholder expectations.
Dissertations / Theses on the topic "Categorical Principal Component Analysis (CATPCA)"
1
Marques, Luís Miguel Valente. "E-government e participação política em Portugal." Master's thesis, Instituto Superior de Ciências Sociais e Políticas, 2016. http://hdl.handle.net/10400.5/11544.
Full textAbstract:
Dissertação de Mestrado em MPA - Administração PúblicaEsta dissertação pretende dar-nos um conhecimento mais fino sobre uma problemática ainda pouco desenvolvida no contexto da Ciência da Administração Pública, a relação entre o e-government e a participação política. Partindo da questão de partida “o e-government a nível local, em Portugal, contribui para uma maior participação política?”, realizou-se um trabalho de investigação quantitativa sobre a presença de funcionalidades de e-government destinadas a uma interação direta entre cidadãos e municípios. Pretendendo explorar as relações de associação entre e-government e participação política que decorrem do uso de ferramentas de e-participation, os conceitos de e-government, participação política e e-participation foram decompostos em dimensões, de acordo com a literatura de referência. As variáveis que constituem as dimensões foram escolhidas de acordo com o levantamento realizado sobre estudos acerca de local e-government e bases de dados nacionais, incidindo sobre os 308 municípios de Portugal, com dados referentes a 2011 e 2013. Estes foram analisados com recurso às técnicas estatísticas, Análise Categorial de Componentes Principais (CATPCA), Análise Fatorial e Análise de Clusters. Os resultados obtidos evidenciaram uma tendência de associação negativa entre a participação eleitoral e as dimensões e-information, capital humano e infraestruturas de telecomunicações. As ferramentas associadas à dimensão e-decision não se encontram significativamente desenvolvidas, neste contexto ainda nos encontramos num estado embrionário em direção ao New Public Service
This dissertation aims to give a finer understanding of a yet not very studied field in the context of Public Administration Science, the relation between e-government and political participation. Starting from the hypothesis “does local level e-government in Portugal contribute to a higher political participation?” a quantitative research was developed about the presence of e-government functionalities geared towards direct interaction between citizens and Municipalities. Aiming to explore the relation between e-government, political participation and e-participation derived from the use of e-participation applications, the concepts of e-government, political participation and e-participation were disaggregated into dimensions, based on current literature. The variables which compose the dimensions were chosen in accordance to the survey of studies on local e-government and national databases, encompassing the 308 Portuguese municipalities, using data from 2011 and 2013. The data was analyzed using the statistical techniques Categorical Principal Components Analysis (CATPCA), Factorial Analysis and Clusters Analysis. The results obtained uncovered a negative association tendency between electoral participation and the dimensions e-information, human capital and telecommunication infrastructures. The applications associated with the dimension e-decision are not significantly developed, in this context we are still in an embryonic stage of development towards New Public Service.
2
Vasquez, Modesto Cal. "Eficiência e produtividade no ensino superior público." Doctoral thesis, Instituto Superior de Ciências Sociais e Políticas, 2013. http://hdl.handle.net/10400.5/6183.
Full textAbstract:
tese de Doutoramento em Ciências Sociais na Especialidade de Administração PúblicaNesta investigação foram estudadas as eficiências e as produtividades das instituições portuguesas do ensino superior público: universidades e institutos politécnicos. Na análise consideraram-se oito períodos académicos: 2000-2001 a 2007-2008. Recorrendo à metodologia Data Envelopment Analysis (DEA), propôs-se uma abordagem que, até à presente data, não foi utilizada para o ensino superior: Analisar a eficiência e a produtividade tendo como base os conceitos de eficiência e super eficiência não radial. Foi também analisado o impacto na eficiência de factores não discricionários. As análises foram efectuadas tendo como suporte o modelo desenvolvido e programado. As conclusões mais relevantes foram: Globalmente, no período 2000-2008, as universidades e os institutos politécnicos melhoraram as suas eficiências relativas e produtividades; A assimetria regional tem impacto na eficiência.
In this research efficiency and productivity of the Portuguese public higher education institutions were studied: universities and polytechnics. The analysis covered eight academic periods: 2000-2001 to 2007-2008. Using Data Envelopment Analysis methodology (DEA), proposed an approach that, to date, has not been used for higher education: Analyze the efficiency and productivity based on the concept of non-radial efficiency and super-efficiency. It was also analysed the impact on efficiency of non-discretionary factors. The analysis were carried out with support of the model developed and programmed. The most relevant conclusions were: overall, in the 2000-2008 period, universities and polytechnics have improved their relative efficiencies and productivity; regional development has impact on efficiency.
