Luigi Marongiu
2015-Dec-13 16:17 UTC
[R] define number of clusters in kmeans/apcluster analysis
Dear all, I am trying to do some cluster analysis, both with the base R and the apcluster. Both methods give 2 clusters, which is what I am looking for since I am interested in identifying positive and negative results. However I could not find a way to fine-tuning the analysis in order to properly allocate the points; essentially the negative points should be all those in the lower left portion of the plot (see example) but some in the top centre are also given to the negative cluster. So how can I change the parameters to get better results? Thank you L>>>x <- c(3.15, 3.07, 2, 3, 2.97, 45, 3.21, 45, 40.55, 2, 22.09, 2.47, 2.97, 2.77, 2.6, 7.35, 4.11, 37.12, 2.73, 36.36, 45, 2.33, 2.49, 45, 2.4, 2.74, 2.64, 45, 2.47, 38.1, 2.47, 37.4, 2.77, 2.37, 45, 2.69, 2.97, 2.7, 2, 2, 2.55, 11.86, 2.51, 2.68, 2.31, 2.6, 2.45, 2, 2.72, 2.57, 2.09, 3.04, 45, 45, 2.13, 43.82, 2.92, 4.94, 24.82, 2.64, 4.96, 3.65, 2.67, 2.64, 8.04, 4.56, 44.87, 37.42, 45, 6.2, 2.84, 4.08, 2, 5.03, 2.27, 44.89, 2.41, 2.47, 2.78, 37.47, 45, 2.76, 45, 2.51, 2.8, 44.8, 6.2, 2.87, 2.23, 18.32, 3.14, 2.1, 2.38, 2.72, 2, 2, 44.41, 3.15, 3.06, 4.8, 2.77, 2.8, 2.71, 44.77, 2.25, 2.69, 28.38, 2, 2.95, 45, 2.79, 2.46, 2.61, 2.78, 2.94, 38.47, 3.29, 2.89, 2.4, 2.23, 2.62, 4.21, 2.61, 2.81, 2.41, 41.98, 2.39, 36.41, 44.84, 4.73, 2, 2.66, 4.57, 3.01, 42.64, 2.04, 5.49, 15.48, 3.08, 2.7, 2, 2, 2.09, 2, 2.29, 2.92, 3.39, 3.1, 2, 6.14, 7.03, 4.77, 2.55, 32.36, 20.61, 3.09, 4.46, 44.75, 2, 2.73, 2, 36.05, 3.61, 34.84, 2.69, 5.28, 3.04, 45, 2.47, 2.58, 2.16, 2.59, 45, 44.08, 2, 37.05, 2.48, 2.46, 38.71, 7.32, 2.95, 2.8, 44.58, 42.24, 36.99, 13.84, 45, 2, 2, 2.38, 45, 45, 43.59, 2.69, 2.81, 3.05, 2.8, 4.65, 45, 41.46, 2.33, 7.12, 19.18, 4.82, 4.76, 2.51, 3.1, 2.74, 4.99, 38.06, 2.53, 2.94, 2.93, 6.59, 2.72, 2.94, 2.56, 2.91, 44.79, 2.98, 42.95, 45, 2.63, 38.44, 2.71, 2, 37.92, 2.69, 2.91, 2.65, 44.48, 6.35, 2.56, 21.94, 3.08, 2.6, 45, 2, 2.62, 2.47, 2.62, 2.73, 2.87, 2.83, 4.56, 44.22, 5.15, 5.13, 2.76, 7.02, 28.61, 4.87, 5.02, 44.35, 2.26, 2.89, 5.26, 38.01, 44.79, 39.26, 2.91, 4.59, 2.69, 2.61, 34.97, 3, 45, 2.81, 2, 2.65, 2, 37.33, 4.69, 3.26, 38.24, 4.97, 4.62, 2.47, 45, 4.52, 2.73, 15.66, 6.06, 2.79, 2.87, 45, 45, 45, 4.84, 3.05, 4.89, 4.64, 4.92, 2.74, 7.83, 42.31, 2.88, 6.89, 23.06, 2.94, 4.72, 4.55, 5.52, 4.48, 4.86, 3.12, 7.68, 43.89, 2.82, 2.64, 3.05, 42.95, 2.33, 3.55, 45, 2.79, 2.47, 45, 2.56, 38.33, 2.73, 2.87, 2.61, 3.01, 2.86, 2.74, 44.46, 44.54, 2.62, 16.94, 2.53, 2.24, 2.72, 2, 3.1, 2.88, 7.4, 4.64, 8.25, 3.01, 2.86, 2.46, 5.67, 44.52, 2.47, 2, 29.01, 2.61, 3.23, 12.3, 3.9, 2.91, 43.99, 36.99, 43.72, 42.29, 2.63, 3.03, 2.85, 2.58, 2.63, 2.73, 2.57, 2.37, 2.57, 2.75, 44.14, 39.4, 40.02, 3.08, 45, 4.96, 3, 2.83, 2.74, 2.8, 2.8, 18.88, 4.69, 2.51, 4.32, 2, 2.56, 2.81 ) y <- c(0.014, 0.04, 0.001, 0.023, 0.008, 0, 0.008, 0.001, -0.001, 0.002, 0.103, 0, 0.013, 0.005, 0.008, 0.001, 0.011, 0.076, 0.005, 0.045, -0.001, 0, 0.008, -0.002, 0.002, 0.016, 0.006, 0.001, 0.002, 0.001, 0.004, 0.086, 0.009, 0.011, 0.002, 0.013, 0.019, 0.007, 0, 0.002, 0.024, 0.119, 0.015, 0.009, 0.013, 