An entropy criterion for assessing the number of clusters in a mixture model
Open accessThe DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.
Overview
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
Items 201–300 of 1,010
Social support networks of older migrants in England and Wales: the role of collectivist culture
Burholt, V., Dobbs, C., & Victor, C. R. (2017) · Ageing and Society · doi:10.1017/s0144686x17000034 · Cited by 80
Liu, M., & Oga-Baldwin, W. L. Q. (2022) · System · doi:10.1016/j.system.2022.102762 · Cited by 81
“Similarly, entropy (Celeux & Soromenho, 1996), with higher values (ranging from 0 to 1) indicating better class separation, is not recommended for final model selection (Collins & Lanza, 2010).”
Bartolucci, F., Lupparelli, M., & Montanari, G. E. (2009) · The Annals of Applied Statistics · doi:10.1214/08-aoas230 · Cited by 68
Assessing metastatic potential of breast cancer cells based on EGFR dynamics
Liu, Y., Chou, C., Kim, M., Vasisht, R., Kuo, Y.-A., Ang, P., Liu, C., Perillo, E. P., Chen, Y., Blocher, K., et al. (2019) · Scientific Reports · doi:10.1038/s41598-018-37625-0 · Cited by 67
Seng, J. J. B., Kwan, Y. H., Lee, V. S. Y., Tan, C. S., Zainudin, S. B., Thumboo, J., & Low, L. L. (2020) · Diabetes Care · doi:10.2337/dc19-2519 · Cited by 62
“Indices used included the Akaike information criterion, Bayesian information criterion (BIC), sample size–adjusted BIC,andentropy, forwhichhigherentropy and smaller values of Akaike information criterion, BIC, and sample size–adjusted BIC indicate better model fit (16).”
MethodDeLisi, M., Vaughn, M. G., Salas‐Wright, C. P., & Jennings, W. G. (2015) · Journal of Drug Issues · doi:10.1177/0022042615579237 · Cited by 62
“In interpreting these criteria, higher log likelihood values and lower BIC, AIC, and CAIC values reflect better model fit (Celeux & Soromenho, 1996).”
Rose, T., McDonald, A. R., Mach, T. V., Witherspoon, D. P., & Lambert, S. F. (2019) · Journal of Youth and Adolescence · doi:10.1007/s10964-019-01135-7 · Cited by 71
Yuan, Y., Yang, M., Guo, Y., Rasouli, S., Gan, Z., & Ren, Y. (2020) · Journal of Safety Research · doi:10.1016/j.jsr.2020.12.012 · Cited by 68
Cunningham, C. E., Chen, Y., Deal, K., Rimas, H. L., McGrath, P. J., Reid, G. J., Lipman, E. L., & Corkum, P. V. (2013) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-013-9728-x · Cited by 67
Södergren, M., Wang, W., Salmon, J., Ball, K., Crawford, D. A., & McNaughton, S. A. (2013) · Maturitas · doi:10.1016/j.maturitas.2013.09.010 · Cited by 69
Currier, J. M., Holland, J. M., & Drescher, K. D. (2014) · PLoS ONE · doi:10.1371/journal.pone.0101741 · Cited by 61
A two-step method for detecting selection signatures using genetic markers
Daniel Gianola, B., Simianer, H., & Qanbari, S. (2010) · Genetics Research · doi:10.1017/s0016672310000121 · Cited by 61
Social climate profiles in adolescent sports: Associations with enjoyment and intention to continue
Gardner, L. A., Magee, C., & Vella, S. A. (2016) · Journal of Adolescence · doi:10.1016/j.adolescence.2016.08.003 · Cited by 86
“540 Other limitations include the relatively small percentage of variance explained by the 541 profiles and the entropy level in the latent profile analysis falling short of the recommended 542 level of .80 (Celeux & Soromenho, 1996) which may reflect issues of distinctiveness.”