3
村上, 隆., and Takashi Murakami. "カテゴリカル・データの非計量的主成分分析の応用." 名古屋大学教育学部, 1997. http://hdl.handle.net/2237/2866.
Full text
4
Gao, Huanhuan. "Categorical structural optimization : methods and applications." Thesis, Compiègne, 2019. http://www.theses.fr/2019COMP2471/document.
Full textAbstract:
La thèse se concentre sur une recherche méthodologique sur l'optimisation structurelle catégorielle au moyen d'un apprentissage multiple. Dans cette thèse, les variables catégorielles non ordinales sont traitées comme des variables discrètes multidimensionnelles. Afin de réduire la dimensionnalité, les nombreuses techniques d'apprentissage sont introduites pour trouver la dimensionnalité intrinsèque et mapper l'espace de conception d'origine sur un espace d'ordre réduit. Les mécanismes des techniques d'apprentissage à la fois linéaires et non linéaires sont d'abord étudiés. Ensuite, des exemples numériques sont testés pour comparer les performances de nombreuses techniques d’apprentissage. Sur la base de la représentation d'ordre réduit obtenue par Isomap, les opérateurs de mutation et de croisement évolutifs basés sur les graphes sont proposés pour traiter des problèmes d'optimisation structurelle catégoriels, notamment la conception du dôme, du cadre rigide de six étages et des structures en forme de dame. Ensuite, la méthode de recherche continue consistant à déplacer des asymptotes est exécutée et fournit une solution compétitive, mais inadmissible, en quelques rares itérations. Ensuite, lors de la deuxième étape, une stratégie de recherche discrète est proposée pour rechercher de meilleures solutions basées sur la recherche de voisins. Afin de traiter le cas dans lequel les instances de conception catégorielles sont réparties sur plusieurs variétés, nous proposons une méthode d'apprentissage des variétés k-variétés basée sur l'analyse en composantes principales pondéréesThe thesis concentrates on a methodological research on categorical structural optimizationby means of manifold learning. The main difficulty of handling the categorical optimization problems lies in the description of the categorical variables: they are presented in a category and do not have any orders. Thus the treatment of the design space is a key issue. In this thesis, the non-ordinal categorical variables are treated as multi-dimensional discrete variables, thus the dimensionality of corresponding design space becomes high. In order to reduce the dimensionality, the manifold learning techniques are introduced to find the intrinsic dimensionality and map the original design space to a reduced-order space. The mechanisms of both linear and non-linear manifold learning techniques are firstly studied. Then numerical examples are tested to compare the performance of manifold learning techniques mentioned above. It is found that the PCA and MDS can only deal with linear or globally approximately linear cases. Isomap preserves the geodesic distances for non-linear manifold however, its time consuming is the most. LLE preserves the neighbour weights and can yield good results in a short time. KPCA works like a non-linear classifier and we proves why it cannot preserve distances or angles in some cases. Based on the reduced-order representation obtained by Isomap, the graph-based evolutionary crossover and mutation operators are proposed to deal with categorical structural optimization problems, including the design of dome, six-story rigid frame and dame-like structures. The results show that the proposed graph-based evolutionary approach constructed on the reduced-order space performs more efficiently than traditional methods including simplex approach or evolutionary approach without reduced-order space. In chapter 5, the LLE is applied to reduce the data dimensionality and a polynomial interpolation helps to construct the responding surface from lower dimensional representation to original data. Then the continuous search method of moving asymptotes is executed and yields a competitively good but inadmissible solution within only a few of iteration numbers. Then in the second stage, a discrete search strategy is proposed to find out better solutions based on a neighbour search. The ten-bar truss and dome structural design problems are tested to show the validity of the method. In the end, this method is compared to the Simulated Annealing algorithm and Covariance Matrix Adaptation Evolutionary Strategy, showing its better optimization efficiency. In chapter 6, in order to deal with the case in which the categorical design instances are distributed on several manifolds, we propose a k-manifolds learning method based on the Weighted Principal Component Analysis. And the obtained manifolds are integrated in the lower dimensional design space. Then the method introduced in chapter 4 is applied to solve the ten-bar truss, the dome and the dame-like structural design problems
5
Audigier, Vincent. "Imputation multiple par analyse factorielle : Une nouvelle méthodologie pour traiter les données manquantes." Thesis, Rennes, Agrocampus Ouest, 2015. http://www.theses.fr/2015NSARG015/document.