0.017, 0.009, 0.009, 0.006, 0.012, 0.002, 0.015, 0, 0.001, 0.002, 0.001, 0.007, 0.004, 0.113, 0.016, 0.013, 0.004, 0.015, 0.005, 0.004, 0.007, 0, 0.081, 0.001, 0.002, 0.014, 0.002, 0, 0.01, 0.003, 0.002, 0.004, 0.004, 0.006, 0.064, 0, 0.014, 0, 0.01, 0.019, 0.002, 0.006, 0.005, 0.003, 0.103, 0.007, 0.008, 0.002, 0.013, 0.007, 0.004, 0.001, 0.04, 0.017, 0.018, 0.002, 0.006, 0.011, 0.003, 0.004, 0.008, 0.115, 0, 0.02, 0, 0.012, 0.009, 0.011, 0.013, 0.004, 0.058, 0.019, 0.006, 0.005, 0.004, 0.012, 0.003, 0.003, 0.004, 0.002, 0.001, 0.002, 0.102, -0.001, 0.008, 0.002, 0.016, 0.023, 0.014, 0.053, 0.009, 0.001, 0.124, 0.009, 0.008, 0.002, 0.002, 0.013, 0.002, 0.001, 0.042, 0.011, 0.009, 0, 0.004, 0.003, 0.002, 0.005, 0, 0.101, 0.013, 0.009, 0.005, 0.002, 0.007, 0.008, 0.067, 0.002, 0.064, 0.028, 0.007, 0.006, 0, 0.007, 0.006, 0, 0.001, 0.001, 0.001, 0, 0.088, 0.005, 0.008, 0.098, 0.005, 0.019, 0.007, 0.05, -0.002, 0.002, 0.129, 0.001, 0.004, -0.001, 0.002, -0.001, 0, 0.043, 0.018, 0.019, 0.015, 0.003, 0.006, 0.002, 0.001, 0.002, 0.004, 0.097, 0.025, 0.022, 0.007, 0.011, 0.007, 0.013, 0.061, 0.008, 0.013, 0.028, 0.004, 0.013, 0.005, 0.01, 0.004, 0, 0.006, -0.001, 0.001, 0.01, 0.061, 0.002, 0.004, 0, 0.011, 0.029, 0.018, 0, 0.003, 0.012, 0.085, 0.015, 0.007, 0.002, 0.003, 0.008, 0.002, 0.007, 0.02, 0.011, 0.02, 0.008, 0.001, 0.003, 0.01, 0.014, 0.001, 0.096, 0.027, 0.024, 0, 0.005, 0.006, 0.024, 0.087, 0.001, 0.083, 0.02, 0.009, 0.009, 0.001, 0, 0.019, 0, 0.003, -0.001, 0.002, 0, 0.089, 0.016, 0.01, 0.103, 0.003, 0.01, 0.002, 0.008, 0.005, 0.014, 0.1, 0.007, 0.009, 0.011, -0.001, 0, 0.002, 0.015, 0.036, 0.018, 0.026, 0.009, 0.008, 0.004, 0.001, 0.014, 0.009, 0.1, 0.026, 0.032, 0.008, 0.011, 0.004, 0.013, 0.019, 0.004, 0.02, 0.015, 0.005, 0.013, -0.001, 0.013, 0.012, 0, 0.01, 0.002, 0.001, 0.013, 0.066, 0.009, 0.005, 0.002, 0.013, 0.025, 0.006, 0, 0, 0.015, 0.121, 0.006, 0.003, 0.008, 0, 0.012, 0.011, 0.003, 0.022, 0.008, 0.032, 0.007, 0.002, 0.006, 0.007, 0, 0.003, 0.11, 0.01, 0.008, 0, 0.018, 0.008, 0.001, 0.087, 0, 0.028, 0.011, 0.014, 0.007, 0.001, 0.018, 0.033, 0.021, 0.003, 0.003, 0.007, -0.001, 0.07, 0.022, 0.009, 0.001, 0.007, 0.031, 0.008, 0.013, 0.01, 0.018, 0.125, 0.01, 0.015, 0.006, 0, 0.015, 0.019 ) z <- cbind(x, y) k <- kmeans(z, 2) plot(z, col=k$cluster) library(apcluster) m <- apclusterK(negDistMat(r=2), z, K=2, verbose=TRUE) plot(m, z)
Bert Gunter
2015-Dec-13 16:28 UTC
[R] define number of clusters in kmeans/apcluster analysis
It sounds to me like you don't understand cluster analysis. You should not expect perfect "allocation" of points. I suggest that you consult references in the man pages of your functions or on the web. You might also find it useful to post on stats.stackexchange.com or a machine learning help site, as this is a statistical issue, not an R programming issue AFAICS (corrections welcome if I'm wrong about this), and so is somewhat OT here. Cheers, Bert Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Sun, Dec 13, 2015 at 8:17 AM, Luigi Marongiu <marongiu.luigi at gmail.com> wrote:> Dear all, > I am trying to do some cluster analysis, both with the base R and the > apcluster. Both methods give 2 