BackgroundHolzberger, D., Praetorius, A., Seidel, T., & Kunter, M. (2019) · European Journal of Psychology of Education · doi:10.1007/s10212-018-00410-8 · Cited by 80
Lian, J., Kiely, K. M., & Anstey, K. J. (2022) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2022.105486 · Cited by 72
Onselen, C. V., Cooper, B. A., Lee, K. A., Dunn, L. B., Aouizerat, B. E., West, C. M., Dodd, M. J., Paul, S. M., & Miaskowski, C. A. (2012) · Supportive Care in Cancer · doi:10.1007/s00520-012-1381-3 · Cited by 67
Stapleton, J. L., Turrisi, R., Hillhouse, J., Robinson, J. K., & Abar, B. W. (2010) · Journal of Behavioral Medicine · doi:10.1007/s10865-009-9246-z · Cited by 65
“An entropy value above 0.80 suggests well separated and distinct latent profiles as well as good predictive value of the tanner profiles by the indoor tanning motive indicator variables (Celeux and Soromenho 1996).”
BackgroundInfluentialMäkikangas, A., Hyvönen, K., & Feldt, T. (2017) · Burnout Research · doi:10.1016/j.burn.2017.04.002 · Cited by 76
“Both the entropy and AvePP values range from 0 to 1; for entropy, values close to 1 indicate a clear classification (Celeux & Soromenho, 1996).”
BackgroundThe Generalized Cross Entropy Method, with Applications to Probability Density Estimation
Botev, Z. I., & Kroese, D. P. (2009) · Methodology And Computing In Applied Probability · doi:10.1007/s11009-009-9133-7 · Cited by 62
“The performance of these methods, however, has been contentious for various theoretical and practical reasons [60, 12, 4].”
BackgroundMagno, L., Dourado, I. C., da Silva, L. A. V., Brignol, S., Amorim, L. D. A. F., & MacCarthy, S. (2018) · PLoS ONE · doi:10.1371/journal.pone.0194306 · Cited by 94
Corso, L. D., CARLO, A. D., Carluccio, F., Colledani, D., & Falco, A. (2020) · PLoS ONE · doi:10.1371/journal.pone.0242267 · Cited by 79
Langford, D. J., Paul, S. M., Cooper, B. A., Kober, K. M., Mastick, J., Melisko, M., Levine, J. D., Wright, F., Hammer, M. J., Cartwright, F., et al. (2015) · Supportive Care in Cancer · doi:10.1007/s00520-015-2819-1 · Cited by 66
“Statistical fit indices are used to determine the number of latent classes and to evaluate model fit [24, 25].”
MethodPerko, K., Kinnunen, U. M., Tolvanen, A. J., & Feldt, T. (2015) · European Journal of Work and Organizational Psychology · doi:10.1080/1359432x.2015.1011136 · Cited by 85
Accurate photometric redshift probability density estimation – method comparison and application
Rau, M. M., Seitz, S., Brimioulle, F., Frank, E., Friedrich, O., Gruen, D., & Hoyle, B. (2015) · Monthly Notices of the Royal Astronomical Society · doi:10.1093/mnras/stv1567 · Cited by 73
“(Celeux & Soromenho 1996) why we cannot use the entropy E(K) directly.”
MethodInfluentialShip Detection in SAR Images via Local Contrast of Fisher Vectors
Wang, X., Li, G., Zhang, X.-P. S., & He, Y. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.2976880 · Cited by 58
“We refer readers to [53] for more alternative methods to choose the value of Q.”
MethodDemaray, M. K., Malecki, C. K., Ryoo, J. H., & Summers, K. H. (2021) · Journal of School Psychology · doi:10.1016/j.jsp.2021.02.006 · Cited by 78
Herrenkohl, T. I., Lee, J. O., Kosterman, R., & Hawkins, J. D. (2012) · Journal of Adolescent Health · doi:10.1016/j.jadohealth.2011.11.003 · Cited by 66
Beyond Young, Highly Educated Males: A Typology of VAA Users
de Pol, J. V., Holleman, B. C., Kamoen, N., Krouwel, A. P. M., & de Vreese, C. H. (2014) · Journal of Information Technology & Politics · doi:10.1080/19331681.2014.958794 · Cited by 58
The Development of Early Profiles of Temperament: Characterization, Continuity, and Etiology
Beekman, C. R., Neiderhiser, J. M., Buss, K. A., Loken, E., Moore, G. A., Leve, L. D., Ganiban, J. M., Shaw, D. S., & Reiss, D. (2015) · Child Development · doi:10.1111/cdev.12417 · Cited by 73
“Model entropy values range from 0 to 1; values closer to 1 typically indicate greater confidence in appropriate classification (Celeux & Soromenho, 1996).”
