Multidimensional Expert Scoring Model of Institutional Repository Based on PCA

Abstract

Generally, expert information scoring has many dimensions in institutional repository. In order to reduce the computational time and improve the computational efficiency in the process of expert recommendation based on institutional repository, a multidimensional expert scoring model of institutional repository based on PCA is proposed. Firstly, the multidimensional expert scoring information matrix of institutional repository is constructed according to the feature dimension of the expert in the institutional repository. Then, using PCA algorithm to calculate the dimensionality reduction of multidimensional expert scoring information matrix of institutional repository. Finally, the multidimensional expert scoring model of institutional repository is obtained according to the 95% component information of the original space. Comparing the reconstruction errors between expert scoring test samples and non-expert scoring test samples, it shows that the model can effectively represent multidimensional expert scoring information of institutional repository.