clusters, which is what I am looking > for since I am interested in identifying positive and negative > results. However I could not find a way to fine-tuning the analysis > in order to properly allocate the points; essentially the negative > points should be all those in the lower left portion of the plot (see > example) but some in the top centre are also given to the negative > cluster. > So how can I change the parameters to get better results? > Thank you > L > >>>> > x <- c(3.15, 3.07, 2, 3, 2.97, 45, 3.21, 45, > 40.55, 2, 22.09, 2.47, 2.97, 2.77, 2.6, 7.35, > 4.11, 37.12, 2.73, 36.36, 45, 2.33, 2.49, 45, > 2.4, 2.74, 2.64, 45, 2.47, 38.1, 2.47, 37.4, > 2.77, 2.37, 45, 2.69, 2.97, 2.7, 2, 2, 2.55, > 11.86, 2.51, 2.68, 2.31, 2.6, 2.45, 2, 2.72, > 2.57, 2.09, 3.04, 45, 45, 2.13, 43.82, 2.92, > 4.94, 24.82, 2.64, 4.96, 3.65, 2.67, 2.64, 8.04, > 4.56, 44.87, 37.42, 45, 6.2, 2.84, 4.08, 2, > 5.03, 2.27, 44.89, 2.41, 2.47, 2.78, 37.47, 45, > 2.76, 45, 2.51, 2.8, 44.8, 6.2, 2.87, 2.23, > 18.32, 3.14, 2.1, 2.38, 2.72, 2, 2, 44.41, > 3.15, 3.06, 4.8, 2.77, 2.8, 2.71, 44.77, 2.25, > 2.69, 28.38, 2, 2.95, 45, 2.79, 2.46, 2.61, > 2.78, 2.94, 38.47, 3.29, 2.89, 2.4, 2.23, 2.62, > 4.21, 2.61, 2.81, 2.41, 41.98, 2.39, 36.41, > 44.84, 4.73, 2, 2.66, 4.57, 3.01, 42.64, 2.04, > 5.49, 15.48, 3.08, 2.7, 2, 2, 2.09, 2, 2.29, > 2.92, 3.39, 3.1, 2, 6.14, 7.03, 4.77, 2.55, > 32.36, 20.61, 3.09, 4.46, 44.75, 2, 2.73, 2, > 36.05, 3.61, 34.84, 2.69, 5.28, 3.04, 45, 2.47, > 2.58, 2.16, 2.59, 45, 44.08, 2, 37.05, 2.48, > 2.46, 38.71, 7.32, 2.95, 2.8, 44.58, 42.24, > 36.99, 13.84, 45, 2, 2, 2.38, 45, 45, 43.59, > 2.69, 2.81, 3.05, 2.8, 4.65, 45, 41.46, 2.33, > 7.12, 19.18, 4.82, 4.76, 2.51, 3.1, 2.74, 4.99, > 38.06, 2.53, 2.94, 2.93, 6.59, 2.72, 2.94, 2.56, > 2.91, 44.79, 2.98, 42.95, 45, 2.63, 38.44, > 2.71, 2, 37.92, 2.69, 2.91, 2.65, 44.48, 6.35, > 2.56, 21.94, 3.08, 2.6, 45, 2, 2.62, 2.47, > 2.62, 2.73, 2.87, 2.83, 4.56, 44.22, 5.15, 5.13, > 2.76, 7.02, 28.61, 4.87, 5.02, 44.35, 2.26, > 2.89, 5.26, 38.01, 44.79, 39.26, 2.91, 4.59, > 2.69, 2.61, 34.97, 3, 45, 2.81, 2, 2.65, 2, > 37.33, 4.69, 3.26, 38.24, 4.97, 4.62, 2.47, 45, > 4.52, 2.73, 15.66, 6.06, 2.79, 2.87, 45, 45, > 45, 4.84, 3.05, 4.89, 4.64, 4.92, 2.74, 7.83, > 42.31, 2.88, 6.89, 23.06, 2.94, 4.72, 4.55, 5.52, > 4.48, 4.86, 3.12, 7.68, 43.89, 2.82, 2.64, > 3.05, 42.95, 2.33, 3.55, 45, 2.79, 2.47, 45, > 2.56, 38.33, 2.73, 2.87, 2.61, 3.01, 2.86, 2.74, > 44.46, 44.54, 2.62, 16.94, 2.53, 2.24, 2.72, 2, > 3.1, 2.88, 7.4, 4.64, 8.25, 3.01, 2.86, 2.46, > 5.67, 44.52, 2.47, 2, 29.01, 2.61, 3.23, 12.3, > 3.9, 2.91, 43.99, 36.99, 43.72, 42.29, 2.63, > 3.03, 2.85, 2.58, 2.63, 2.73, 2.57, 2.37, 2.57, > 2.75, 44.14, 39.4, 40.02, 3.08, 45, 4.96, 3, > 2.83, 2.74, 2.8, 2.8, 18.88, 4.69, 2.51, 4.32, > 2, 2.56, 2.81 > ) > y <- c(0.014, 0.04, 0.001, 0.023, 0.008, 0, 0.008, > 0.001, -0.001, 0.002, 0.103, 0, 0.013, 0.005, > 0.008, 0.001, 0.011, 0.076, 0.005, 0.045, -0.001, > 0, 0.008, -0.002, 0.002, 0.016, 0.006, 0.001, > 0.002, 0.001, 0.004, 0.086, 0.009, 0.011, 0.002, > 0.013, 0.019, 0.007, 0, 0.002, 0.024, 0.119, > 0.015, 0.009, 0.013, 0.017, 0.009, 0.009, 0.006, > 0.012, 0.002, 0.015, 0, 0.001, 0.002, 