Microbial Decolorization of Triazo Dye, Direct Blue 71: An Optimization Approach Using Response Surface Methodology (RSM) and Artificial Neural Network (ANN)

Table 2

Box–Behnken matrix for experimental design and predicted response using RSM and ANN.

Run	A: dye conc (ppm)	B: yeast extract (g/L)	C: pH (g/L)	Decolorization (%)	Predicted RSM (%)	Predicted ANN (%)
1	100	1.75	6.75	90.97	89.84	89.356
2	100	1.75	6.75	87.15	89.84	89.382
3	50	3	6.75	94.9	94.1	92.076
4	150	3	6.75	89.86	92.09	89.862
5	150	1.75	6	77.03	74.87	77.03
6	150	0.5	6.75	54.43	55.23	58.118
7	100	3	6	88.69	88.62	88.689
8	100	0.5	7.5	66.18	66.25	66.18
9	150	1.75	7.5	77.01	76.14	78.499
10	100	1.75	6.75	88.57	89.84	89.347
11	100	3	7.5	89.8	88.44	89.799
12	50	0.5	6.75	79.06	76.83	79.06
13	100	0.5	6	55.31	56.67	61.547
14	50	1.75	6	82.38	83.25	82.381
15	100	1.75	6.75	90.72	89.84	89.328
16	100	1.75	6.75	91.77	89.84	90.356
17	50	1.75	7.5	89.22	91.38	89.222

ANN training dataset—unbold numbers. ANN testing dataset—bold numbers.