Research Article
A New Dual-Mode GEP Prediction Algorithm Based on Irregularity and Similar Period
| | Begin: | | | Step1:Data preprocessing | | (1) | For i = 1 to G-L do//base array | | (2) | | | (3) | End for; | | (4) | For j = 1 to W do//subgoal modeling and forecast sets | | (5) | Datas [j] ← Y_datas [j]...Y_datas [j + (N-3) L]; | | (6) | End for; | | (7) | Com_Chromosme [W] ← structure container for complex subgoals; | | | Step2:genetic evolution process | | (8) | For t = 1 to W do | | (9) | CHROMOSME [ ] ← stochastic original population using Datas [t]; | | (10) | BEST ← Using IPM algorithm get the best individual; | | (11) | Com_Chromosme [t] ← BEST | | (12) | End for; | | (13) | Return Com_Chromosme [W] as result; | | | Step3:use compound target | | (14) | For x = 1 to P do | | (15) | FORECAST gets using Com_Chromosme [x mod L]; | | (16) | SUM_ERROR += gets calculate the error; | | (17) | End for; | | (18) | AVE_ERROR gets SUM_ERROR/P; | | | End; |
|