similar to: pam() clustering for large data sets

Greetings, I have had good success using the clara() function to perform a simple cluster analysis on a large dataset (1 million+ records with 9 variables). Since the clara function is a wrapper to pam(), which will accept known medoid data - I am wondering if this too is possible with clara() ... The documentation does not suggest that this is possible. Essentially I am trying to

Hello R-community: Much of the time I want to use loops to look at graphs, etc. For example, I have 25 plots, for which the names are m.1$medoids, m.2$medoids, ..., m.25$medoids. I want to index the object number (1:25) as below (just to show concept). for (i in 1:25){ plot(m.i$medoids) } I've tried the following, with negative results for ...

Hello Martin Maechler and All, A simple question (I hope): How can I compute the "sum of the dissimilarities" that appears in the pam command (from the cluster package) ? Is it the "manhattan" distance (such as the one implemented by "dist") ? I am asking since I am running clustering on a dataset. I found 7 medoids with the pam command, and from it I have the

Hola. Yo en vez de utilizar análisis cluster que impliquen distancias, probaría con un kmedias o con un pam (partition around medoids) pero utilizando muestras, la función clara de la librería cluster puede ayudarte. Pego el details de la ayuda de 'clara' Details clara is fully described in chapter 3 of Kaufman and Rousseeuw (1990). Compared to other partitioning methods such as pam,

Dear R helpers, I have a large data set with 36 variables and about 50.000 cases. The variabels represent labour market status during 36 months, there are 8 different variable values (e.g. Full-time Employment, Student,...) Only cases with at least one change in labour market status is included in the data set. To analyse sub sets of the data, I have used daisy in the cluster-package to create

Hello all, I am using the clustering functions in R in order to work with large masses of binary time series data, however the clustering functions do not seem able to fit this size of practical problem. Library 'hclust' is good (though it may be sub par for this size of problem, thus doubly poor for this application) in that I do not want to make assumptions about the number of

El inconveniente con un K-medias, es que se tiene que se tiene que pre definir el número de segmentos, pero eso es algo con lo q no cuento. La solución de Javier me parece q sería la única opción. Atte. Ricardo Alva Valiente -----Mensaje original----- De: R-help-es [mailto:r-help-es-bounces en r-project.org] En nombre de javier.ruben.marcuzzi en gmail.com Enviado el: miércoles, 29 de abril de

Hello all. I wish to run k-means with "manhattan" distance. Since this is not supported by the function "kmeans", I turned to the "pam" function in the "fpc" package. Yet, when I tried to have the algorithm run with different starting points, I found that pam ignores and keep on starting the algorithm from the same starting-points (medoids). For my

Hello I am using k medoids in R to generate sets of clusters for datasets through time. I can plot the individual clusters OK but what I cannot find is a way of pulling out the co-ordinates of the individual points in the cluster diagrams - none of the kmed$... info sets seems to be this. Beneath is an example of a k medoid prog using the built in US arrests dataset - this is not the data I am

Hi, I wanted the DICE coefficient (similarity measure for binary variables) to be calculated in R and found that the "igraph" package has the option of "similarity.dice" to do this. But, for this command, the input object should be an igraph object. But, I have a dataframe of columns containing 1's and 0's. Can I convert this dataframe into an igraph object, so that

HI R users With clara function I get a data frame (maybe this is not the exact word, I'm new to R) with the following variables: > names(myclara) [1] "sample" "medoids" "i.med" "clustering" "objective" [6] "clusinfo" "diss" "call" "silinfo" "data" I want to

Buen aporte?excelente!! Atte. Ricardo Alva Valiente De: Jose Luis Cañadas Reche [mailto:canadasreche en gmail.com] Enviado el: miércoles, 29 de abril de 2015 12:51 PM Para: Alva Valiente, Ricardo (RIAV); 'javier.ruben.marcuzzi en gmail.com'; R-help-es en r-project.org Asunto: Re: [R-es] cantidad de datos Podrías hacer varios kmedias con diferente número de clusters y comprobar como

I am using the PAM algorithm in the CLUSTER library. When I allow PAM to seed the medoids using the default __build__ algorithm things work well: > pam(stats.table, metric="euclidean", stand=TRUE, k=5) But I have some clusters from a Hierarchical analysis that I would like to use as seeds for the PAM algorithm. I can't figure what the mediod argument wants. When I put in the

Hello, When using PAM (partitioning around medoids), I would like to skip the build-step and give the fonction my own medoids. Do you know if it is possible, and how ? Thank you very much. Isabel

Dear all, I have a problem when using some classification functions (Kmeans, PAM, FANNY...) with a distance matrix, and i would to understand how it proceeds for the positioning of centroids after one execution step. In fact, in the classical formulation of the algorithm, after each step, to re-position the center, it calculates the distance between any elements of the old cluster and its

Hello list, I am new to R, so if the question is rather silly, please ignore it. I was wondering wether it would be possible to use the models generated by pam, clara and the like as predictors? Scanning through the available documentation shed no light (for me) upon the subject. Regards, Renald

Hi All, For the function "bclust"(e1071), the argument "base.method" is explained as "must be the name of a partitioning cluster function returning a list with the same components as the return value of 'kmeans'. In my understanding, there are three partitioning cluster functions in R, which are "clara, pam, fanny". Then I check each of them to

Estimados Creo que se puede presentar un problema con el sistema operativo, al ser de 32 bit si no recuerdo mal soporta hasta 4 GB, aunque no estoy del todo seguro. Los 292 GB que informa Carlos son una enormidad, esos requerimientos son complicados. ¿Qué posibilidad hay de trabajar con memoria virtual en windows? Aunque me parece que no sería optimo, prefiero intentar en Linux y R. Su

I am using some functions from package clusterSim to evaluate the best clusters layout. Here is the features vector I am using to cluater 12 signals: > alpha.vec [1] 0.8540039 0.8558350 0.8006592 0.8066406 0.8322754 0.8991699 0.8212891 [8] 0.8815918 0.9050293 0.9174194 0.8613281 0.8425293 In the following I pasted an excerpt of my program:

Dear R users, My data have repeating "beh" parameter : 1 or 2 - type of animal behavior in subsequent locations. I need to assign unique number to each sequence of locations. My data is: >data=data.frame(row=seq(1:10),beh=c(1,1,1,2,2,2,1,1,2,2)) >attach(data) >data row beh 1 1 1 2 2 1 3 3 1 4 4 2 5 5 2 6 6 2 7 7 1 8