0.001, > 0.007, 0.004, 0.113, 0.016, 0.013, 0.004, 0.015, > 0.005, 0.004, 0.007, 0, 0.081, 0.001, 0.002, > 0.014, 0.002, 0, 0.01, 0.003, 0.002, 0.004, > 0.004, 0.006, 0.064, 0, 0.014, 0, 0.01, 0.019, > 0.002, 0.006, 0.005, 0.003, 0.103, 0.007, 0.008, > 0.002, 0.013, 0.007, 0.004, 0.001, 0.04, 0.017, > 0.018, 0.002, 0.006, 0.011, 0.003, 0.004, 0.008, > 0.115, 0, 0.02, 0, 0.012, 0.009, 0.011, 0.013, > 0.004, 0.058, 0.019, 0.006, 0.005, 0.004, 0.012, > 0.003, 0.003, 0.004, 0.002, 0.001, 0.002, 0.102, > -0.001, 0.008, 0.002, 0.016, 0.023, 0.014, 0.053, > 0.009, 0.001, 0.124, 0.009, 0.008, 0.002, 0.002, > 0.013, 0.002, 0.001, 0.042, 0.011, 0.009, 0, > 0.004, 0.003, 0.002, 0.005, 0, 0.101, 0.013, > 0.009, 0.005, 0.002, 0.007, 0.008, 0.067, 0.002, > 0.064, 0.028, 0.007, 0.006, 0, 0.007, 0.006, 0, > 0.001, 0.001, 0.001, 0, 0.088, 0.005, 0.008, > 0.098, 0.005, 0.019, 0.007, 0.05, -0.002, 0.002, > 0.129, 0.001, 0.004, -0.001, 0.002, -0.001, 0, > 0.043, 0.018, 0.019, 0.015, 0.003, 0.006, 0.002, > 0.001, 0.002, 0.004, 0.097, 0.025, 0.022, 0.007, > 0.011, 0.007, 0.013, 0.061, 0.008, 0.013, 0.028, > 0.004, 0.013, 0.005, 0.01, 0.004, 0, 0.006, > -0.001, 0.001, 0.01, 0.061, 0.002, 0.004, 0, > 0.011, 0.029, 0.018, 0, 0.003, 0.012, 0.085, > 0.015, 0.007, 0.002, 0.003, 0.008, 0.002, 0.007, > 0.02, 0.011, 0.02, 0.008, 0.001, 0.003, 0.01, > 0.014, 0.001, 0.096, 0.027, 0.024, 0, 0.005, > 0.006, 0.024, 0.087, 0.001, 0.083, 0.02, 0.009, > 0.009, 0.001, 0, 0.019, 0, 0.003, -0.001, 0.002, > 0, 0.089, 0.016, 0.01, 0.103, 0.003, 0.01, > 0.002, 0.008, 0.005, 0.014, 0.1, 0.007, 0.009, > 0.011, -0.001, 0, 0.002, 0.015, 0.036, 0.018, > 0.026, 0.009, 0.008, 0.004, 0.001, 0.014, 0.009, > 0.1, 0.026, 0.032, 0.008, 0.011, 0.004, 0.013, > 0.019, 0.004, 0.02, 0.015, 0.005, 0.013, -0.001, > 0.013, 0.012, 0, 0.01, 0.002, 0.001, 0.013, > 0.066, 0.009, 0.005, 0.002, 0.013, 0.025, 0.006, > 0, 0, 0.015, 0.121, 0.006, 0.003, 0.008, 0, > 0.012, 0.011, 0.003, 0.022, 0.008, 0.032, 0.007, > 0.002, 0.006, 0.007, 0, 0.003, 0.11, 0.01, 0.008, > 0, 0.018, 0.008, 0.001, 0.087, 0, 0.028, > 0.011, 0.014, 0.007, 0.001, 0.018, 0.033, 0.021, > 0.003, 0.003, 0.007, -0.001, 0.07, 0.022, 0.009, > 0.001, 0.007, 0.031, 0.008, 0.013, 0.01, 0.018, > 0.125, 0.01, 0.015, 0.006, 0, 0.015, 0.019 > ) > z <- cbind(x, y) > k <- kmeans(z, 2) > plot(z, col=k$cluster) > > library(apcluster) > m <- apclusterK(negDistMat(r=2), z, K=2, verbose=TRUE) > plot(m, z) > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
Boris Steipe
2015-Dec-13 17:04 UTC
[R] define number of clusters in kmeans/apcluster analysis
You could use the clValid package to run your problem through a variety of different algorithms and evaluate cluster quality, you will learn a lot. https://cran.r-project.org/web/views/Cluster.html gives you many more options. All of the algorithms I am aware of have tunable parameters - but what is the "best" approach really depends on the detailed context of the problem domain. Sometimes dimension reduction with PCA, or even the magical tsne package is useful for preprocessing data. Finally, if you already know how your data set should be partitioned, perhaps you are really looking for a machine learning