Stock selection for effective investment decisions is a valuable and attractive research interest for many years. Owing to the uncertainty and complexity of the stock market, many fuzzy multicriteria decision-making (MCDM) methods were proposed to solve stock selection problems. However, these methods have difficulty in characterizing unreliable information, which is widespread in the stock market, and handling the non-compensation among multiple criteria. In this paper, an innovative method is developed from the perspectives of information reliability and criterion non-compensation to manage stock selection problems. First, the Z-number, which is a powerful tool for describing real-life information and identifying information reliability, is introduced to depict stock evaluation information. Second, the outranking degree of Z-numbers is defined based on the fuzzy and probability information. Subsequently, some outranking aggregation and exploitation procedures are presented based on the idea of Elimination and Choice Translating Reality (ELECTRE) I to handle the non-compensation among stock evaluation criteria. By integrating the above studies, a Z-number ELECTRE I MCDM method is developed. Finally, a stock investment object selection problem is solved, and some discussions and analyses are conducted to testify the applicability and validity of this method.

Peer-to-Peer (P2P) lending brings many benefits for both lenders and borrowers, as well as risk for them, especially for the lenders. To help the lenders select a reliable P2P lending platform with high return and low risk, a more comprehensive selection method is necessary. However, vast information with fluctuant, fuzzy and subjective data on the P2P lending platform make it more difficult for lenders to make the right choice. This paper intends to propose a selection method by using intuitionistic fuzzy soft set to organize the data, and using D-S theory of evidence to integrate the intuitionistic fuzzy values. Besides, the paper analyses the critical factors affecting the lenders’ decision, and presents some indicators for evaluating. Finally, an empirical experiment was given by choosing several P2P lending platforms with real data, and the results are compared with the intuitionistic fuzzy weighted average and Development Index from Wangdaizhijia, that verify the validity and superiority of the proposed method.

With the advent of the era of big data, data has become bigger than ever. Recently, as a fundamental task of pattern recognition, predict and data mining, feature selection has aroused wide public concern. However, extant methods on feature selection have an $O(\left|C\right|^x\left|U\right|^y)$ time complexity, which is the bottleneck preventing people from exploring knowledge in large-scale or high-dimensional datasets. Based on bijective soft sets, we propose a new rationale for feature selection, which can help break that bottleneck. Subsequently, this paper proposes an $O(\left|U\right|)$ feature-selection method whose computational time increases linearly only with the number of instances. To validate the proposed method, we conduct extensive experiments on the University of California Irvine (UCI) datasets in which large-scale and high-dimensional datasets containing four million instances and over three million features are included. The results reveal that the proposed method is an efficient, effective, and outperforms traditional methods in runtime, which can save massive computing resources. Moreover, the proposed method can be applied to feature selection for large-scale and gigantic-dimensional datasets, which are difficult to process with traditional methods.

Well known for its simplicity and effectiveness in classification, AdaBoost, however, suffers from overfitting when class-conditional distributions have significant overlap. Moreover, it is very sensitive to noise that appears in the labels. This paper tackles the above limitations simultaneously via optimizing a modified loss function (i.e., the conditional risk). The proposed approach has the following two advantages. First, it is able to directly take into account label uncertainty with an associated label confidence. Second, it introduces a trustworthiness measure on training samples via the Bayesian risk rule, and hence the resulting classifier tends to have finite sample performance that is superior to that of the original AdaBoost when there is a large overlap between class conditional distributions. Theoretical properties of the proposed method are investigated. Extensive experimental results using synthetic data and real-world data sets from UCI machine learning repository are provided. The empirical study shows the high competitiveness of the proposed method in predication accuracy and robustness when compared with the original AdaBoost and several existing robust AdaBoost algorithms.