The DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.
An entropy criterion for assessing the number of clusters in a mixture model
Open accessOverview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Entropy above .80 also indicates a good-fitting model [40, 41].”
From Symptom experiences in post-treatment cancer survivors: associations with acceptance and commitment therapy constructs. (2021) · cites this paper as [40]
In the citing paper
“These values can vary between 0 and 1 with higher values representing greater accuracy of classification and values above 0.80 evidencing classification with minimal uncertainty.”
From Coping profiles and their association with psychological functioning: A latent profile analysis of coping strategies during the COVID-19 pandemic (2022) · cites this paper as Celeux & Soromenho, 1996
In the citing paper
“the entropy values of greater than .80 showed a 220 clear delineation of classes in the 'quadratic' GMM models with two or three classes 221 (Celeux and Soromenho, 1996).”
From Time in treatment: Examining mental illness trajectories across inpatient psychiatric treatment (2020) · cites this paper as Celeux and Soromenho, 1996
Citations
Cites
A new look at the statistical model identification
Akaike, H. (1974) · IEEE Transactions on Automatic Control · doi:10.1109/tac.1974.1100705 · Cited by 51,879
Maximum Likelihood from Incomplete Data Via the EM Algorithm
Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1977.tb01600.x · Cited by 50,444
Estimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033
(1967) · Cited by 23,719
Pattern Recognition with Fuzzy Objective Function Algorithms
Bezdek, J. C. (1981) · doi:10.1007/978-1-4757-0450-1 · Cited by 9,881
Statistical Analysis of Finite Mixture Distributions.
Lindsay, B. G., Titterington, D. M., Smith, A. F. M., & Makov, U. E. (1987) · Journal of the American Statistical Association · doi:10.2307/2289504 · Cited by 2,905
Model-Based Gaussian and Non-Gaussian Clustering
Banfield, J. D., & Raftery, A. E. (1993) · Biometrics · doi:10.2307/2532201 · Cited by 2,395 · cited 2×
“…have considered the problem of choosing the number of clusters, in a partitioning setting, within the context of multivariate mixture models (see Bock 1985; Celeux 1986; Bozdogan 1993; Banfield and Raftery 1993; McLachlan and Basford 1988; Windham and Cutler 1992; Bryant 1993, among others).”
MethodMixture models : inference and applications to clustering
McLachlan, G. J., & Basford, K. E. (1988) · M. Dekker eBooks · Cited by 2,094
Statistical Analysis of Finite Mixture Distributions.
Newton, J., Titterington, D. M., Smith, A. F. M., & Makov, U. E. (1986) · Biometrics · doi:10.2307/2531224 · Cited by 1,986
Proceedings of the 6th Berkeley Symposium on Mathematical Statistics and Probability
Cam, L. M. L., Neyman, J., & Scott, E. L. (1972) · Medical Entomology and Zoology · Cited by 1,679
Research works printed but not linked: 1
- BRYANT, P. G., and WILLIAMSON, J. A. (1986), ?Maximum Likelihood and Classification: a Comparison of Three Approaches,? inClassification as a tool of research, Eds., W. Gaul and M. Schader, North-Holland, pp. 33?45.
Cited by
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705510701575396 · Cited by 11,415
“In other studies, the AIC has been shown to overestimate the correct number of components in finite mixture models (Celeux & Soromenho, 1996; Soromenho, 1993), whereas the BIC has been reported to perform well (Roeder & Wasserman, 1997).”
Testing the number of components in a normal mixture
Lo, Y. (2001) · Biometrika · doi:10.1093/biomet/88.3.767 · Cited by 4,640
Model-Based Clustering, Discriminant Analysis, and Density Estimation
Fraley, C., & Raftery, A. E. (2002) · Journal of the American Statistical Association · doi:10.1198/016214502760047131 · Cited by 4,257
“Cheeseman and Stutz (1995) and Chickering and Heckerman (1997) used a different approximation to the integrated likelihood; other approaches include an informational complexity criterion called ICOMP (Bozdogan 1994), an entropy criterion called NEC (Celeux and Soromenho 1996; Biernacki, Celeux, and Govaert 1999), the…”
MethodLatent class analysis: a guide to best practice
Weller, B. E., Bowen, N. K., & Faubert, S. J. (2020) · Journal of Black Psychology · doi:10.1177/0095798420930932 · Cited by 2,223
“In general, an entropy value close to 1 is ideal (Celeux & Soromenho, 1996) and above .8 is acceptable.”