approach: see here https://cran.r-project.org/web/views/MachineLearning.html B. On Dec 13, 2015, at 11:17 AM, Luigi Marongiu <marongiu.luigi at gmail.com> wrote:> Dear all, > I am trying to do some cluster analysis, both with the base R and the > apcluster. Both methods give 2 clusters, which is what I am looking > for since I am interested in identifying positive and negative > results. However I could not find a way to fine-tuning the analysis > in order to properly allocate the points; essentially the negative > points should be all those in the lower left portion of the plot (see > example) but some in the top centre are also given to the negative > cluster. > So how can I change the parameters to get better results? > Thank you > L > >>>> > x <- c(3.15, 3.07, 2, 3, 2.97, 45, 3.21, 45, > 40.55, 2, 22.09, 2.47, 2.97, 2.77, 2.6, 7.35, > 4.11, 37.12, 2.73, 36.36, 45, 2.33, 2.49, 45, > 2.4, 2.74, 2.64, 45, 2.47, 38.1, 2.47, 37.4, > 2.77, 2.37, 45, 2.69, 2.97, 2.7, 2, 2, 2.55, > 11.86, 2.51, 2.68, 2.31, 2.6, 2.45, 2, 2.72, > 2.57, 2.09, 3.04, 45, 45, 2.13, 43.82, 2.92, > 4.94, 24.82, 2.64, 4.96, 3.65, 2.67, 2.64, 8.04, > 4.56, 44.87, 37.42, 45, 6.2, 2.84, 4.08, 2, > 5.03, 2.27, 44.89, 2.41, 2.47, 2.78, 37.47, 45, > 2.76, 45, 2.51, 2.8, 44.8, 6.2, 2.87, 2.23, > 18.32, 3.14, 2.1, 2.38, 2.72, 2, 2, 44.41, > 3.15, 3.06, 4.8, 2.77, 2.8, 2.71, 44.77, 2.25, > 2.69, 28.38, 2, 2.95, 45, 2.79, 2.46, 2.61, > 2.78, 2.94, 38.47, 3.29, 2.89, 2.4, 2.23, 2.62, > 4.21, 2.61, 2.81, 2.41, 41.98, 2.39, 36.41, > 44.84, 4.73, 2, 2.66, 4.57, 3.01, 42.64, 2.04, > 5.49, 15.48, 3.08, 2.7, 2, 2, 2.09, 2, 2.29, > 2.92, 3.39, 3.1, 2, 6.14, 7.03, 4.77, 2.55, > 32.36, 20.61, 3.09, 4.46, 44.75, 2, 2.73, 2, > 36.05, 3.61, 34.84, 2.69, 5.28, 3.04, 45, 2.47, > 2.58, 2.16, 2.59, 45, 44.08, 2, 37.05, 2.48, > 2.46, 38.71, 7.32, 2.95, 2.8, 44.58, 42.24, > 36.99, 13.84, 45, 2, 2, 2.38, 45, 45, 43.59, > 2.69, 2.81, 3.05, 2.8, 4.65, 45, 41.46, 2.33, > 7.12, 19.18, 4.82, 4.76, 2.51, 3.1, 2.74, 4.99, > 38.06, 2.53, 2.94, 2.93, 6.59, 2.72, 2.94, 2.56, > 2.91, 44.79, 2.98, 42.95, 45, 2.63, 38.44, > 2.71, 2, 37.92, 2.69, 2.91, 2.65, 44.48, 6.35, > 2.56, 21.94, 3.08, 2.6, 45, 2, 2.62, 2.47, > 2.62, 2.73, 2.87, 2.83, 4.56, 44.22, 5.15, 5.13, > 2.76, 7.02, 28.61, 4.87, 5.02, 44.35, 2.26, > 2.89, 5.26, 38.01, 44.79, 39.26, 2.91, 4.59, > 2.69, 2.61, 34.97, 3, 45, 2.81, 2, 2.65, 2, > 37.33, 4.69, 3.26, 38.24, 4.97, 4.62, 2.47, 45, > 4.52, 2.73, 15.66, 6.06, 2.79, 2.87, 45, 45, > 45, 4.84, 3.05, 4.89, 4.64, 4.92, 2.74, 7.83, > 42.31, 2.88, 6.89, 23.06, 2.94, 4.72, 4.55, 5.52, > 4.48, 4.86, 3.12, 7.68, 43.89, 2.82, 2.64, > 3.05, 42.95, 2.33, 3.55, 45, 2.79, 2.47, 45, > 2.56, 38.33, 2.73, 2.87, 2.61, 3.01, 2.86, 2.74, > 44.46, 44.54, 2.62, 16.94, 2.53, 2.24, 2.72, 2, > 3.1, 2.88, 7.4, 4.64, 8.25, 3.01, 2.86, 2.46, > 5.67, 44.52, 2.47, 2, 29.01, 2.61, 3.23, 12.3, > 3.9, 2.91, 43.99, 36.99, 43.72, 42.29, 2.63, > 3.03, 2.85, 2.58, 2.63, 2.73, 2.57, 2.37, 2.57, > 2.75, 44.14, 39.4, 40.02, 3.08, 45, 4.96, 3, > 2.83, 2.74, 2.8, 2.8, 18.88, 4.69, 2.51, 4.32, > 2, 2.56, 2.81 > ) > y <- c(0.014, 0.04, 0.001, 0.023, 0.008, 0, 0.008, > 0.001, -0.001, 0.002, 0.103, 0, 0.013, 0.005, > 0.008, 