In this paper, for predicting the temperature of thermostable proteins on some useful physicochemical quantities of each amino symbolic sequence with different lengths, a novel integrated algorithm of Hurst exponent and Choquet integral with respect to extensional Lambda-measure was proposed. A real data set of the temperature of thermostable proteins with 5-fold cross-validation MSE was used for evaluation. The performances of the Choquet integral regression models based on extensional Lambdameasure, completed L-measure, Lambda-measure, and the P-measure, respectively, along with two traditional regression models, the multiple regression model and the ridge regression model were compared. The results show the integrated algorithm of Hurst exponent and Choquet integral regression model with respect to extensional Lambda-measure outperforms the others.
|Number of pages||6|
|Journal||ICIC Express Letters|
|Publication status||Published - 2013 Jun 1|
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Computer Science(all)