BackgroundPredictors for Persistence of Functional Somatic Symptoms in Adolescents
Janssens, K. A. M., Klis, S., Kingma, E. M., Oldehinkel, A. J., & Rosmalen, J. G. M. (2014) · The Journal of Pediatrics · doi:10.1016/j.jpeds.2013.12.003 · Cited by 69
Papalia, N. L., Luebbers, S., Ogloff, J. R. P., Cutajar, M. C., Mullen, P. E., & Mann, E. (2017) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2017.02.040 · Cited by 70
Development and Validation of the Mukbang Addiction Scale
Kırcaburun, K., Stavropoulos, V., Harris, A., Calado, F., Emirtekin, E., & Griffiths, M. D. (2020) · International Journal of Mental Health and Addiction · doi:10.1007/s11469-019-00210-1 · Cited by 61
“Models with progressively increasing number of classes/profiles were examined and their fit was compared based on the combination of the following fit indices: (i) Akaike Information Criterion (AIC; Vrieze 2012); (ii) Bayesian Information Criterion (BIC; Vrieze 2012); (iii) Vuong-Lo-Mendell-Rubin test (VLMR-LRT; Jung…”
Methodvan Alten, D. C. D., Phielix, C., Janssen, J. J. H. M., & Kester, L. (2020) · Computers in Human Behavior · doi:10.1016/j.chb.2020.106676 · Cited by 76
Sanchez-Cerezo, J., Neale, J., Julius, N., Croudace, T., Lynn, R. M., Hudson, L. D., & Nicholls, D. (2024) · EClinicalMedicine · doi:10.1016/j.eclinm.2024.102440 · Cited by 56
Ing, M. M., & Nylund-Gibson, K. K. (2013) · Educational Research and Evaluation · doi:10.1080/13803611.2013.806218 · Cited by 64
“Entropy ranges between 0 and 1, where 1 is perfect classification; and values approaching 1 indicate clear delineation of classes (Celeux & Soromenho, 1996).”
BackgroundNational identity profiles and support for the European Union
Aichholzer, J., Kritzinger, S., & Plescia, C. (2021) · European Union Politics · doi:10.1177/1465116520980068 · Cited by 70
“…as relative fit indices, where smaller values indicate better model fit, and entropy as a marker of the clarity of class delineation, with values closer to 1 indicating greater classification accuracy and more precise assignment of individuals to latent profiles (e.g., Celeux and Soromenho, 1996).”
BackgroundHolzberger, D., Maurer, C., Kunina‐Habenicht, O., & Kunter, M. (2021) · Teaching and Teacher Education · doi:10.1016/j.tate.2021.103285 · Cited by 79
A latent class analysis of trauma based on a nationally representative sample of US adolescents
McChesney, G. C., Adamson, G., & Shevlin, M. E. (2015) · Social Psychiatry and Psychiatric Epidemiology · doi:10.1007/s00127-015-1075-5 · Cited by 64
“Entropy measures were used to determine how accurately participants were classified, with higher values (ranging from 0 to 1) indicating better classification [37].”
MethodProperties of the Cannabis Abuse Screening Test (CAST) in the general population
Legleye, S., Guignard, R., Richard, J., Kraus, L., Pabst, A., & Beck, F. (2015) · International Journal of Methods in Psychiatric Research · doi:10.1002/mpr.1465 · Cited by 67
“The model with the optimal number of classes was the one with the lowest BIC (Bayesian information criterion) and the highest entropy (Celeux and Soromenho, 1996).”