0.001, 0.011, 0.076, 0.005, 0.045, -0.001, > 0, 0.008, -0.002, 0.002, 0.016, 0.006, 0.001, > 0.002, 0.001, 0.004, 0.086, 0.009, 0.011, 0.002, > 0.013, 0.019, 0.007, 0, 0.002, 0.024, 0.119, > 0.015, 0.009, 0.013, 0.017, 0.009, 0.009, 0.006, > 0.012, 0.002, 0.015, 0, 0.001, 0.002, 0.001, > 0.007, 0.004, 0.113, 0.016, 0.013, 0.004, 0.015, > 0.005, 0.004, 0.007, 0, 0.081, 0.001, 0.002, > 0.014, 0.002, 0, 0.01, 0.003, 0.002, 0.004, > 0.004, 0.006, 0.064, 0, 0.014, 0, 0.01, 0.019, > 0.002, 0.006, 0.005, 0.003, 0.103, 0.007, 0.008, > 0.002, 0.013, 0.007, 0.004, 0.001, 0.04, 0.017, > 0.018, 0.002, 0.006, 0.011, 0.003, 0.004, 0.008, > 0.115, 0, 0.02, 0, 0.012, 0.009, 0.011, 0.013, > 0.004, 0.058, 0.019, 0.006, 0.005, 0.004, 0.012, > 0.003, 0.003, 0.004, 0.002, 0.001, 0.002, 0.102, > -0.001, 0.008, 0.002, 0.016, 0.023, 0.014, 0.053, > 0.009, 0.001, 0.124, 0.009, 0.008, 0.002, 0.002, > 0.013, 0.002, 0.001, 0.042, 0.011, 0.009, 0, > 0.004, 0.003, 0.002, 0.005, 0, 0.101, 0.013, > 0.009, 0.005, 0.002, 0.007, 0.008, 0.067, 0.002, > 0.064, 0.028, 0.007, 0.006, 0, 0.007, 0.006, 0, > 0.001, 0.001, 0.001, 0, 0.088, 0.005, 0.008, > 0.098, 0.005, 0.019, 0.007, 0.05, -0.002, 0.002, > 0.129, 0.001, 0.004, -0.001, 0.002, -0.001, 0, > 0.043, 0.018, 0.019, 0.015, 0.003, 0.006, 0.002, > 0.001, 0.002, 0.004, 0.097, 0.025, 0.022, 0.007, > 0.011, 0.007, 0.013, 0.061, 0.008, 0.013, 0.028, > 0.004, 0.013, 0.005, 0.01, 0.004, 0, 0.006, > -0.001, 0.001, 0.01, 0.061, 0.002, 0.004, 0, > 0.011, 0.029, 0.018, 0, 0.003, 0.012, 0.085, > 0.015, 0.007, 0.002, 0.003, 0.008, 0.002, 0.007, > 0.02, 0.011, 0.02, 0.008, 0.001, 0.003, 0.01, > 0.014, 0.001, 0.096, 0.027, 0.024, 0, 0.005, > 0.006, 0.024, 0.087, 0.001, 0.083, 0.02, 0.009, > 0.009, 0.001, 0, 0.019, 0, 0.003, -0.001, 0.002, > 0, 0.089, 0.016, 0.01, 0.103, 0.003, 0.01, > 0.002, 0.008, 0.005, 0.014, 0.1, 0.007, 0.009, > 0.011, -0.001, 0, 0.002, 0.015, 0.036, 0.018, > 0.026, 0.009, 0.008, 0.004, 0.001, 0.014, 0.009, > 0.1, 0.026, 0.032, 0.008, 0.011, 0.004, 0.013, > 0.019, 0.004, 0.02, 0.015, 0.005, 0.013, -0.001, > 0.013, 0.012, 0, 0.01, 0.002, 0.001, 0.013, > 0.066, 0.009, 0.005, 0.002, 0.013, 0.025, 0.006, > 0, 0, 0.015, 0.121, 0.006, 0.003, 0.008, 0, > 0.012, 0.011, 0.003, 0.022, 0.008, 0.032, 0.007, > 0.002, 0.006, 0.007, 0, 0.003, 0.11, 0.01, 0.008, > 0, 0.018, 0.008, 0.001, 0.087, 0, 0.028, > 0.011, 0.014, 0.007, 0.001, 0.018, 0.033, 0.021, > 0.003, 0.003, 0.007, -0.001, 0.07, 0.022, 0.009, > 0.001, 0.007, 0.031, 0.008, 0.013, 0.01, 0.018, > 0.125, 0.01, 0.015, 0.006, 0, 0.015, 0.019 > ) > z <- cbind(x, y) > k <- kmeans(z, 2) > plot(z, col=k$cluster) > > library(apcluster) > m <- apclusterK(negDistMat(r=2), z, K=2, verbose=TRUE) > plot(m, z) > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
William Dunlap
2015-Dec-13 18:28 UTC
[R] define number of clusters in kmeans/apcluster analysis
In addition to the other fine replies, you should also know that
kmeans's results
depend on the relative scales of the data columns (since it is based
on distances
between points). Your x and y have quite different scales so the distance is
essentially determined only by the differences in the variable with
the larger scale.