BackgroundUnsupervised learning of finite mixture models
Figueiredo, M. A. T., & Jain, A. K. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.990138 · Cited by 2,125
Spurk, D., Hirschi, A., Wang, M., Valero, D., & Kauffeld, S. (2020) · Journal of Vocational Behavior · doi:10.1016/j.jvb.2020.103445 · Cited by 2,119
Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis
Tein, J.-Y., Coxe, S., & Cham, H. (2013) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2013.824781 · Cited by 1,959
Assessing a mixture model for clustering with the integrated completed likelihood
Biernacki, C., Celeux, G., & Govaert, G. (2000) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.865189 · Cited by 1,535
“The entropy EOm;KU (see [ 9 ]) is a measure of the ability of the”
BackgroundMethodInfluentialSelf-citationK‐means clustering: A half‐century synthesis
Steinley, D. (2006) · British Journal of Mathematical and Statistical Psychology · doi:10.1348/000711005x48266 · Cited by 1,120
“…to overestimate the number of parameters, and it has been suggested that a penalized measure may be more appropriate, creating the Bayesian information criterion (BIC) (Kass & Wasserman, 1995; Schwarz, 1978), while some related methods based on entropy are provided in Celeux and Soromenho (1996).”
MethodPROC LCA: A SAS Procedure for Latent Class Analysis
Lanza, S. T., Collins, L. M., Lemmon, D. R., & Schafer, J. L. (2007) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705510701575602 · Cited by 1,070
“The average posterior probabilities can then be used to calculate the odds of correct classification for each class (Nagin, 2005) and the overall entropy (Celeux & Soromenho, 1996).”
MethodBauer, D. J., & Curran, P. J. (2003) · Psychological Methods · doi:10.1037/1082-989x.8.3.338 · Cited by 1,059
“These statistics include the normalized entropy criterion (NEC; Biernacki & Govaert, 1999; Celeux & Soromenho, 1996) and the classification likelihood criterion (CLC; Biernacki & Govaert, 1997).”
BackgroundResultInfluentialFerguson, S. L., G. Moore, E. W., & Hull, D. M. (2020) · International Journal of Behavioral Development · doi:10.1177/0165025419881721 · Cited by 911
“SABIC has been supported as the most accurate information criteria index in simulation studies, particularly with smaller samples and low class separation (Kim, 2014; Morgan, 2015): SABICðKÞ ¼ 2LðKÞ þ vðKÞlnn ðnþ 2Þ=24 ð7Þ Finally, AIC is an inconsistent model fit measure as it does not have strong parsimony constrain…”
BackgroundMethodInfluentialThe GRoLTS-Checklist: Guidelines for Reporting on Latent Trajectory Studies
van de Schoot, R., Sijbrandij, M., Winter, S. D., Depaoli, S., & Vermunt, J. K. (2017) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2016.1247646 · Cited by 721
“The higher the entropy index, the fewer the classi fi cation errors, and the less bias in the prediction of class membership (Celeux & Soromenho, 1996).”
Latent burnout profiles: A new approach to understanding the burnout experience
Leiter, M. P., & Maslach, C. (2016) · Burnout Research · doi:10.1016/j.burn.2016.09.001 · Cited by 655
“Models with higher entropy and latent class probabilities are generally regarded as better models (Celeux & Soromenho, 1996).”
BackgroundKrueger, R. F., Markon, K. E., Patrick, C. J., & Iacono, W. G. (2005) · Journal of Abnormal Psychology · doi:10.1037/0021-843x.114.4.537 · Cited by 566
“BIC is enjoying widespread use and has been examined extensively in the context of latent variable modeling (e.g., Celeux & Soromenho, 1996; Fishler, Grosmann, & Messer, 2002; Lin & Dayton, 1997; Yang, 1998) and elsewhere (e.g., Zhang, 1993).”
Bauer, D. J., & Curran, P. J. (2004) · Psychological Methods · doi:10.1037/1082-989x.9.1.3 · Cited by 536
“…to favor models in which classes are distinguished by well-separated mean vectors, as it is typically in these cases that one could assign cases to classes with the most precision (Biernacki, Celeux, & Govaert, 1999, 2000; Celeux & Soromenho, 1996; Ramaswamy, DeSarbo, Reibstein, & Robinson, 1993).”
BackgroundChakravarty, A., Grewal, R., & Sambamurthy, V. (2013) · Information Systems Research · doi:10.1287/isre.2013.0500 · Cited by 502
“The entropy of separation of the two-regime solution of 0.85 indicated a good separation of latent regimes (Celeux and Soromenho 1996).”
BackgroundInfluentialHenson, J. M., Reise, S. P., & Kim, K. H. (2007) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705510709336744 · Cited by 429
“Finally, to evaluate Celeux and Soromenho’s (1996) proposed methodology for comparing the NEC statistics between a k-component model and a one-component model, constrained mixture models were estimated as well.”
