Research Article

Biomedical Text Categorization Based on Ensemble Pruning and Optimized Topic Modelling

Table 8

The macro-averaged F-measure results obtained with different LDA-based configurations.

ā€‰Naive Bayes (NB)Support Vector Machines (SVM)

Configurationoh5oh10oh15ohscalOhsu-medoh5oh10oh15ohscalOhsu-med

LDA (k=50)0.750.680.710.610.300.770.800.850.730.36
LDA (k=100)0.720.650.690.620.310.790.800.850.750.40
LDA (k=150)0.700.670.670.610.310.770.810.860.760.43
LDA (k=200)0.670.650.650.610.290.780.800.860.760.44
GA-LDA (BIC)0.760.690.760.740.370.790.700.770.760.37
PSO-LDA (BIC)0.760.700.780.750.370.790.740.770.780.38
FA-LDA (BIC)0.760.730.790.750.370.800.760.780.790.39
CSA-LDA (BIC)0.770.730.800.750.370.800.780.800.790.40
BA-LDA (BIC)0.800.740.810.750.380.810.780.810.800.41
GA-LDA (CH)0.800.740.820.770.380.810.790.820.810.41
PSO-LDA (CH)0.810.740.820.770.390.820.790.820.810.41
FA-LDA (CH)0.820.740.820.770.390.830.800.820.820.41
CSA-LDA (CH)0.820.750.830.780.400.830.800.820.820.41
BA-LDA (CH)0.820.750.830.790.410.840.810.830.820.41
GA-LDA (DB)0.850.780.860.810.420.860.830.880.850.45
PSO-LDA (DB)0.850.820.870.820.440.860.830.880.870.45
FA-LDA (DB)0.870.820.870.830.460.870.840.890.870.46
CSA-LDA (DB)0.870.830.880.840.470.880.840.890.880.49
BA-LDA (DB)0.880.830.890.860.490.900.840.900.880.52
GA-LDA (SI)0.820.750.840.790.410.840.810.840.820.42
PSO-LDA (SI)0.830.750.840.790.410.840.810.850.830.43
FA-LDA (SI)0.840.760.850.800.410.850.810.850.830.43
CSA-LDA (SI)0.850.770.850.800.410.850.820.860.840.43
BA-LDA (SI)0.850.780.850.810.420.850.830.870.850.44

LDA: latent Dirichlet allocation, GA-LDA: genetic algorithm based LDA, PSO-LDA: particle swarm optimization based LDA, FA-LDA: firefly algorithm based LDA, CSA-LDA: cuckoo search algorithm based LDA, BA-LDA: bat algorithm based LDA, BIC: Bayesian information criterion, CH: Calinski-Harabasz index, DB: Davies-Bouldin index, and SI: Silhouette index.