Skip to main content

Table 1 Clustering parameters

From: A big data methodology for categorising technical support requests using Hadoop and Mahout

Parameter Value
k (k-means) 40
T1 (Canopy clustering) 1.0
T2 (Canopy clustering) 1.4
Fuzziness Factor (Fuzzy-kmeans) 1.05
Dirichlet Clusters 50
Max Dirichlet Iterations 10
LDA Clusters 50
Words in Corpus 12886
Max LDA Iterations 30