BackgroundMethodInfluentialRisk and Rationality: Uncovering Heterogeneity in Probability Distortion
(2010) · Econometrica · doi:10.3982/ecta7139 · Cited by 384
A Latent Transition Mixture Model Using the Three-Step Specification
Nylund-Gibson, K., Grimm, R., Quirk, M., & Furlong, M. (2014) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2014.915375 · Cited by 372
Comparison of childhood wheezing phenotypes in 2 birth cohorts: ALSPAC and PIAMA
Savenije, O. E., Granell, R., Caudri, D., Koppelman, G. H., Smit, H. A., Wijga, A., de Jongste, J. C., Brunekreef, B., Sterne, J. A., Postma, D. S., et al. (2011) · Journal of Allergy and Clinical Immunology · doi:10.1016/j.jaci.2011.02.002 · Cited by 351
Modeling Unobserved Heterogeneity Using Latent Profile Analysis: A Monte Carlo Simulation
Peugh, J., & Fan, X. (2013) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2013.824780 · Cited by 341
“Two entropy-based enumeration indexes that reward LPA models producing well-defined profiles have been proposed (e.g, Bauer & Curran, 2003, 2004): the normalized entropy criterion (NEC; Celeux & Soromenho, 1996), and the classification likelihood criterion (CLC; Biernacki & Govaert, 1997).”
BackgroundChildren’s Classroom Engagement and School Readiness Gains in Prekindergarten
Chien, N. C., Howes, C., Burchinal, M., Pianta, R. C., Ritchie, S., Bryant, D. M., Clifford, R. M., Early, D. M., & Barbarin, O. A. (2010) · Child Development · doi:10.1111/j.1467-8624.2010.01490.x · Cited by 334
Study engagement and burnout profiles among Finnish higher education students.
Salmela-Aro, K., & Read, S. (2017) · Burnout Research · doi:10.1016/j.burn.2017.11.001 · Cited by 319
Job security and job protection
Clark, A., & Postel-Vinay, F. (2009) · Oxford Economic Papers · doi:10.1093/oep/gpn017 · Cited by 312
“φ̂i = ( φ̂ ppriv i , φ̂ ppub i , φ̂ temp i )′ , (10)”
Sarstedt, M., Becker, J.-M., Ringle, C. M., & Schwaiger, M. (2011) · Schmalenbach Business Review · doi:10.1007/bf03396886 · Cited by 304
Coping with boredom in school: An experience sampling perspective
Nett, U. E., Goetz, T., & Hall, N. C. (2011) · Contemporary Educational Psychology · doi:10.1016/j.cedpsych.2010.10.003 · Cited by 303
van der Nest, G., Lima Passos, V., Candel, M. J. J. M., & van Breukelen, G. J. P. (2020) · Advances in Life Course Research · doi:10.1016/j.alcr.2019.100323 · Cited by 287
Hennig, C., & Liao, T. F. (2013) · Journal of the Royal Statistical Society Series C: Applied Statistics · doi:10.1111/j.1467-9876.2012.01066.x · Cited by 278
Successful Aging: Early Influences and Contemporary Characteristics
Pruchno, R. A., Wilson-Genderson, M., Rose, M., & Cartwright, F. (2010) · The Gerontologist · doi:10.1093/geront/gnq041 · Cited by 271
“…was in empirically identifying two-group solutions for objective and subjective success that maximize the distinction between the successful and unsuccessful , we present the entropy statistic (Celeux & Soromenho, 1996) and statistics regarding the percentage of sample correctly classified.”
BackgroundAn Approach for Determining the Number of Clusters in a Model-Based Cluster Analysis
Akogul, S., & Erisoglu, M. (2017) · Entropy · doi:10.3390/e19090452 · Cited by 269
Heterogeneity of postpartum depression: a latent class analysis
(2015) · The Lancet Psychiatry · doi:10.1016/s2215-0366(14)00055-8 · Cited by 254
On the number of components in a Gaussian mixture model
McLachlan, G. J., & Rathnayake, S. (2014) · WIREs Data Mining and Knowledge Discovery · doi:10.1002/widm.1135 · Cited by 249
“Celeux and Soromenho 63 proposed using the estimated entropy EN ( ̂𝜏 (after normalization) as a criterion in its own right for choosing the number of clusters.”
BackgroundEliminating Bias in Classify-Analyze Approaches for Latent Class Analysis
Bray, B. C., Lanza, S. T., & Tan, X. (2015) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2014.935265 · Cited by 255
“Other methods for quantifying the extent of classification error (i.e., classification uncertainty) have been proposed in the context of assessing the overall quality of a latent class model (Celeux & Soromenho, 1996; Goodman, 2007; Vermunt & Magidson, 2002).”