MethodDifferential Developmental Profiles of Adolescents Using Sexually Explicit Internet Material
Doornwaard, S. M., van den Eijnden, R. J. J. M., Overbeek, G. J., & ter Bogt, T. (2014) · The Journal of Sex Research · doi:10.1080/00224499.2013.866195 · Cited by 61
“The accuracy of the classification of individuals in latent trajectories is judged with the entropy, where values of .8 or higher indicate a good classification (Celeux & Soromenho, 1996).”
Mixture Modeling: Applications in Educational Psychology
Harring, J. R., & Hodis, F. A. (2016) · Educational Psychologist · doi:10.1080/00461520.2016.1207176 · Cited by 62
“Classification accuracy is another important consideration when choosing on a final number of classes and is typically captured through a measure of entropy (Celeux & Soromenho, 1996).”
BackgroundCareless responding in internet-based quality of life assessments
Schneider, S., May, M., & Stone, A. A. (2017) · Quality of Life Research · doi:10.1007/s11136-017-1767-2 · Cited by 70
“1 3 which means that respondents could be assigned to the latent classes with very little uncertainty [46].”
Model Selection for Mixture Models – Perspectives and Strategies
Celeux, G., Frühwirth‐Schnatter, S., & Robert, C. P. (2019) · Chapman & Hall/CRC eBooks · doi:10.1201/9780429055911-7 · Cited by 59
“, z (m) n ) remains instead concentrated around one single or a subset of the modes of the posterior distribution, the approximation of log p̂(θo|y, G) based on Chib’s representation fails, in that it is usually off by a numerical factor of order O(log G!). Furthermore, this order cannot be used as a reliable correcti…”
MethodSelf-citationWilliams, N. J., Frank, H. E., Frederick, L., Beidas, R. S., Mandell, D. S., Aarons, G. A., Green, P., & Locke, J. (2019) · Implementation Science · doi:10.1186/s13012-019-0863-9 · Cited by 54
“, values closer to 1 indicate greater separation between classes; [64]); sufficiently populated classes (i.”
BackgroundJinnin, R., Okamoto, Y., Takagaki, K., Nishiyama, Y., Yamamura, T., Okamoto, Y., Miyake, Y., Takebayashi, Y., Tanaka, K., Sugiura, Y., et al. (2016) · Neuropsychiatric Disease and Treatment · doi:10.2147/ndt.s117846 · Cited by 61
de Lijster, J. M., van den Dries, M. A., van der Ende, J., Utens, E. M. W. J., Jaddoe, V. W. V., Dieleman, G. C., Hillegers, M. H. J., Tiemeier, H. W., & Legerstee, J. S. (2019) · Research on Child and Adolescent Psychopathology · doi:10.1007/s10802-019-00550-5 · Cited by 59
Zamecnik, A., Kovanović, V., Joksimovíc, S., & Liu, L. (2022) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2022.100051 · Cited by 88
Paternal Work Stress and Latent Profiles of Father-Infant Parenting Quality
Goodman, W. B., Crouter, A. C., Lanza, S. T., Cox, M. J., Vernon‐Feagans, L., & The Family Life Project Key Investigators (2011) · Journal of Marriage and the Family · doi:10.1111/j.1741-3737.2011.00826.x · Cited by 72
“…multiple fit indices, including the Bayesian information criterion (BIC), Akaike information criterion (AIC), Lo-Mendell-Rubin likelihood ratio test (Lo, Mendell, & Rubin, 2001), and entropy statistic (Celeux & Soromenho, 1996) for each model were examined to determine the optimal number of classes.”
InfluentialCHull as an alternative to AIC and BIC in the context of mixtures of factor analyzers
Bulteel, K., Wilderjans, T. F., Tuerlinckx, F., & Ceulemans, E. (2013) · Behavior Research Methods · doi:10.3758/s13428-012-0293-y · Cited by 60
“Consequently, AIC will often select a too complex model, which is a tendency of AIC that has already been observed in the context of many models (e.g., Celeux & Soromenho, 1996; Kass & Raftery, 1995).”
