Feed-forward Artificial Neural Network based inference system applied in bioinformatics data-mining

Inference system using softcomputing and mixed data applied in metabolic pathway datamining

“…Several FANN configurations were tested and evaluated (Leiva et al, 2009) with three different input data models: without previous BLAST parameter normalisation, with a normalisation approach previously discussed in literature (Arredondo et al, 2008), and with a new parameter normalisation approach. Several FANN configurations were tested and evaluated (Leiva et al, 2009) with three different input data models: without previous BLAST parameter normalisation, with a normalisation approach previously discussed in literature (Arredondo et al, 2008), and with a new parameter normalisation approach.…”

“…The inference systems were evaluated and catalogued according to the following test cases: two sets of fuzzy rules generated by human naive and expert users (H-N and H-E), a standard SVM classifier implementation (Weka SMO Waikato environment for knowledge analysis, http://sourceforge.net/projects/weka/), two sets of fuzzy rules (GA-1 and GA-2) that were generated by a GA training method (Arredondo et al, 2008) and five FANN-based inference systems (i.e., NN-1 through NN-5) (Leiva et al, 2009). The other 50% were used as the test set to evaluate the obtained models and associated inference systems.…”