MethodA Comparison of Segment Retention Criteria for Finite Mixture Logit Models
Andrews, R. L., & Currim, I. S. (2003) · Journal of Marketing Research · doi:10.1509/jmkr.40.2.235.19225 · Cited by 253
Model-Based Clustering and Classification for Data Science: With Applications in R
Bouveyron, C., Celeux, G., Murphy, T. B., & Raftery, A. E. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108644181 · Cited by 230
Self-citationDi Gessa, G., Glaser, K., & Tinker, A. (2016) · Social Science & Medicine · doi:10.1016/j.socscimed.2016.01.041 · Cited by 230
“…who provide 15 h of care per week or more are more likely to experience health declines and to report depressive symptoms in comparison to thosewho provide lower levels of grandchild care or no grandchild care (Hughes et al., 2007; Blustein et al., 2004; Minkler et al., 1997; Chen and Liu, 2012).”
InfluentialRossi, R. A., & Ahmed, N. K. (2015) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2014.2349913 · Cited by 212
“There has been a substantial amount of research on selecting the number of clusters automatically [156], [157], [158], [159], [160].”
BackgroundMixed Mode Latent Class Analysis: An Examination of Fit Index Performance for Classification
Morgan, G. B. (2015) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2014.935751 · Cited by 209
“Despite being a very commonly used indicator of class enumeration, AIC tends to overestimate the correct number of classes in mixture models (Celeux & Soromenho, 1996; Soromenho, 1993).”
MethodWhat to do when feeling bored?
Nett, U. E., Goetz, T., & Daniels, L. M. (2010) · Learning and Individual Differences · doi:10.1016/j.lindif.2010.09.004 · Cited by 202
“The value is bounded from [0; 1] with values closer to 1 indicating greater classification certainty (Celeux & Soromenho, 1996).”
BackgroundLougheed, J. P., & Hollenstein, T. (2012) · Social Development · doi:10.1111/j.1467-9507.2012.00663.x · Cited by 200
Liang, Y., & Lee, S. A. (2017) · International Journal of Social Robotics · doi:10.1007/s12369-017-0401-3 · Cited by 197
“0indicatedcleardelineationofclasses [2].”
Dissolution point and isolation robustness: Robustness criteria for general cluster analysis methods
Hennig, C. (2008) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2007.07.002 · Cited by 186
“It is interesting if this holds for other methods to estimate k (see, e.g., Section 3.5 of Gordon 1999, Celeux and Soromenho 1996, McLachlan 1987).”
MethodInfluentialIlli, J., Miaskowski, C., Cooper, B., Levine, J. D., Dunn, L., West, C., Dodd, M., Dhruva, A., Paul, S. M., Baggott, C., et al. (2012) · Cytokine · doi:10.1016/j.cyto.2012.02.015 · Cited by 185
“In addition, better-fitting models should produce higher entropy values [71].”
BackgroundNájera Catalán, H. E. (2019) · Social Indicators Research · doi:10.1007/s11205-018-1950-z · Cited by 183
“Values closer to 1, preferably E ≥ 0.8 are associated with very low class-membership probabilities ( < 5% ) and therefore with good classification in the literature (Celeux and Soromenho 1996).”
BackgroundInfluentialModel-based cluster and discriminant analysis with the MIXMOD software
Biernacki, C., Celeux, G., Govaert, G., & Langrognet, F. (2006) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2005.12.015 · Cited by 178
“The NEC (Normalized Entropy Criterion) criterion proposed by Celeux and Soromenho (1996) uses a similar entropy term EK = − ∑n i=m+1 ∑K k=1 tik ln(tik), but this criterion is intended to be used principally in determining the number of mixture components K, rather than the model parameterization M…”
BackgroundMethodSelf-citationStability of ARDS subphenotypes over time in two randomised controlled trials
Delucchi, K., Famous, K. R., Ware, L. B., Parsons, P. E., Thompson, B. T., & Calfee, C. S. (2018) · Thorax · doi:10.1136/thoraxjnl-2017-211090 · Cited by 176
An improvement of the NEC criterion for assessing the number of clusters in a mixture model
Biernacki, C., Celeux, G., & Govaert, G. (1999) · Pattern Recognition Letters · doi:10.1016/s0167-8655(98)00144-5 · Cited by 175
“As stressed in (Celeux and Soromenho, 1996) and shown through numerical experiments in (Biernacki and Govaert, 1997) and (Celeux and Soromenho, 1996), the NEC criterion is designed to choose the mixture model providing the greatest evidence for partitioning data.”
BackgroundMethodSelf-citationPanel Attrition: Separating Stayers, Fast Attriters, Gradual Attriters, and Lurkers
Lugtig, P. (2014) · Sociological Methods & Research · doi:10.1177/0049124113520305 · Cited by 174
“Apart from the values of BIC and the LMRT and BLRT, we chose to also rely on values of the entropy (Celeux and Soromenho 1996) as a criterion for the classification quality, and the substantive results of the best-performing models.”