BackgroundMethodFryer, L. K. (2016) · Higher Education · doi:10.1007/s10734-016-0094-9 · Cited by 59
McElroy, E., Shevlin, M. E., & Murphy, J. (2017) · Comprehensive Psychiatry · doi:10.1016/j.comppsych.2017.03.003 · Cited by 64
“Entropy values range from 0 – 1, with higher values reflecting more accurate classification [33].”
MethodInfluentialBeyond the local-newcomer divide: Village attachment in the era of mobilities
Gieling, J., Vermeij, L., & Haartsen, T. (2017) · Journal of Rural Studies · doi:10.1016/j.jrurstud.2017.08.015 · Cited by 66
“Secondly, entropy was examined as an estimation of how distinct the latent classes identified were from one another: values approaching 1 indicate a clear delineation of classes (Celeux and Soromenho, 1996).”
BackgroundThe clustering of risk behaviours in adolescence and health consequences in middle age
Akasaki, M., Ploubidis, G. B., Dodgeon, B., & Bonell, C. P. (2019) · Journal of Adolescence · doi:10.1016/j.adolescence.2019.11.003 · Cited by 74
Trajectories of Economic Disconnection among Families in the Child Welfare System
Hook, J. L., Romich, J. L., Lee, J. A., Marcenko, M. O., & Kang, J. Y. (2016) · Social Problems · doi:10.1093/socpro/spw006 · Cited by 54
Kruikemeier, S., Boerman, S. C., & Bol, N. (2020) · Media Psychology · doi:10.1080/15213269.2019.1598434 · Cited by 54
“Another measure to assess the fit is the entropy value (Celeux & Soromenho, 1996), a higher value indicates better fit.”
BackgroundMethodCognitive profiles in children and adolescents with Down syndrome
Onnivello, S., Pulina, F., Locatelli, C., Marcolin, C., Ramacieri, G., Antonaros, F., Vione, B., Caracausi, M., & Lanfranchi, S. (2022) · Scientific Reports · doi:10.1038/s41598-022-05825-4 · Cited by 74
Chan, P. A., Rose, J. S., Maher, J. A., Benben, S., Pfeiffer, K., Almonte, A. A., Poceta, J., Oldenburg, C. E., Parker, S., Marshall, B. D. L., et al. (2015) · AIDS Patient Care and STDs · doi:10.1089/apc.2015.0113 · Cited by 55
Latent class and latent profile models.
Flaherty, B. P., & Kiff, C. J. (2012) · American Psychological Association eBooks · doi:10.1037/13621-019 · Cited by 54
Astrup, G. L., Hofsø, K., Bjordal, K., Guren, M. G., Vistad, I., Cooper, B. A., Miaskowski, C. A., & Rustøen, T. (2017) · Acta Oncologica · doi:10.1080/0284186x.2016.1273546 · Cited by 58
Drossel, K., Eickelmann, B., & Vennemann, M. (2020) · Large-scale Assessments in Education · doi:10.1186/s40536-020-00087-w · Cited by 98
“…of model-parsimony we decided to choose the 3-group-solution which is characterized by a relatively low BIC and at the same time by a high entropy which is a measure of classification quality and values approaching one indication a clear delineation of profiles (Celeux and Soromenho 1996).”
BackgroundSteinmann, I., Strietholt, R., & Braeken, J. (2021) · Psychological Methods · doi:10.1037/met0000392 · Cited by 81
“The entropy values of the constrained FMAs ranged between .810 and .954 across the datasets, which we interpret as an indication for appropriate classification (Celeux & Soromenho, 1996).”
BackgroundJiang, X., Zhang, C., Pan, Y., Cheng, X., & Zhang, W. (2023) · Scientific Reports · doi:10.1038/s41598-023-42352-2 · Cited by 62
“Generally, an entropy value > 0.80 is considered indicative of high classification accuracy, with a higher entropy value indicating a better goodness-of-fit of the model 18 .”