E.g., look at the difference in clustering when x is scaled by
different amounts:
par(mfrow=c(2,2), mar=c(3,3,0,0))
for(i in 1:4){k <- kmeans(zs<-cbind(x/(i*5),y/.05), 2); plot(zs,
col=k$cluster, asp=1)}
Bill Dunlap
TIBCO Software
wdunlap tibco.com
On Sun, Dec 13, 2015 at 8:17 AM, Luigi Marongiu
<marongiu.luigi at gmail.com> wrote:> Dear all,
> I am trying to do some cluster analysis, both with the base R and the
> apcluster. Both methods give 2 clusters, which is what I am looking
> for since I am interested in identifying positive and negative
> results. However I could not find a way to fine-tuning the analysis
> in order to properly allocate the points; essentially the negative
> points should be all those in the lower left portion of the plot (see
> example) but some in the top centre are also given to the negative
> cluster.
> So how can I change the parameters to get better results?
> Thank you
> L
>
>>>>
> x <- c(3.15, 3.07, 2, 3, 2.97, 45, 3.21, 45,
> 40.55, 2, 22.09, 2.47, 2.97, 2.77, 2.6, 7.35,
> 4.11, 37.12, 2.73, 36.36, 45, 2.33, 2.49, 45,
> 2.4, 2.74, 2.64, 45, 2.47, 38.1, 2.47, 37.4,
> 2.77, 2.37, 45, 2.69, 2.97, 2.7, 2, 2, 2.55,
> 11.86, 2.51, 2.68, 2.31, 2.6, 2.45, 2, 2.72,
> 2.57, 2.09, 3.04, 45, 45, 2.13, 43.82, 2.92,
> 4.94, 24.82, 2.64, 4.96, 3.65, 2.67, 2.64, 8.04,
> 4.56, 44.87, 37.42, 45, 6.2, 2.84, 4.08, 2,
> 5.03, 2.27, 44.89, 2.41, 2.47, 2.78, 37.47, 45,
> 2.76, 45, 2.51, 2.8, 44.8, 6.2, 2.87, 2.23,
> 18.32, 3.14, 2.1, 2.38, 2.72, 2, 2, 44.41,
> 3.15, 3.06, 4.8, 2.77, 2.8, 2.71, 44.77, 2.25,
> 2.69, 28.38, 2, 2.95, 45, 2.79, 2.46, 2.61,
> 2.78, 2.94, 38.47, 3.29, 2.89, 2.4, 2.23, 2.62,
> 4.21, 2.61, 2.81, 2.41, 41.98, 2.39, 36.41,
> 44.84, 4.73, 2, 2.66, 4.57, 3.01, 42.64, 2.04,
> 5.49, 15.48, 3.08, 2.7, 2, 2, 2.09, 2, 2.29,
> 2.92, 3.39, 3.1, 2, 6.14, 7.03, 4.77, 2.55,
> 32.36, 20.61, 3.09, 4.46, 44.75, 2, 2.73, 2,
> 36.05, 3.61, 34.84, 2.69, 5.28, 3.04, 45, 2.47,
> 2.58, 2.16, 2.59, 45, 44.08, 2, 37.05, 2.48,
> 2.46, 38.71, 7.32, 2.95, 2.8, 44.58, 42.24,
> 36.99, 13.84, 45, 2, 2, 2.38, 45, 45, 43.59,
> 2.69, 2.81, 3.05, 2.8, 4.65, 45, 41.46, 2.33,
> 7.12, 19.18, 4.82, 4.76, 2.51, 3.1, 2.74, 4.99,
> 38.06, 2.53, 2.94, 2.93, 6.59, 2.72, 2.94, 2.56,
> 2.91, 44.79, 2.98, 42.95, 45, 2.63, 38.44,
> 2.71, 2, 37.92, 2.69, 2.91, 2.65, 44.48, 6.35,
> 2.56, 21.94, 3.08, 2.6, 45, 2, 2.62, 2.47,
> 2.62, 2.73, 2.87, 2.83, 4.56, 44.22, 5.15, 5.13,
> 2.76, 7.02, 28.61, 4.87, 5.02, 44.35, 2.26,
> 2.89, 5.26, 38.01, 44.79, 39.26, 2.91, 4.59,
> 2.69, 2.61, 34.97, 3, 45, 2.81, 2, 2.65, 2,
> 37.33, 4.69, 3.26, 38.24, 4.97, 4.62, 2.47, 45,
> 4.52, 2.73, 15.66, 6.06, 2.79, 2.87, 45, 45,
> 45, 4.84, 3.05, 4.89, 4.64, 4.92, 2.74, 7.83,
> 42.31, 2.88, 6.89, 23.06, 2.94, 4.72, 4.55, 5.52,
> 4.48, 4.86, 3.12, 7.68, 43.89, 2.82, 2.64,
> 3.05, 42.95, 2.33, 3.55, 45, 2.79, 2.47, 45,
> 2.56, 38.33, 2.73, 2.87, 2.61, 3.01, 2.86, 2.74,
> 44.46, 44.54, 2.62, 16.94, 2.53, 2.24, 2.72, 2,
> 3.1, 2.88, 7.4, 4.64, 8.25, 3.01, 2.86, 2.46,
> 5.67, 44.52, 2.47, 2, 29.01, 2.61, 3.23, 12.3,
> 3.9, 2.91, 43.99, 36.99, 43.72, 42.29, 2.63,