Full textAbstract:
Cette thèse est centrée sur le développement de nouvelles méthodes d'imputation multiples, basées sur des techniques d'analyse factorielle. L'étude des méthodes factorielles, ici en tant que méthodes d'imputation, offre de grandes perspectives en termes de diversité du type de données imputées d'une part, et en termes de dimensions de jeux de données imputés d'autre part. Leur propriété de réduction de la dimension limite en effet le nombre de paramètres estimés.Dans un premier temps, une méthode d'imputation simple par analyse factorielle de données mixtes est détaillée. Ses propriétés sont étudiées, en particulier sa capacité à gérer la diversité des liaisons mises en jeu et à prendre en compte les modalités rares. Sa qualité de prédiction est éprouvée en la comparant à l'imputation par forêts aléatoires.Ensuite, une méthode d'imputation multiple pour des données quantitatives basée sur une approche Bayésienne du modèle d'analyse en composantes principales est proposée. Elle permet d'inférer en présence de données manquantes y compris quand le nombre d'individus est petit devant le nombre de variables, ou quand les corrélations entre variables sont fortes.Enfin, une méthode d'imputation multiple pour des données qualitatives par analyse des correspondances multiples (ACM) est proposée. La variabilité de prédiction des données manquantes est reflétée via un bootstrap non-paramétrique. L'imputation multiple par ACM offre une réponse au problème de l'explosion combinatoire limitant les méthodes concurrentes dès lors que le nombre de variables ou de modalités est élevThis thesis proposes new multiple imputation methods that are based on principal component methods, which were initially used for exploratory analysis and visualisation of continuous, categorical and mixed multidimensional data. The study of principal component methods for imputation, never previously attempted, offers the possibility to deal with many types and sizes of data. This is because the number of estimated parameters is limited due to dimensionality reduction.First, we describe a single imputation method based on factor analysis of mixed data. We study its properties and focus on its ability to handle complex relationships between variables, as well as infrequent categories. Its high prediction quality is highlighted with respect to the state-of-the-art single imputation method based on random forests.Next, a multiple imputation method for continuous data using principal component analysis (PCA) is presented. This is based on a Bayesian treatment of the PCA model. Unlike standard methods based on Gaussian models, it can still be used when the number of variables is larger than the number of individuals and when correlations between variables are strong.Finally, a multiple imputation method for categorical data using multiple correspondence analysis (MCA) is proposed. The variability of prediction of missing values is introduced via a non-parametric bootstrap approach. This helps to tackle the combinatorial issues which arise from the large number of categories and variables. We show that multiple imputation using MCA outperforms the best current methods
6
Jha, Rajesh. "Combined Computational-Experimental Design of High-Temperature, High-Intensity Permanent Magnetic Alloys with Minimal Addition of Rare-Earth Elements." FIU Digital Commons, 2016. http://digitalcommons.fiu.edu/etd/2621.
Full textAbstract:
AlNiCo magnets are known for high-temperature stability and superior corrosion resistance and have been widely used for various applications. Reported magnetic energy density ((BH) max) for these magnets is around 10 MGOe. Theoretical calculations show that ((BH) max) of 20 MGOe is achievable which will be helpful in covering the gap between AlNiCo and Rare-Earth Elements (REE) based magnets. An extended family of AlNiCo alloys was studied in this dissertation that consists of eight elements, and hence it is important to determine composition-property relationship between each of the alloying elements and their influence on the bulk properties.
In the present research, we proposed a novel approach to efficiently use a set of computational tools based on several concepts of artificial intelligence to address a complex problem of design and optimization of high temperature REE-free magnetic alloys. A multi-dimensional random number generation algorithm was used to generate the initial set of chemical concentrations. These alloys were then examined for phase equilibria and associated magnetic properties as a screening tool to form the initial set of alloy. These alloys were manufactured and tested for desired properties. These properties were fitted with a set of multi-dimensional response surfaces and the most accurate meta-models were chosen for prediction. These properties were simultaneously extremized by utilizing a set of multi-objective optimization algorithm. This provided a set of concentrations of each of the alloying elements for optimized properties. A few of the best predicted Pareto-optimal alloy compositions were then manufactured and tested to evaluate the predicted properties. These alloys were then added to the existing data set and used to improve the accuracy of meta-models. The multi-objective optimizer then used the new meta-models to find a new set of improved Pareto-optimized chemical concentrations. This design cycle was repeated twelve times in this work. Several of these Pareto-optimized alloys outperformed most of the candidate alloys on most of the objectives. Unsupervised learning methods such as Principal Component Analysis (PCA) and Heirarchical Cluster Analysis (HCA) were used to discover various patterns within the dataset. This proves the efficacy of the combined meta-modeling and experimental approach in design optimization of magnetic alloys.
