Table of Content

  

20 May 2022
Volume 57 Issue 5

Data classification method based on Bayesian intuitionistic fuzzy rough sets

XUE Zhan-ao, LI Yong-xiang, YAO Shou-qian, JING Meng-meng

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  1-10.  doi:10.6040/j.issn.1671-9352.4.2021.250

Abstract ( 722 )   PDF (1387KB) ( 501 )   Save

References | Related Articles | Metrics

This paper proposed a Bayesian intuitionistic fuzzy rough set model based on the theory of intuitionistic fuzzy sets and rough sets, Bayesian probability and approximate relations is combined, and conducted research on it. Firstly, on the basis of rough sets, Bayesian probability based on intuitionistic fuzzy rough set is defined, combined with intuitionistic fuzzy approximation relation and fuzzy matrix, the intuitionistic fuzzy equivalence relation is given, and its properties are discussed. Secondly, according to the characteristics of intuitionistic fuzzy sets and cut sets, the equivalence class basis of Bayesian intuitionistic fuzzy rough sets is obtained, and the upper and lower approximation division method are further given, the positive, negative and boundary fields are calculated and the approximate accuracy is calculated. Finally, the effectiveness of the model is analyzed and verified, and the data with fuzzy information can be better classified on the UCI data sets.

Generalized multi-granularity rough sets based on parameter granular

SUN Wen-xin, LIU Yu-feng

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  11-19.  doi:10.6040/j.issn.1671-9352.4.2021.014

Abstract ( 695 )   Save

References | Related Articles | Metrics

A method of obtaining information granules by parameters is proposed aiming at the problem of obtaining information granule when information source interferes with the same problem or decision-making differently. Firstly, the definitions of counting function, parameter granule and parameter support function are given. Secondly, two kinds of generalized multi-granularity parameter granular rough set models are constructed by parameter support function, and the properties of upper and lower approximation operators of generalized multi granularity parameter granular rough set are discussed. The algorithm of upper approximation operator of type Ⅰ generalized multi-granularity parameter granular rough set is given. Thirdly, the measurement of two kinds of generalized multi-granularity parameter granule rough sets is discussed by defining the accuracy and roughness of two kinds of generalized multi-granularity parameter granule. In addition, an example is given to show the effectiveness of the model. Finally, through the analysis of experimental data, it is found that the larger the parameter is, the higher the accuracy of type Ⅰ generalized multi-granularity parameter is.

α-lower and upper approximation reductions in inconsistent interval-valued decision systems

ZHANG Xiao-yu, LI Tong-jun

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  20-27.  doi:10.6040/j.issn.1671-9352.7.2021.216

Abstract ( 565 )   PDF (473KB) ( 234 )   Save

References | Related Articles | Metrics

The focus is on the attribute reductions of inconsistent interval-valued decision systems based on α-tolerance relation. With respect to one kind of α-tolerance relations in inconsistent interval-valued decision systems, notions of α-lower and upper approximate reductions are defined, which keep the lower and upper approximations of all the decision classes unchanged, respectively. Subsequently, the equivalences among the α-upper approximate reduction and α-generalized decision reduction, α-uncertainty maintained reduction, and between α-lower approximate reduction and α-certainty maintained reduction are proved. Meanwhile, the relationships between α-lower approximate reduction and α-upper approximate reduction are discussed in detail, and verified by some examples.

Loss function determination method based on interval number analytic hierarchy process

SHI Ji, SUO Zhong-ying

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  28-37.  doi:10.6040/j.issn.1671-9352.4.2021.145

Abstract ( 758 )   PDF (4155KB) ( 283 )   Save

References | Related Articles | Metrics

In terms of three-way decision models based on decision-theoretic rough sets, the loss function is difficult to determine and rational. Considering the adaptability of interval-number analytic hierarchy process in dealing with multi-attribute decision-making problems and the tolerance of matrix error, the steps of determining the loss function based on interval-number analytic hierarchy process are given, and the effectiveness of the loss function is proved. It is applied to an example of air combat target threat assessment, and the assessment results are explained from the perspective of three decision-making. The example analysis shows that the loss function determined by this method is reasonable, the threshold pair is more compact, the uncertainty caused by data error is reduced, the risk of wrong decision is reduced, and it meets the needs of modern information air combat.

Three branches of fuzzy soft Tibetan Hetuvidya reasonings and its application

ZHANG Hai-dong, Jia-hua Duo-jie, NAN Tai-ben

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  38-45.  doi:10.6040/j.issn.1671-9352.4.2021.075

Abstract ( 622 )   PDF (467KB) ( 142 )   Save

Related Articles | Metrics

In order to deal with the problem of the combination of inference and inductive deduction in uncertainty environment, we first initiate the concept of three branches of fuzzy soft Tibetan hetuvidya reasonings by integrating three branches of Tibetan hetuvidya reasonings with fuzzy soft sets, and investigate some of interesting properties. Finally, the method to three branches of fuzzy soft Tibetan hetuvidya reasonings is applied to the determination of criminal cases, and its rationality and effective practicability are also expounded.

Robust recommendation algorithm combining random forest and target item identification

YI Hua-wei, NIU Zai-sen, LI Xiao-hui, LI Bo, JING Rong

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  46-56.  doi:10.6040/j.issn.1671-9352.2.2020.018

Abstract ( 632 )   PDF (1960KB) ( 254 )   Save

References | Related Articles | Metrics

The collaborative recommendation algorithms have lower robustness in the presence of shilling attacks. To address this problem, a robust recommendation algorithm combining random forest and target item identification is proposed. Firstly, chi-square statistics is used to extract the effective features which can distinguish normal users and attacking users. Then, the random forest classifier is used to classify user profiles, which is the first stage of the attack profile detection. Next, the further detection to the cluster including attack profiles is performed by identifying the target item, which is the second stage of the attack profile detection. Finally, based on the detection results of attack profiles, the corresponding robust recommendation algorithm is constructed. Experimental results show that the proposed algorithm outperforms the existing methods in terms of both recommendation accuracy and robustness.

