Uncertainty measurement for a fuzzy relation information system

Li, Zhaowen; Zhang, Pengfei; Ge, Xun; Xie, Ningxin; Zhang, Gangqiang; Wen, Ching‐Feng

doi:10.1109/tfuzz.2019.2898158

Cited by 28 publications

(11 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RST can effectively deal with the uncertainty of information system (IS). In recent years, RST has attracted many researchers' attention, and its application is mostly related to an IS [18], [42].…”

Section: A Research Background and Related Workmentioning

confidence: 99%

Feature Selection for Interval-Valued Data Based on D-S Evidence Theory

Peng

Zhang

2021

IEEE Access

View full text Add to dashboard Cite

Feature selection is one basic and critical technology for data mining, especially in current "big data era". Rough set theory (RST) is sensitive to noise in feature selection due to the strict condition of equivalence relation. However, D-S evidence theory is flexible to measure uncertainty of information. This paper introduces robust feature evaluation metrics "belief function" and "plausibility function" into feature selection algorithm to avoid the defect that classification effect is affected by noise. First of all, similarity between information values in an interval-valued information system (IVIS) is given and a variable parameter to control the similarity of samples is introduced. Then, θ-lower approximation and θ-upper approximation in IVIS are put forward. Next, belief function and plausibility function based on θlower approximation and θ-upper approximation are put forward. Finally, four feature selection algorithms in an IVIS based on D-S evidence theory are proposed. The experimental results on four real interval-valued datasets show that the proposed metrics are robust to noise, and the proposed algorithms are more effective than the existing algorithms.

show abstract

Section: A Research Background and Related Workmentioning

confidence: 99%

Feature Selection for Interval-Valued Data Based on D-S Evidence Theory

Peng

Zhang

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Fuzzy measure is a powerful method to describe the crossinteractions, especially the non-linear Choquet integral. It has been successfully applied in evidence fusion and intelligence information processing [5], [9], [19], [26]. Since the massive MIMO system's throughput is a multi-dimensional factor influenced conception, the interaction among contribution of contributions towards the objective system throughput can be measured adequately through a non-additive FM, which is the main feature of the Choquet integral model.…”

Section: Choquet Integral-based Interdependency and Significance Analysis Algorithmmentioning

confidence: 99%

DRESIA: Deep Reinforcement Learning-Enabled Gray Box Approach for Large-Scale Dynamic Cyber-Twin System Simulation

Lin

Yang

et al. 2021

IEEE Open J. Comput. Soc.

View full text Add to dashboard Cite

The massive data generated by large-scale dynamic systems makes its optimization facing a tough challenge. Traditional White Box-based methods directly model the internal operating mechanism of the system, so massive amounts of measured data need to be handled, which is costly and time-consuming. The poor interpretability of the Black Box-based methods makes it difficult to adapt to the dynamic environment. Thus we propose a novel Gray Box-based approach namely Deep Reinforcement Learning-enabled Constraint Set Inversion Algorithm (DRESIA), which establishes a quantitative model of the nonlinear interoperability effects of system internal states which simplifies the White Box's complex mechanism of reconstruction and prediction and retains the interpretability of the model, therefore improves the prediction efficiency of feasible region while also improving the generalization ability. It further improves the dynamic adaptability of the modeling environment, which provides a new performance balancing scheme for system modeling. Under the premise that the large-scale 5G Cyber-Twin system satisfies the given Quality of Service(QoS) requirements, we perform DRESIA to realize the efficient and dynamic optimal search of feasible region, the results show that the DRESIA reduces the computational cost, and balances the accuracy and robustness of the feasible region, which validate the effectiveness and superiority of Gray Box-based approach.

show abstract

“…Dai et al [8] presented UM for an interval-valued decision IS based on extended conditional entropy. Li et al [18] studied uncertainty measurement for a fuzzy relation IS. Zhang et al [40]…”

Section: A Research Background and Related Workmentioning

confidence: 99%

Uncertainty Measurement for a Hybrid Information System With Images: An Application in Attribute Reduction

Wang

2020

IEEE Access

View full text Add to dashboard Cite

A hybrid information system with images (HISI) is an information system (IS) where there exist many kinds of data or attributes (e.g., boolean, categorical, real-valued, set-valued, interval-valued, imaged, decision and missing data or attributes). Uncertainty measurement (UM) is an effective tool for evaluation. This paper inquires into UM for a HISI with application to attribute reduction. The distance between attribute values in a HISI is first constructed. Then, the tolerance relations on an object set of a HISI are obtained by means of this distance. Next, information structures in a HISI are proposed. After that, UM for a HISI is considered by using its information structures. Moreover, effectiveness analysis for the proposed measures is conducted from the angle of statistics. Finally, an application of the proposed measures in attribute reduction for a HISI is given, and the corresponding algorithms are put forward.

show abstract

Uncertainty measurement for a fuzzy relation information system

Cited by 28 publications

References 45 publications

Feature Selection for Interval-Valued Data Based on D-S Evidence Theory

Feature Selection for Interval-Valued Data Based on D-S Evidence Theory

DRESIA: Deep Reinforcement Learning-Enabled Gray Box Approach for Large-Scale Dynamic Cyber-Twin System Simulation

Uncertainty Measurement for a Hybrid Information System With Images: An Application in Attribute Reduction

Contact Info

Product

Resources

About