7
Costa, Anabela da Silva Jorge da. "Análise fatorial exploratória e análise de componentes principais aplicadas a um questionário para avaliação da qualidade escolar." Master's thesis, 2012. http://hdl.handle.net/10400.2/2173.
Full textAbstract:
Dissertação de Mestrado em Estatística, Matemática e Computação apresentada à Universidade AbertaNeste trabalho apresenta-se uma aplicação de técnicas de Estatística Multivariada. Comparam-se os resultados de duas técnicas, a Análise de Componentes Principais (ACP) e a Análise de Componentes Principais para Dados Categóricos (CATPCA), utilizando para tal dois conjuntos de dados de respostas a dois questionários. Pretende-se verificar se a violação de alguns pressupostos da ACP relativamente ao tipo de escala da variável e à forma dos dados (simétrica ou não simétrica) influencia a qualidade e o número das componentes extraídas e se difere significativamente da solução da CATPCA. Para esta análise comparativa são utilizados dados recolhidos através de dois questionários de autoavaliação interna de uma escola secundária do interior do distrito de Coimbra. Um dos questionários foi administrado a uma amostra de 161 alunos do 10º ao 12ºanos, e o outro foi aplicado a uma amostra de 97 encarregados de educação, sendo assim representados dois tamanhos amostrais. Os itens de cada questionário estão agrupados em 6 ou 8 grupos respetivamente, no caso do questionário dos alunos e no questionário dos encarregados de educação. Usando a técnica de estatística multivariada Análise Fatorial Exploratória (AFE) exploram-se os fatores subjacentes à estrutura dos questionários e verifica-se se estes refletem os que tinham sido definidos previamente pela equipa da avaliação escolar. Apresenta-se ainda uma caracterização teórica das 3 técnicas de estatística multivariada aplicadas neste trabalho e, finalmente, procura-se dar um contributo para a melhoria do instrumento utilizado na avaliação da qualidade escolar.
In this work, is presented an application of exploratory multivariate statistical techniques. It has been designed to compare the results of two techniques, the Principal Component Analysis (PCA) and Principal Component Analysis for Categorical variables. The two techniques are compared through two sets of data taken from an application of two questionnaires. The purpose of this comparison is to verify whether the violation of some assumptions of the PCA, related with the type of variables, the form of the data distribution (symmetric or nonsymmetrical), influences significantly the solution obtained, the quality and the quantity of the extracted components. To carry out the comparative analysis, the data used was collected through two auto-administered questionnaires of internal self-assessment of a secondary school in the district of Coimbra. One of the questionnaires was administered to a sample of 161 students from 10th to 12th grade, and the other survey was applied to a sample of 97 educational responsible. The items of the student’s questionnaire are theoretically grouped into 6 groups while the items of the parent’s questionnaire are grouped into 8 groups. This theoretical structure was explored using the multivariate technique Exploratory Factor Analysis (EFA), in order to determine how this structure is reflected in the data collected. It also presents a more developed theoretical characterization of three multivariate statistical techniques applied in this study and, finally, some suggestions are made to contribute to improving the tool used in the evaluation of the quality of school services.
Book chapters on the topic "Categorical Principal Component Analysis (CATPCA)"
1
Casacci, Sara. "Categorical Principal Component Analysis." In Encyclopedia of Quality of Life and Well-Being Research, 1–2. Cham: Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-319-69909-7_104643-1.
Full text
2
Diday, Edwin. "Principal Component Analysis for Categorical Histogram Data: Some Open Directions of Research." In Classification and Multivariate Analysis for Complex Data Structures, 3–15. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010. http://dx.doi.org/10.1007/978-3-642-13312-1_1.
Full text
3
Murakami, Takashi. "Orthonormal Principal Component Analysis for Categorical Data as a Transformation of Multiple Correspondence Analysis." In Advanced Studies in Behaviormetrics and Data Science, 211–31. Singapore: Springer Singapore, 2020. http://dx.doi.org/10.1007/978-981-15-2700-5_13.
Full text
4
Markos, Angelos I., Manolis G. Vozalis, and Konstantinos G. Margaritis. "An Optimal Scaling Approach to Collaborative Filtering Using Categorical Principal Component Analysis and Neighborhood Formation." In IFIP Advances in Information and Communication Technology, 22–29. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010. http://dx.doi.org/10.1007/978-3-642-16239-8_6.