ECDSA-based auditing scheme for cloud storage

LI Rui-feng, YI Zheng-ge, YUAN Wen-yong, YANG Xiao-yuan, NIU Ke

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  57-65.  doi:10.6040/j.issn.1671-9352.2.2021.004

Abstract ( 624 )   PDF (2936KB) ( 235 )   Save

References | Related Articles | Metrics

Aiming at the problem of difficult rights protection after the integrity of user cloud data is destroyed, a cloud audit scheme based on elliptic curve signature is proposed. The scheme uses blockchain and smart contract technology to realize fair payment of cloud audit. Support data privacy protection for cloud service providers and third-party auditors. There is no need to calculate power exponent and bilinear pair in the audit process, and the calculation overhead is small. Security analysis and experimental results show that the scheme is safe and efficient.

Reversible data hiding in motion vector based on regression prediction

NING Zhi-qiang, PAN Feng, NIU Ke, CHEN Pei

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  66-73.  doi:10.6040/j.issn.1671-9352.2.2021.013

Abstract ( 660 )   PDF (4899KB) ( 133 )   Save

References | Related Articles | Metrics

Aiming at the problem that the existing video data hiding algorithm based on motion vector affects the visual quality after video decoding and the bit rate increases too much, a reversible data hiding algorithm based on motion vector is proposed. The algorithm fine-tunes the difference between the predicted motion vector and the real motion vector, and embeds secret data in the lowest bit of the difference to realize the lossless recovery of the motion vector at the decoding end. Compared with the previous data hiding algorithms based on motion vectors, while ensuring good video quality and stable embedding capacity, the bit rate expansion is reduced by an average of 28.77%.

Research of balanced Boolean functions with high nonlinearity based on GSA

WU Wan-qing, ZHOU Guo-long, WANG Qiao, ZHAO Yong-xin

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  74-84.  doi:10.6040/j.issn.1671-9352.2.2021.007

Abstract ( 580 )   PDF (510KB) ( 238 )   Save

References | Related Articles | Metrics

A two-stage Boolean function search algorithm is proposed, the first stage is based on the Gravitational Search Algorithm, using a floating-point coding rule, which specifies the conversion between the truth table and the floating-point vector, and set the appropriate cost function, the rules of the force between the particles and the movement rules. The second stage is called local traversal, which improves the nonlinearity of Boolean functions. Simulating experiments show that our algorithm can gain a large number of balanced 6-9 variables Boolean functions with high nonlinearity and low autocorrelation, among them with optimal or sub-optimal autocorrelation.

Improved peripheral register category scheme for IoT firmware testing

WANG Li-na, CHEN Si, ZHANG Tong, QIN Peng, XU Lai

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  85-91.  doi:10.6040/j.issn.1671-9352.2.2021.011

Abstract ( 506 )   PDF (913KB) ( 262 )   Save

References | Related Articles | Metrics

By emulating the data interface between microcontroller and peripherals, rehosting proposed a new way to test firmware, but it is affected by the accuracy of peripherals registers it has categorized. By studying the details of firmware hosting, the problems in the register category can be found, and a set of reasonable correction schemes to synthesize the test results of multiple firmware can be adopted to improve the accuracy of its classification of peripheral registers, with very low overhead. After testing 54 different firmware under 4 types of microcontrollers, the solution can improve the accuracy of register classification judgments in most cases, and effectively improve the effect of firmware testing.

Connectivity and Hamiltonian properties of second order circuit graphs of a class of uniform matroid

DENG Zi-Jian, LIU Bin, HUO Bo-feng

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  92-96.  doi:10.6040/j.issn.1671-9352.0.2020.398

Abstract ( 522 )   PDF (611KB) ( 452 )   Save

References | Related Articles | Metrics

The connectivity and Hamiltonian properties of the second order circuit graph of the uniform matroid U2,n have been investigated. The second order circuit graph of U2,n is 3(n-3)-regular graph and the connectivity and edge-connectivity of its are equal to the minimum degree. Furthermore, Hamiltonian properties of the second order circuit graph of U2,n is introduced and it is Hamiltonian connected.

Influence of altruistic behaviors upon service and pricing decisions of dynamic supply chain

JIANG Ji-kun, LI Yu, WANG Wei-hao

JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE). 2022, 57(5):  97-110.  doi:10.6040/j.issn.1671-9352.0.2021.193

Abstract ( 494 )   PDF (3692KB) ( 145 )   Save

References | Related Articles | Metrics

In the dynamic product supply chain system which dominated by manufacturer, considering that retailers and manufacturers have altruistic behavior preferences, three altruistic models are constructed: retailer altruism, manufacturer altruism and bilateral altruism. Using the differential game theory, the dynamic equilibrium strategies and their value functions of three altruistic modes are obtained. The market parameters and altruistic preference parameters of strategies under different modes are analyzed. The results show that retailers altruistic preference is not always beneficial to retailers utility and retailer should restrain this preference. The manufacturers altruistic preference can always encourage retailers to provide better services and increase the utility of both parties. The service price ratio of the manufacturers altruistic mode is the highest which is secondly to the centralized mode.