MethodRisk and Rationality: Uncovering Heterogeneity in Probability Distortion
Bruhin, A., Fehr-Duda, H., & Epper, T. F. (2009) · SSRN Electronic Journal · doi:10.2139/ssrn.1415975 · Cited by 172
“Unfortunately, the AIC is order inconsistent, that is, the probability that it is minimized at the true model size does not approach unity with increasing sample size, and it tends to overfit models (Atkinson (1981), Geweke and Meese (1981), Celeux and Soromenho (1996)).”
BackgroundMethodInfluentialReligiosity Profiles of American Youth in Relation to Substance Use, Violence, and Delinquency
Salas-Wright, C. P., Vaughn, M. G., Hodge, D. R., & Perron, B. E. (2012) · Journal of Youth and Adolescence · doi:10.1007/s10964-012-9761-z · Cited by 170
Kreuter, F., & Muthén, B. (2008) · Journal of Quantitative Criminology · doi:10.1007/s10940-007-9036-0 · Cited by 169
“However, for finite mixture models, the AIC has been shown to overestimate the correct number of components (Soromenho 1994; Celeux and Soromenho 1996).”
Cognitive decline in the elderly: an analysis of population heterogeneity
Hayden, K. M., Reed, B. R., Manly, J. J., Tommet, D., Pietrzak, R. H., Chelune, G. J., Yang, F. M., Revell, A. J., Bennett, D. A., & Jones, R. N. (2011) · Age and Ageing · doi:10.1093/ageing/afr101 · Cited by 168
“Model fit was evaluated with standard approaches for profile mixture modelling, considering absolute fit of model-implied values to observed values (Empirical r(2)) [19], quality of assignment (entropy: range 0–1; higher values signal better models) [20], the probability of class membership (posterior probability) and…”
MethodFirst-grade predictors of mathematical learning disability: A latent class trajectory analysis
Geary, D. C., Bailey, D. H., Littlefield, A., Wood, P., Hoard, M. K., & Nugent, L. (2009) · Cognitive Development · doi:10.1016/j.cogdev.2009.10.001 · Cited by 167
“The entropy value for the 4-class model-bound between 0 and 1 (with values of 1 indicating a clear delineation between classes; Celeux & Soromenho, 1996) was 0.82.”
InfluentialRationality on the rise: Why relative risk aversion increases with stake size
Fehr-Duda, H., Bruhin, A., Epper, T., & Schubert, R. (2010) · Journal of Risk and Uncertainty · doi:10.1007/s11166-010-9090-0 · Cited by 166
“Support We calculated AIC, BIC and NEC for three different model sizes, C ∈ {1, 2, 3}, presented in Table 3.”
BackgroundMethodInfluentialTrajectories of Social Engagement and Limitations in Late Life
Thomas, P. A. (2011) · Journal of Health and Social Behavior · doi:10.1177/0022146511411922 · Cited by 164
Doong, S.-H., Dhruva, A., Dunn, L. B., West, C., Paul, S. M., Cooper, B. A., Elboim, C., Abrams, G., Merriman, J. D., Langford, D. J., et al. (2015) · Biological Research For Nursing · doi:10.1177/1099800414550394 · Cited by 164
“In addition, betterfitting models should produce higher entropy values (Celeux & Soromenho, 1996).”
BackgroundLacey, R. E., Howe, L. D., Kelly-Irving, M., Bartley, M., & Kelly, Y. (2022) · Journal of Interpersonal Violence · doi:10.1177/0886260520935096 · Cited by 165
“Lower AIC, BIC, and SSABIC values indicate a better fitting class solution, and entropy values closer to 1 show clearer distinction of classes (Celeux & Soromenho, 1996).”
BackgroundDescribing disability through individual-level mixture models for multivariate binary data
Erosheva, E. A., Fienberg, S. E., & Joutard, C. (2007) · The Annals of Applied Statistics · doi:10.1214/07-aoas126 · Cited by 159
“ontinue to apply both criteria and to study their performance in a mixture context. Findings in populationlevel mixture models suggest that AIC tends to overestimate the correct number of components [Celeux and Soromenho (1996)], while BIC shows better results [Leroux (1992), Roeder and Wasserman (1997)]. Questions of…”
ResultUher, R., Mors, O., Rietschel, M., Rajewska-Rager, A., Petrovic, A., Zobel, A., Henigsberg, N., Mendlewicz, J., Aitchison, K., Farmer, A., et al. (2011) · Journal of Clinical Psychiatry · doi:10.4088/jcp.10m06419 · Cited by 139
Job crafting profiles and work engagement: A person-centered approach
Mäkikangas, A. (2018) · Journal of Vocational Behavior · doi:10.1016/j.jvb.2018.01.001 · Cited by 196
Lacey, R. E., Pereira, S. M. P., Li, L., & Danese, A. (2020) · Brain Behavior and Immunity · doi:10.1016/j.bbi.2020.03.017 · Cited by 166
Markon, K. E., & Krueger, R. F. (2004) · Behavior Genetics · doi:10.1007/s10519-004-5587-0 · Cited by 165
“However, these studies have also generally concluded that neither criterion is completely satisfactory, and that other fit criteria exhibit greater power to select true models ( Celeux and Soromenho, 1996; Lin and Dayton, 1997; Yang, 1998).”