> 3.03, 2.85, 2.58, 2.63, 2.73, 2.57, 2.37, 2.57,
> 2.75, 44.14, 39.4, 40.02, 3.08, 45, 4.96, 3,
> 2.83, 2.74, 2.8, 2.8, 18.88, 4.69, 2.51, 4.32,
> 2, 2.56, 2.81
> )
> y <- c(0.014, 0.04, 0.001, 0.023, 0.008, 0, 0.008,
> 0.001, -0.001, 0.002, 0.103, 0, 0.013, 0.005,
> 0.008, 0.001, 0.011, 0.076, 0.005, 0.045, -0.001,
> 0, 0.008, -0.002, 0.002, 0.016, 0.006, 0.001,
> 0.002, 0.001, 0.004, 0.086, 0.009, 0.011, 0.002,
> 0.013, 0.019, 0.007, 0, 0.002, 0.024, 0.119,
> 0.015, 0.009, 0.013, 0.017, 0.009, 0.009, 0.006,
> 0.012, 0.002, 0.015, 0, 0.001, 0.002, 0.001,
> 0.007, 0.004, 0.113, 0.016, 0.013, 0.004, 0.015,
> 0.005, 0.004, 0.007, 0, 0.081, 0.001, 0.002,
> 0.014, 0.002, 0, 0.01, 0.003, 0.002, 0.004,
> 0.004, 0.006, 0.064, 0, 0.014, 0, 0.01, 0.019,
> 0.002, 0.006, 0.005, 0.003, 0.103, 0.007, 0.008,
> 0.002, 0.013, 0.007, 0.004, 0.001, 0.04, 0.017,
> 0.018, 0.002, 0.006, 0.011, 0.003, 0.004, 0.008,
> 0.115, 0, 0.02, 0, 0.012, 0.009, 0.011, 0.013,
> 0.004, 0.058, 0.019, 0.006, 0.005, 0.004, 0.012,
> 0.003, 0.003, 0.004, 0.002, 0.001, 0.002, 0.102,
> -0.001, 0.008, 0.002, 0.016, 0.023, 0.014, 0.053,
> 0.009, 0.001, 0.124, 0.009, 0.008, 0.002, 0.002,
> 0.013, 0.002, 0.001, 0.042, 0.011, 0.009, 0,
> 0.004, 0.003, 0.002, 0.005, 0, 0.101, 0.013,
> 0.009, 0.005, 0.002, 0.007, 0.008, 0.067, 0.002,
> 0.064, 0.028, 0.007, 0.006, 0, 0.007, 0.006, 0,
> 0.001, 0.001, 0.001, 0, 0.088, 0.005, 0.008,
> 0.098, 0.005, 0.019, 0.007, 0.05, -0.002, 0.002,
> 0.129, 0.001, 0.004, -0.001, 0.002, -0.001, 0,
> 0.043, 0.018, 0.019, 0.015, 0.003, 0.006, 0.002,
> 0.001, 0.002, 0.004, 0.097, 0.025, 0.022, 0.007,
> 0.011, 0.007, 0.013, 0.061, 0.008, 0.013, 0.028,
> 0.004, 0.013, 0.005, 0.01, 0.004, 0, 0.006,
> -0.001, 0.001, 0.01, 0.061, 0.002, 0.004, 0,
> 0.011, 0.029, 0.018, 0, 0.003, 0.012, 0.085,
> 0.015, 0.007, 0.002, 0.003, 0.008, 0.002, 0.007,
> 0.02, 0.011, 0.02, 0.008, 0.001, 0.003, 0.01,
> 0.014, 0.001, 0.096, 0.027, 0.024, 0, 0.005,
> 0.006, 0.024, 0.087, 0.001, 0.083, 0.02, 0.009,
> 0.009, 0.001, 0, 0.019, 0, 0.003, -0.001, 0.002,
> 0, 0.089, 0.016, 0.01, 0.103, 0.003, 0.01,
> 0.002, 0.008, 0.005, 0.014, 0.1, 0.007, 0.009,
> 0.011, -0.001, 0, 0.002, 0.015, 0.036, 0.018,
> 0.026, 0.009, 0.008, 0.004, 0.001, 0.014, 0.009,
> 0.1, 0.026, 0.032, 0.008, 0.011, 0.004, 0.013,
> 0.019, 0.004, 0.02, 0.015, 0.005, 0.013, -0.001,
> 0.013, 0.012, 0, 0.01, 0.002, 0.001, 0.013,
> 0.066, 0.009, 0.005, 0.002, 0.013, 0.025, 0.006,
> 0, 0, 0.015, 0.121, 0.006, 0.003, 0.008, 0,
> 0.012, 0.011, 0.003, 0.022, 0.008, 0.032, 0.007,
> 0.002, 0.006, 0.007, 0, 0.003, 0.11, 0.01, 0.008,
> 0, 0.018, 0.008, 0.001, 0.087, 0, 0.028,
> 0.011, 0.014, 0.007, 0.001, 0.018, 0.033, 0.021,
> 0.003, 0.003, 0.007, -0.001, 0.07, 0.022, 0.009,
> 0.001, 0.007, 0.031, 0.008, 0.013, 0.01, 0.018,
> 0.125, 0.01, 0.015, 0.006, 0, 0.015, 0.019
> )
> z <- cbind(x, y)
> k <- kmeans(z, 2)
> plot(z, col=k$cluster)
>
> library(apcluster)
> m <- apclusterK(negDistMat(r=2), z, K=2, verbose=TRUE)
> plot(m, z)
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.