ISR11 On Some Clustering Techniques for Information Retrieval chapter J. D. Broffitt H. L. Morgan J. V. Soden Harvard University Gerard Salton Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. i =0 W=D 10 if- i = l,m C = fi) I Wn==Wl[OCRerr]C ?0=W = C(n) 5k D F w Ix-9 = (jJs3 = 1 in "lasttt similarity matrix) k = (ild.E D, i=1,.. .,m) 1 = list of clusters found = subset of D from which clusters are to be dra[OCRerr]m C = possible cluster currently being formed P = list of documents currently being con- n sldered for inclusion in C j = no. of clusters found m = no. of d.E D 1 n = no. of documents currently in C P = S fl P In k n-[OCRerr] P =[OCRerr] N n Y. <$4~Cc~?FN [m=nC:lC(n) ifflz=nC++lPn(l) j =j+l C F=F+T Y n = 0 N k = C(n) F 9 Y N____ 10 __________ LL-j=%l[OCRerr]%l(l LIST THEj CLUSTERS, Ti, I \½[OCRerr]NF Bonner's Cluster Building Algorithm II Figure 2