Uyghur Text Feature Selection Methods by Artificial Fish Optimization
Journal Title: 河南科技大学学报（自然科学版） - Year 2016, Vol 37, Issue 6
Abstract
Feature selection was a key step in text classification,which had direct effect on the classification results. The basic principles of artificial fish behaviors of foraging,gregarious and rear-end were analyzed.Combined with the principle of Uyghur text feature extraction,a kind of improved artificial fish swarm algorithm to Uyghur text feature extraction was proposed. In order to speed up the convergence rate of fish,a strategy was taken into fish behaviors by changing the perspective. At the same time,the mutation strategy was added to the algorithm to avoid falling into local optimal solution. Finally,the sample after feature selection was input into the classifiers to perform the simulation experiment. The results show that the improved artificial fish swarm algorithm can make the classification accuracy reach 94.5%.
Authors and Affiliations
Bingbing WU, •Abudureyimu HALIDAN, •Aierken ALIYA, Yan HE
Keywords
Related Articles
- EP ID EP477490
- DOI 10.15926/j.cnki.issn1672-6871.2016.06.010
- Views 96
- Downloads 0
Gray Relational Analysis of Red Mud Mass Fraction and Asphalt Performance
In order to analyze the variation trend and sensitivity of the asphalt performance index with the mass fraction of red mud changing, six kinds of non-dimensional treatment in the gray theory were used to obtain the effec...
Integral Influence Analysis of Bridge Partly Strengthened by Prestressed Carbon Fibre Reinforced Plastics Plate
1
Construction and Photoelectric Conversion Properties of WO3/BiVO4 Composite Film Photoelectrode
WO3 and BiVO4 semiconductor films eletrodes were prepared by polymer-assisted deposition method and metal-organic decomposition method, respectively. The band structures of WO3 and BiVO4 semiconductor films eletrode were...
Control Strategy of Miniature Pure Electric Truck Based on Driver's Intention Recognition
In order to improve the energy efficiency of miniature pure electric truck, a vehicle control strategy based on driver's intention recognition was proposed.Based on Stateflow finite state machine, a primary driver recogn...
Rolling Bearing Clustering Faults Diagnosis Based on EEMD,Permutation Entropy and PCA-GK
Aiming at the problem that the dimension of entropy eigenvectors decomposed by ensemble empirical mode decomposition( EEMD) of rolling bearing vibration signals was high,and the computational efficiency of sample entropy...