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) |
| Configuration | oh5 | oh10 | oh15 | ohscal | Ohsu-med | oh5 | oh10 | oh15 | ohscal | Ohsu-med |
| LDA (k=50) | 0.75 | 0.68 | 0.71 | 0.61 | 0.30 | 0.77 | 0.80 | 0.85 | 0.73 | 0.36 | LDA (k=100) | 0.72 | 0.65 | 0.69 | 0.62 | 0.31 | 0.79 | 0.80 | 0.85 | 0.75 | 0.40 | LDA (k=150) | 0.70 | 0.67 | 0.67 | 0.61 | 0.31 | 0.77 | 0.81 | 0.86 | 0.76 | 0.43 | LDA (k=200) | 0.67 | 0.65 | 0.65 | 0.61 | 0.29 | 0.78 | 0.80 | 0.86 | 0.76 | 0.44 | GA-LDA (BIC) | 0.76 | 0.69 | 0.76 | 0.74 | 0.37 | 0.79 | 0.70 | 0.77 | 0.76 | 0.37 | PSO-LDA (BIC) | 0.76 | 0.70 | 0.78 | 0.75 | 0.37 | 0.79 | 0.74 | 0.77 | 0.78 | 0.38 | FA-LDA (BIC) | 0.76 | 0.73 | 0.79 | 0.75 | 0.37 | 0.80 | 0.76 | 0.78 | 0.79 | 0.39 | CSA-LDA (BIC) | 0.77 | 0.73 | 0.80 | 0.75 | 0.37 | 0.80 | 0.78 | 0.80 | 0.79 | 0.40 | BA-LDA (BIC) | 0.80 | 0.74 | 0.81 | 0.75 | 0.38 | 0.81 | 0.78 | 0.81 | 0.80 | 0.41 | GA-LDA (CH) | 0.80 | 0.74 | 0.82 | 0.77 | 0.38 | 0.81 | 0.79 | 0.82 | 0.81 | 0.41 | PSO-LDA (CH) | 0.81 | 0.74 | 0.82 | 0.77 | 0.39 | 0.82 | 0.79 | 0.82 | 0.81 | 0.41 | FA-LDA (CH) | 0.82 | 0.74 | 0.82 | 0.77 | 0.39 | 0.83 | 0.80 | 0.82 | 0.82 | 0.41 | CSA-LDA (CH) | 0.82 | 0.75 | 0.83 | 0.78 | 0.40 | 0.83 | 0.80 | 0.82 | 0.82 | 0.41 | BA-LDA (CH) | 0.82 | 0.75 | 0.83 | 0.79 | 0.41 | 0.84 | 0.81 | 0.83 | 0.82 | 0.41 | GA-LDA (DB) | 0.85 | 0.78 | 0.86 | 0.81 | 0.42 | 0.86 | 0.83 | 0.88 | 0.85 | 0.45 | PSO-LDA (DB) | 0.85 | 0.82 | 0.87 | 0.82 | 0.44 | 0.86 | 0.83 | 0.88 | 0.87 | 0.45 | FA-LDA (DB) | 0.87 | 0.82 | 0.87 | 0.83 | 0.46 | 0.87 | 0.84 | 0.89 | 0.87 | 0.46 | CSA-LDA (DB) | 0.87 | 0.83 | 0.88 | 0.84 | 0.47 | 0.88 | 0.84 | 0.89 | 0.88 | 0.49 | BA-LDA (DB) | 0.88 | 0.83 | 0.89 | 0.86 | 0.49 | 0.90 | 0.84 | 0.90 | 0.88 | 0.52 | GA-LDA (SI) | 0.82 | 0.75 | 0.84 | 0.79 | 0.41 | 0.84 | 0.81 | 0.84 | 0.82 | 0.42 | PSO-LDA (SI) | 0.83 | 0.75 | 0.84 | 0.79 | 0.41 | 0.84 | 0.81 | 0.85 | 0.83 | 0.43 | FA-LDA (SI) | 0.84 | 0.76 | 0.85 | 0.80 | 0.41 | 0.85 | 0.81 | 0.85 | 0.83 | 0.43 | CSA-LDA (SI) | 0.85 | 0.77 | 0.85 | 0.80 | 0.41 | 0.85 | 0.82 | 0.86 | 0.84 | 0.43 | BA-LDA (SI) | 0.85 | 0.78 | 0.85 | 0.81 | 0.42 | 0.85 | 0.83 | 0.87 | 0.85 | 0.44 |
|
|
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.
|