BackgroundInfluentialCharacteristics and Comorbid Symptoms of Older Adults Reporting Death Ideation
Van Orden, K. A., Simning, A., Conwell, Y., Skoog, I., & Waern, M. (2013) · The American Journal of Geriatric Psychiatry · doi:10.1016/j.jagp.2013.01.015 · Cited by 54
Patterns of Alcohol Use and Consequences Among Empirically Derived Sexual Minority Subgroups
Talley, A. E., Sher, K. J., Steinley, D., Wood, P. K., & Littlefield, A. K. (2012) · Journal of Studies on Alcohol and Drugs · doi:10.15288/jsad.2012.73.290 · Cited by 58
“Relative entropy (range: 0–1) was used as an indicator of how accurately participants were classifi ed into latent groups (Celeux and Soromenho, 1996), with higher values indicating a solution where each observed response pattern was likely to be assigned accurately to one latent class over others.”
MethodFunctional Outcomes of Young Adults with Childhood ADHD: A Latent Profile Analysis
Merrill, B. M., Molina, B. S. G., Coxe, S. J., Gnagy, E. M., Altszuler, A. R., Macphee, F. L., Morrow, A. S., Trucco, E. M., & Pelham, W. E. (2019) · Journal of Clinical Child & Adolescent Psychology · doi:10.1080/15374416.2018.1547968 · Cited by 63
“…of parameters has significantly better fit; and entropy with values approaching 1 indicating better classification of individuals into profiles; Celeux & Soromenho, 1996; Nylund, Asparouhov, & Muthén, 2007) in conjunction with theory-based decision making and interpretability influenced the…”
BackgroundLanza, S. T., & Bray, B. C. (2010) · PubMed · Cited by 52
Latent class analysis of incomplete data via an entropy-based criterion
Larose, C. D., Harel, O., Kordas, K., & Dey, D. K. (2016) · Statistical Methodology · doi:10.1016/j.stamet.2016.04.004 · Cited by 53
“There is also an entropy-based criterion for mixture model data, but it was applied to complete, normal mixture model data [4].”
BackgroundMethodPenagos-Londoño, G. I., Rodriguez–Sanchez, C., Ruiz-Moreno, F., & Torres, E. (2021) · Journal of Destination Marketing & Management · doi:10.1016/j.jdmm.2020.100532 · Cited by 51
Bayesian Plackett–Luce Mixture Models for Partially Ranked Data
Mollica, C., & Tardella, L. (2016) · Psychometrika · doi:10.1007/s11336-016-9530-0 · Cited by 51
“Tables SM-1, SM-2, andSM-3 show the distribution of Ĝ for the alternative criteria as well as that corresponding to the BNPPLM analysis obtained with the optimization method of Dahl (2006), as suggested by Caron et al.”
MethodTurpyn, C. C., Chaplin, T. M., Cook, E. C., & Martelli, A. M. (2015) · Journal of Experimental Child Psychology · doi:10.1016/j.jecp.2015.02.009 · Cited by 51
Symptoms and quality of life indicators among children with chronic medical conditions
Kim, J., Chung, H., Amtmann, D., Salem, R., Park, R., & Askew, R. L. (2013) · Disability and health journal · doi:10.1016/j.dhjo.2013.08.007 · Cited by 59
Kuo, B., & Landgrebe, D. A. (2002) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2002.805088 · Cited by 51
“And ICL-BIC in [13] is n E L BIC ICL log ) ( ) ( 2 ) ( 2 ) ( Ψ + Ψ + Ψ − = Ψ − ν February 17, 2003 It was observed that AIC is order inconsistent and tends to overfit models [12].”
BackgroundExploring Trajectories of Pornography Use Through Adolescence and Emerging Adulthood
Willoughby, B. J., Young-Petersen, B., & Leonhardt, N. D. (2018) · The Journal of Sex Research · doi:10.1080/00224499.2017.1368977 · Cited by 51
“Models with lower AIC, BIC, and entropy values closer to one indicate betterfitting models (Celeux & Soromenho, 1996; Nylund, Asparouhov, & Muthen, 2007).”