Resultvan der Velden, P. G., Hyland, P., Contino, C., von Gaudecker, H., Muffels, R. J. A., & Das, M. (2021) · PLoS ONE · doi:10.1371/journal.pone.0245057 · Cited by 174
“In addition, we looked at entropy values that are measures of the quality of classification (approaching one indicates clear delineation of classes) [38].”
BackgroundIdentity and the body: Trajectories of body esteem from adolescence to emerging adulthood
Nelson, S. C., Kling, J., Wängqvist, M., Frisén, A., & Syed, M. (2018) · Developmental Psychology · doi:10.1037/dev0000435 · Cited by 151
Self-regulated learning and self-directed study in a pre-college sample
Abar, B. W., & Loken, E. (2009) · Learning and Individual Differences · doi:10.1016/j.lindif.2009.09.002 · Cited by 160
“Entropy approaching 1 indicates little profile overlap (Celeux & Soromenho, 1996) The smallest profile (15”
BackgroundMäkikangas, A., Bakker, A. B., Aunola, K., & Demerouti, E. (2009) · Journal of Occupational and Organizational Psychology · doi:10.1348/096317909x476333 · Cited by 201
“…of a model with at least kþ 1 classes); (c) the classification quality that can be determined by entropy values; entropy values range from 0 to 1, where values close to 1 indicate clear classification (Celeux & Soromenho, 1996); and (d) usefulness and clarity of the latent classes in practice.”
BackgroundCustomer Orientation Structure for Internet-Based Business-to-Business Platform Firms
Chakravarty, A., Kumar, A., & Grewal, R. (2014) · Journal of Marketing · doi:10.1509/jm.12.0442 · Cited by 182
“The entropy of separation of the four-class solution for each equation exceeded .85, indicating a good separation of the latent classes (Celeux and Soromenho 1996).”
BackgroundBishop-Fitzpatrick, L., Hong, J., Smith, L. E., Makuch, R. A., Greenberg, J. S., & Mailick, M. R. (2016) · Journal of Autism and Developmental Disorders · doi:10.1007/s10803-016-2816-3 · Cited by 142
“In addition, entropy values approaching one were deemed to be indicative of clear delineation of classes, which helped to confirm power estimates (Celeux and Soromenho 1996).”
ResultAhrens, K. F., Neumann, R. J., Kollmann, B., Brokelmann, J., von Werthern, N. M., Malyshau, A., Weichert, D., Lutz, B., Fiebach, C. J., Wessa, M., et al. (2021) · Translational Psychiatry · doi:10.1038/s41398-021-01508-2 · Cited by 157
“825 indicates that participants are well-assigned in the three-class model [44].”
BackgroundMcAuley, E., Mailey, E. L., Mullen, S. P., Szabo‐Reed, A. N., Wójcicki, T. R., White, S. M., Gothe, N. P., Olson, E. A., & Kramer, A. F. (2010) · Health Psychology · doi:10.1037/a0021567 · Cited by 164
“These values range from 0 to 1, with 0 corresponding to randomness and 1 to a perfect classification (Celeux & Soromenho, 1996).”
Miaskowski, C. A., Cooper, B. A., Paul, S. M., West, C. M., Langford, D. J., Levine, J. D., Abrams, G., Hamolsky, D., Dunn, L. B., Dodd, M., et al. (2012) · Journal of Pain · doi:10.1016/j.jpain.2012.09.013 · Cited by 155
Danks, N. P., Sharma, P. N., & Sarstedt, M. (2020) · Journal of Business Research · doi:10.1016/j.jbusres.2020.03.019 · Cited by 161
Sasidharan, L., Wu, K.-F., & Menéndez, M. (2015) · Accident Analysis & Prevention · doi:10.1016/j.aap.2015.09.020 · Cited by 135
Assessing emotion regulation repertoires: The Regulation of Emotion Systems Survey
France, K. D., & Hollenstein, T. (2017) · Personality and Individual Differences · doi:10.1016/j.paid.2017.07.018 · Cited by 157
Breakdown points for maximum likelihood estimators of location–scale mixtures
Hennig, C. M. (2004) · The Annals of Statistics · doi:10.1214/009053604000000571 · Cited by 129
“Further suggestions to estimate s, which are more difficult to analyze with respect to the breakdown properties, are given, for example, by Bozdogan (1994) and Celeux and Soromenho (1996).”
