clustering algorithm 0.001810102
document similarity 0.0017851170000000001
document clustering 0.001769159
text documents 0.001681475
vector model 0.001485486
document retrieval 0.001412798
clustering method 0.001393708
document ranking 0.0013694269999999999
following words 0.001333264
document vectors 0.001328185
text collection 0.001327744
document frequency 0.001299013
regression algorithm 0.0012902080000000002
algorithm the 0.001266669
document indexing 0.001252292
document index 0.001248492
query words 0.001213024
virtual documents 0.001190067
similarity threshold 0.00118886
czech documents 0.001153771
cosine similarity 0.001136223
lsi model 0.001134292
individual words 0.001106239
information retrieval 0.001086257
positive function 0.001070778
vector models 0.00106637
hard clustering 0.0010250419999999999
algorithm 0.00102387
fuzzy clustering 0.00101306
specific topics 0.001010387
specific cluster 0.00100506
other nlp 9.964330000000001E-4
document 9.82927E-4
real function 9.69645E-4
large collection 9.59271E-4
experimental results 9.577769999999999E-4
linear functions 9.572059999999999E-4
training collection 9.46997E-4
test collection 9.45223E-4
detailed information 9.40714E-4
membership function 9.40583E-4
different newspapers 9.36594E-4
quality function 9.24203E-4
inverse matrix 9.23126E-4
linear regression 9.14748E-4
model 9.11862E-4
experimental collection 9.09454E-4
linear func 9.07528E-4
linear transformation 9.06058E-4
documents 8.98646E-4
function ﬁﬀﬃﬂ 8.96631E-4
matrix inversion 8.91011E-4
matrix multiplica 8.8537E-4
original method 8.84014E-4
unrelated topics 8.838279999999999E-4
linear programming 8.78583E-4
words 8.7804E-4
linear program 8.73922E-4
similar approach 8.73778E-4
linear inequalities 8.71592E-4
output vector 8.6854E-4
heuristic method 8.52393E-4
positive parameters 8.483060000000001E-4
vector rep 8.43338E-4
classic vector 8.3993E-4
nlp methods 8.29211E-4
sample set 8.28422E-4
corresponding set 8.26371E-4
important parameters 8.24654E-4
high value 8.23745E-4
threshold weight 8.237419999999999E-4
formative cluster 8.171070000000001E-4
meable topics 8.14232E-4
collection let 8.11265E-4
ing values 8.10921E-4
local search 8.105580000000001E-4
topic 8.09603E-4
error threshold 8.07435E-4
vector technique 8.04768E-4
similarity 8.0219E-4
threshold size 8.021199999999999E-4
related work 8.0008E-4
czech collection 8.000399999999999E-4
sic vector 7.96057E-4
ple set 7.95763E-4
whole collection 7.93985E-4
clustering 7.86232E-4
annotated collection 7.714169999999999E-4
analyzed collection 7.69048E-4
computational complexity 7.663990000000001E-4
search procedure 7.55992E-4
positive value 7.53996E-4
crucial algorithms 7.53282E-4
several approaches 7.5039E-4
main issues 7.38112E-4
probability distribu 7.29076E-4
crucial parameters 7.2326E-4
high dimension 7.22554E-4
list paper 7.22492E-4
high quality 7.197379999999999E-4