MethodFryer, L. K., & Bovee, H. N. (2018) · Computers & Education · doi:10.1016/j.compedu.2018.01.006 · Cited by 64
Kim, S. Y., Hou, Y., Song, J., Schwartz, S. J., Chen, S., Zhang, M., Perreira, K. M., & Parra-Medina, D. M. (2018) · Journal of Youth and Adolescence · doi:10.1007/s10964-018-0851-4 · Cited by 61
“Entropy with values approaching 1 indicate clearer delineation of classes (Celeux and Soromenho 1996).”
BackgroundJago, R., Salway, R., Lawlor, D. A., Emm-Collison, L., Heron, J., Thompson, J. L., & Sebire, S. J. (2018) · International Journal of Behavioral Nutrition and Physical Activity · doi:10.1186/s12966-018-0735-8 · Cited by 56
“We also reported the relative entropy, an overall measure of classification on a scale of 0 (random) to 1 (perfect classification) [39] and the smallest class size, to identify problematic models with very small class sizes.”
BackgroundNa, J., Grossmann, I., Varnum, M. E. W., Karasawa, M., Cho, Y., Kitayama, S., & Nisbett, R. E. (2019) · Journal of Personality · doi:10.1111/jopy.12536 · Cited by 60
Animal welfare information affects consumers’ hedonic and emotional responses towards milk
Jiang, R., Sharma, C., Bryant, R. H., Mohan, M. S., Al-Marashdeh, O., Harrison, R., & Torrico, D. D. (2020) · Food Research International · doi:10.1016/j.foodres.2020.110006 · Cited by 58
Satiety responsiveness and eating behavior among Chilean adolescents and the role of breastfeeding
Reyes, M., Hoyos, V., Martinez, S. M., Lozoff, B., Castillo, M., Burrows, R. A., Blanco, E., & Gahagan, S. (2013) · International Journal of Obesity · doi:10.1038/ijo.2013.191 · Cited by 62
Magee, C., Voyer, M., McIlgorm, A., & Li, O. (2017) · Fisheries Research · doi:10.1016/j.fishres.2017.11.026 · Cited by 59
“Classification accuracy was also inspected (reflected by entropy) along with the size of the identified latent classes (Celeux and Soromenho, 1996).”
BackgroundWillie, T. C., Kershaw, T., & Sullivan, T. P. (2018) · Journal of Interpersonal Violence · doi:10.1177/0886260518802852 · Cited by 55
“In addition, a p value of .05 on the Lo–Mendell–Rubin likelihood ratio test indicated that, compared with the model with one less profile, the present model was a better fit (Celeux & Soromenho, 1996).”
MethodRangel, V. S., Vaval, L., & Bowers, A. J. (2020) · Science Education · doi:10.1002/sce.21593 · Cited by 56
Bereavement-Related Regret Trajectories Among Widowed Older Adults
Holland, J. M., Thompson, K. L., Rozalski, V., & Lichtenthal, W. G. (2013) · The Journals of Gerontology Series B · doi:10.1093/geronb/gbt050 · Cited by 62
Mäkikangas, A., Kinnunen, S. M., Rantanen, J., Mauno, S., Tolvanen, A. J., & Bakker, A. B. (2013) · Anxiety Stress & Coping · doi:10.1080/10615806.2013.860968 · Cited by 58
“…(iii) the Vuong-Lo-Mendel-Rubin (VLMR); (iv) the Lo-Mendell-Rubin (LMR); (v) the Bootstrap Likelihood Ratio Test (BLRT) and (vi) classification quality as determined by entropy (entropy values range from 0 to 1, where values close to 1 indicate a clear classification; Celeux & Soromenho, 1996).”
MethodExploring Symptom Clusters in People With Heart Failure
Park, J., Moser, D. K., Griffith, K. A., Harring, J. R., & Johantgen, M. (2017) · Clinical Nursing Research · doi:10.1177/1054773817729606 · Cited by 56
Nguyen, H., Chua, K., Dregan, A., Vitoratou, S., Bayes‐Marin, I., Olaya, B., & Prina, M. A. (2019) · Journal of Aging and Health · doi:10.1177/0898264319891026 · Cited by 55
“Entropy statistic and average posterior probabilities for each latent class, which showed class membership classification errors, were also taken into account (Celeux & Soromenho, 1996).”