MethodMiaskowski, C. A., Dunn, L. B., Ritchie, C. S., Paul, S. M., Cooper, B. A., Aouizerat, B. E., Alexander, K. E., Skerman, H. M., & Yates, P. M. (2015) · Journal of Pain and Symptom Management · doi:10.1016/j.jpainsymman.2014.12.011 · Cited by 124
Choosing models in model-based clustering and discriminant analysis
Biernacki, C., & Govaert, G. (1999) · Journal of Statistical Computation and Simulation · doi:10.1080/00949659908811966 · Cited by 125
Miaskowski, C. A., Cooper, B. A., Melisko, M., Chen, L.-M., Mastick, J., West, C. M., Paul, S. M., Dunn, L. B., Schmidt, B. L., Hammer, M. J., et al. (2014) · Cancer · doi:10.1002/cncr.28699 · Cited by 122
“80.(27) Finally, well-fitting models “make sense” conceptually and the estimated classes differ as might be expected on variables not used in the generation of the model.”
BackgroundOn Becoming (Un)Committed: A Taxonomy and Test of Newcomer Onboarding Scenarios
Solinger, O. N., van Olffen, W., Roe, R. A., & Hofmans, J. (2013) · Organization Science · doi:10.1287/orsc.1120.0818 · Cited by 149
“Entropy values range from 0 to 1, with 0 corresponding to random classification and 1 corresponding to a perfect classification (Celeux and Soromenho 1996).”
BackgroundUnhealthy behavior clustering and mental health status in United States college students
Jao, N. C., Robinson, L. D., Kelly, P. J., Ciecierski, C. C., & Hitsman, B. L. (2018) · Journal of American College Health · doi:10.1080/07448481.2018.1515744 · Cited by 136
“Entropy is an indication of the delineation of classes within the data.(57) In LCA, the local independence is assumed (ie, health behaviors are independent from each other within each class).”
BackgroundSmith, T. E., & Short, A. (2022) · Addictive Behaviors · doi:10.1016/j.addbeh.2022.107259 · Cited by 155
Dunn, L. B., Cooper, B. A., Neuhaus, J. M., West, C. M., Paul, S. M., Aouizerat, B. E., Abrams, G., Edrington, J., Hamolsky, D., & Miaskowski, C. A. (2011) · Health Psychology · doi:10.1037/a0024366 · Cited by 144
“Better-fitting models should produce higher entropy values (Celeux & Soromenho, 1996; Muthén & Muthén, 1998-2008).”
Socialization in Open Source Software Projects: A Growth Mixture Modeling Approach
Qureshi, I., & Fang, Y. (2010) · Organizational Research Methods · doi:10.1177/1094428110375002 · Cited by 137
“…used indicator of classification quality and it ranges from 0.0 to 1.0, where 1.0 represents a better classification (Hix-Small, Duncan, Duncan, & Okut, 2004; Jedidi, Ramaswamy, & Desarbo, 1993; Nagin, 1999) in the sense that there is clear delineation between classes (Celeux & Soromenho, 1996).”
BackgroundMethodInfluentialBarhight, L. R., Hubbard, J. A., & Hyde, C. (2012) · Child Development · doi:10.1111/j.1467-8624.2012.01839.x · Cited by 159
“Finally, the normalized entropy criterion (NEC), proposed by Celeux and Soromenho (1996), was used to indicate how well the model classified individuals into groups.”
MethodBacker, A. D., Martínez, G., Rosenauer, A., & Aert, S. V. (2013) · Ultramicroscopy · doi:10.1016/j.ultramic.2013.05.003 · Cited by 120
“Since BIC and CLC tend to underestimate and overestimate the number of components of the probability distribution respectively [45,46], the Integrated Classification Likelihood criterion (ICL) is introduced to overcome these shortcomings [47]:”
BackgroundFinite mixtures of multivariate Poisson distributions with application
Karlis, D., & Meligkotsidou, L. (2006) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2006.07.001 · Cited by 119
“We also used the Bayesian information criterion given by BIC(K)=−2L(K)+ dK log(n) which uses a penalty that contains the sample size and the normalized entropy criterion proposed by Celeux and Soromenho (1996). The latter is a classification criterion in the sense that it measures how good the classification offered b…”
MethodChildhood sexual abuse and adolescent substance use: A latent class analysis
Shin, S. H., Hong, H. G., & Hazen, A. L. (2010) · Drug and Alcohol Dependence · doi:10.1016/j.drugalcdep.2010.01.013 · Cited by 142
Support of marijuana tourism in Colorado: A residents’ perspective using social exchange theory
Kang, S. K., & Lee, J. (2018) · Journal of Destination Marketing & Management · doi:10.1016/j.jdmm.2018.03.003 · Cited by 132
Retention of latent segments in regression-based marketing models
Andrews, R. L., & Currim, I. S. (2003) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2003.04.001 · Cited by 114
A latent profile analysis of university students’ self-regulated learning strategies
Ning, H. K., & Downing, K. J. (2014) · Studies in Higher Education · doi:10.1080/03075079.2014.880832 · Cited by 154
Mixture model clustering for mixed data with missing information
Hunt, L., & Jorgensen, M. A. (2003) · Computational Statistics & Data Analysis · doi:10.1016/s0167-9473(02)00190-1 · Cited by 114
“Determination of the appropriate number of underlying clusters is still an unresolved problem, and there does not appear to be a universally superior method of determining the cluster number (see for example, Celeux and Soromenho (1996) and the references therein).”