Full text
5
Tawfik, Amal, and Stephan Davidshofer. "Multiple Correspondence Analysis and Geometric Data Analysis." In Research Methods in the Social Sciences: An A-Z of key concepts, 177–84. Oxford University Press, 2021. http://dx.doi.org/10.1093/hepl/9780198850298.003.0043.
Full textAbstract:
This chapter focuses on multiple correspondence analysis (MCA), which is a factor analysis statistical method used to analyse relations between a large set of categorical variables. Developed by Jean-Paul Benzécri in the early 1970s, MCA is one of the principal methods of geometric data analysis (GDA). Three different statistical methods can be identified as GDA: correspondence analysis (CA), which enables the cross-tabulation of two categorical variables; MCA for the analysis of a matrix of individuals and categorical variables; and principal component analysis (PCA), which uses numerical variables. In GDA, data is represented as a cloud of points to allow statistical interpretations. Although MCA is a relational method, it differs from social network analysis (SNA) as it focuses on the objective relations that characterize actors or groups, rather than the effective relations.
Conference papers on the topic "Categorical Principal Component Analysis (CATPCA)"
1
Rotimi, Oluwatosin John, David Nnaemeka Ukwu, Wang Zhenli, Yao Liang, Anthony A. Ameloko, Temitope Fred Ogunkunle, Kehinde David Oyeyemi, Kouamelan Serge Kouamelan, Ufouma Frank Asaboro, and Ifunanya Sylvia Onuigbo. "Sequential Prediction of Drilling Fluid Loss Using Support Vector Machine and Decision Tree Methods." In SPE Nigeria Annual International Conference and Exhibition. SPE, 2021. http://dx.doi.org/10.2118/207185-ms.
Full textAbstract:
Abstract Machine learning methods have been applied to predict depths of fluid loss in hydrocarbon exploration.During drilling, lost circulation can be described as the unpleasant loss of all or part of drilling mud or fluid into the immediate formations or affected formation by excessive hydrostatic pressure, sufficient to fracture the formation or expand existing fractures encountered during the drilling process. In this study, we deployed Python codes of Support Vector Machine (SVM) and Decision Tree (DT) methodsto categorical data obtained from drilling operations in a producing field to predict lost circulation occurrence. The modelsleveraged the capability of both SVM and DT to achieve binary classification by adopting flow-out percentage of less than 70 percent as the points of lost circulation. That is, < 70% is represented as Loss and > 70% represented asNo Loss. Prediction models were applied to 10 input variables preprocessed with principal component analysis (PCA) to reduce dimensionality and focus on essential variables. The preprocessed SVM model gave an improved result while preprocessing does not affect DT models. Overall, DT models predicted accurate fluid losszones and can be scaled up to field operations with options ofcontinuous sampled variables.
2
Sangelkar, Shraddha, and Daniel A. McAdams. "Automated Graph Grammar Generation for Engineering Design With Frequent Pattern Mining." In ASME 2017 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. American Society of Mechanical Engineers, 2017. http://dx.doi.org/10.1115/detc2017-67520.
Full textAbstract:
Graph grammars, a technique for formulating new graphs based on a set of rules, is a very powerful tool for computational design synthesis. It is particularly suitable for discrete categorical data where principal component analysis is generally not applicable. Furthermore, this technique utilizes three different programs in conjunction with a design repository, which is opposed to traditional methods that require experts to empirically derive graph grammars. This technique can be separated into three steps. These steps are the creation of the input, graph data mining, and interpretation of the output with the intention of these steps being to automate or assist an expert with the process of extracting engineering graph grammars. Graph grammars that can then serve as guidelines during concept generation. The results of this paper show that this technique is very applicable to computational design synthesis by testing only a small number of products and still producing tangible results that coincide with empirically derived graphs. Fifty electromechanical products from the design repository are used in this study. When comparing, the machine generated grammar rules with expert derived grammar rules, it can be seen that only 14% cannot be developed, 58% cannot be mined with the current setup and 28% were mined with the current set up. However, it is important to keep in mind a few considerations. Specifically, the technique does not replace the expert. Instead, the technique acts as more of an aid than a replacement. Also, while this technique has great potential in regards to computational design synthesis, it is limited to the products in the design repository and the current implementation of the aforementioned programs. Despite these minor considerations, this work proposes application of graph data mining to derive engineering grammars.