BackgroundBartholomew, T. S., Tookes, H., Bullock, C., Onugha, J., Forrest, D. W., & Feaster, D. J. (2020) · International Journal of Drug Policy · doi:10.1016/j.drugpo.2020.102716 · Cited by 50
Byrd, C. M., & Ahn, L. H. (2020) · Journal of Community Psychology · doi:10.1002/jcop.22393 · Cited by 62
“Entropy represents whether there are clear profiles, with an entropy of over 0.8 indicating distinct classes (Celeux & Soromenho, 1996).”
BackgroundInfluentialIdentifying connected components in Gaussian finite mixture models for clustering
Scrucca, L. (2015) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2015.01.006 · Cited by 49
“Some authors (Celeux and Soromenho, 1996; Ray and 19 Lindsay, 2008) noted the tendency of BIC to underestimate the number of clusters.”
BackgroundSchool Principals’ Stress Profiles During COVID-19, Demands, and Resources
Upadyaya, K., Toyama, H., & Salmela‐Aro, K. (2021) · Frontiers in Psychology · doi:10.3389/fpsyg.2021.731929 · Cited by 70
Racial Identity From Adolescence to Young Adulthood: Does Prior Neighborhood Experience Matter?
Rivas-Drake, D., & Witherspoon, D. (2013) · Child Development · doi:10.1111/cdev.12095 · Cited by 49
Benameur, S., Mignotte, M., Destrempes, F., & de Guise, J. A. (2005) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2005.855717 · Cited by 57
“, by using information theoretic criteria [Akaike information criterion (AIC) or maximum description lenght (MDL)] [17], [18], [25], [26].”
MethodYapan, S., Türkçapar, M. H., & Boysan, M. (2022) · Current Psychology · doi:10.1007/s12144-020-01086-4 · Cited by 48
Factors Influencing School Social Work Practice: A Latent Profile Analysis
Thompson, A. M., Frey, A. J., & Kelly, M. S. (2018) · School Mental Health · doi:10.1007/s12310-018-9279-y · Cited by 53
“Lastly, cohesiveness of each model is assessed using entropy—a statistic with values above .7 indicating persons fit into their identified profile well (Celeux & Soromenho, 1996).”
MethodDunn, K., & Iwaniec, J. (2021) · Studies in Second Language Acquisition · doi:10.1017/s0272263121000759 · Cited by 81
Garden, F. L., Simpson, J. M., Mellis, C. M., & Marks, G. B. (2015) · European Respiratory Journal · doi:10.1183/13993003.00284-2015 · Cited by 56
The cusp catastrophe model as cross-sectional and longitudinal mixture structural equation models.
Chow, S., Witkiewitz, K. A., Grasman, R. P. P. P., & Maisto, S. A. (2015) · Psychological Methods · doi:10.1037/a0038962 · Cited by 52
“In addition, not only was the entropy notably higher in the three-regime model than in the two-regime model, entropy of the two-regime MSEM-RS actually fell below .90, the commonly adopted “rule of thumb” for indicating clarity in class assignment (Celeux & Soromenho, 1996).”
InfluentialUnderstanding and predicting classes of college students who use pornography
Brown, C. C., Durtschi, J. A., Carroll, J. S., & Willoughby, B. J. (2016) · Computers in Human Behavior · doi:10.1016/j.chb.2016.09.008 · Cited by 63
Sleep quality subtypes predict health-related quality of life in children
Magee, C., Robinson, L. D., & Keane, C. A. (2017) · Sleep Medicine · doi:10.1016/j.sleep.2017.04.007 · Cited by 55
Changes in Religious Doubt and Physical and Mental Health in Emerging Adulthood
Upenieks, L. (2021) · Journal for the Scientific Study of Religion · doi:10.1111/jssr.12712 · Cited by 56
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.