MethodBécares, L., & Priest, N. (2015) · PLoS ONE · doi:10.1371/journal.pone.0141363 · Cited by 144
“Entropy statistics, which measure the separation of the classes based on the posterior class membership probabilities, were also examined; entropy values approaching 1 indicate clear separation between classes [69].”
Berger, N., Mackenzie, E., & Holmes, K. (2020) · The Australian Educational Researcher · doi:10.1007/s13384-020-00379-8 · Cited by 157
Work engagement and its antecedents in remote work: A person-centered view
Mäkikangas, A., Juutinen, S., Mäkiniemi, J.-P., Sjöblom, K., & Oksanen, A. (2022) · Work & Stress · doi:10.1080/02678373.2022.2080777 · Cited by 200
“The entropy value indicates the quality of the profiling: the closer the entropy value is to 1, the higher is the precision with which the cases are classified into profiles (Celeux & Soromenho, 1996).”
BackgroundRecommended Practices in Latent Class Analysis Using the Open-Source R-Package tidySEM
Lissa, C. J. V., Garnier‐Villarreal, M., & Anadria, D. (2023) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2023.2250920 · Cited by 141
“The entropy criterion summarizes class separability in a single statistic (Celeux & Soromenho, 1996).”
BackgroundMiaskowski, C. A., Paul, S. M., Cooper, B. A., West, C. M., Levine, J. D., Elboim, C., Hamolsky, D., Abrams, G., Luce, J. A., Dhruva, A., et al. (2014) · European Journal of Oncology Nursing · doi:10.1016/j.ejon.2013.12.002 · Cited by 125
Not all roads lead to Rome — Comparing different types of motivational regulation profiles
Schwinger, M., Steinmayr, R., & Spinath, B. (2011) · Learning and Individual Differences · doi:10.1016/j.lindif.2011.12.006 · Cited by 138
The many Faces of Human Sociality: Uncovering the Distribution and Stability of Social Preferences
Bruhin, A., Fehr, E., & Schunk, D. (2018) · Journal of the European Economic Association · doi:10.1093/jeea/jvy018 · Cited by 156
Lebret, R., Iovleff, S., Langrognet, F., Biernacki, C., Celeux, G., & Govaert, G. (2015) · Journal of Statistical Software · doi:10.18637/jss.v067.i06 · Cited by 120
“The index K is used to highlight that NEC is essentially devoted to choosing the number of mixture components K , not the model parameterization (Celeux and Soromenho 1996; Biernacki, Celeux, and Govaert 1999).”
Self-citation
Metadata
- Authors
- Celeux, G., & Soromenho, G.
- Affiliations
- Institut national de recherche en sciences et technologies du numérique (FR)
- University of Lisbon (PT)
- ORCID
- Soromenho, Gilda
- Venue
- Journal of Classification
- Published
- 1996-09
- Volume
- 13
- Issue
- 2
- Pages
- 195-212
- ISSN
- 0176-4268 (ISSN-L)
- 1432-1343
- OpenAlex
- W2079817125
- Cited by
- 2,448
- References
- 37
- Citations by year
- 2026: 248 · 2025: 223 · 2024: 213 · 2023: 208 · 2022: 184 · 2021: 204 · 2020: 176 · 2019: 183 · 2018: 112 · 2017: 100 · 2016: 105 · 2015: 92 · 2014: 85 · 2013: 75 · 2012: 49
- FWCI
- 1.91
- Type
- journal article
- Language
- en
- Open access
- green
- inria.hal.science
- Licence
- other-oa
- www.springer.com/tdm (tdm)
- Locations
- Journal of Classification (publishedVersion)
- OpenGrey (Institut de l'Information Scientifique et Technique) (submittedVersion)
- HAL (submittedVersion, other-oa)
- RePEc: Research Papers in Economics (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Methods and Mixture Models
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Clustering Algorithms Research
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Keywords
- entropy criterion, number of clusters, mixture models, model selection, likelihood ratios, optimal number of clusters
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.