DORA :: Browsing by Author "Liu, Sifeng"

Browsing by Author "Liu, Sifeng"

Now showing 1 - 20 of 35

Embargo
A generalized grey model with symbolic regression algorithm and its application in predicting aircraft remaining useful life
(Elsevier, 2024-07-18) Liu, Lianyi; Liu, Sifeng; Yang, Yingjie; Guo, Xiaojun; Sun, Jinghe
As a sparse data analysis method, a grey model faces challenges in interpretability for its effective application in uncertain systems. This study proposes a generalized grey model (GGM) based on symbolic regression, designed to improve the intelligence and adaptability of grey models. The GGM serves as a unified framework, integrating various grey model families and addresses regression challenges to determine the model structure. Symbolic regression in the GGM identifies symbolic input-output relationships, offering an interpretable approach for structure determination. By leveraging the non-uniqueness principle in grey system theory and employing structural penalty parameters, the model balances complexity and interpretability. A comparative analysis between GGM and conventional grey function models is conducted focusing on the differences in modeling, structure identification, and parameter optimization. Validation on the M3 competition dataset demonstrated the GGM's superior performance, achieving a significant reduction in prediction error compared to other grey forecasting models. Additionally, a rigorous analysis of aircraft lifespan data underscored the robustness and accuracy of GGM in practical engineering applications.
Embargo
A recursive polynomial grey prediction model with adaptive structure and its application
(Elsevier, 2024-03-26) Liu, Lianyi; Liu, Sifeng; Yang, Yingjie; Fang, Zhigeng; Shuqi Xu
As a sparse data analysis algorithm, ensuring a reasonable model structure is an important challenge for grey models to identify the control mechanism of the uncertain system from observational data. To improve the intelligence and adaptability of the model, this study presents a synchronized optimization strategy for data prioritization and model structure for discrete polynomial grey prediction model. The proposed polynomial grey model contains two hyper-parameters: memory factor parameter and structural parameter. The memory factor is introduced into the discrete model to reconstruct the objective function of structural parameter optimization, thereby avoiding the problem of information superposition. The structural parameter is used to enhance the adaptability of grey prediction model in uncertain data analysis tasks. By employing a recursive estimation approach, an adaptive strategy for estimating model hyper-parameters is proposed, which focuses on minimizing prediction errors within the in-sample data. Additionally, a comparison is made between the proposed improved polynomial grey model and existing polynomial grey models in terms of data information mining, estimation stability, and robustness against measurement noise. The proposed model is applied to the practical engineering application of wear prediction, further validating the effectiveness of the proposed approach in non-equidistant time series prediction tasks.
Metadata only
Agent-based modelling in grey economic systems
(Springer, 2023-02-05) Delcea, Camelia; Yang, Yingjie; Liu, Sifeng; Cotfas, Liviu-Adrian
The economic systems are basically grey systems due to their components and to their interactions which enable the occurrence of uncertainty. First, the human component plays an important role as a consequence of its usually unpredictable and sometimes irrational behavior, a situation strictly related to the way the humans are thinking and acting. From here, it can easily be demonstrated that when analyzing a system, we are facing grey knowledge. This kind of knowledge exists and it represents that small piece of puzzle needed to successfully fill the gap separating the explicit knowledge form the tacit one, also conducting to uncertainty.
Open Access
Comparative analysis of properties of weakening buffer operators in time series prediction models
(Elsevier, 2018-08-23) Li, Chong; Yang, Yingjie; Liu, Sifeng
Reducing the negative influence of stochastic disturbances in sample data has always been a difficult problem in time series analysis. In this paper, three new fractional weakening buffer operators are proposed, and then some desirable properties of these proposed se- quence operators are investigated. Their potential effect in smoothing unexpected distur- bances while maintaining the normal trend in sample series is analyzed and compared with other widely used sequence operators in time series modeling. Results of theoretical and empirical research show that the proposed novel fractional weakening buffer oper- ators are effective in improving the development pattern analysis of time series in dis- turbance scenarios, while also avoid too subjectively weighting experimental data from collected samples. The robust of the proposed operator-based prediction algorithm against noise effect is tested in five different types of noise scenarios. Result of empirical study demonstrates that the proposed method improves the series prediction performance and it also improves the robustness of corresponding forecasting algorithms. These unique prop- erties of the proposed weakening buffer operators make them more attractive in time se- ries analysis.
Open Access
Designing optimal proactive replacement strategies for degraded systems subject to two types of external shocks
(Taylors & Francis, 2023-02-27) Dong, Wenjie; Yang, Yingjie; Cao, Yingsai; Zhang, Jingru; Liu, Sifeng
This paper mainly investigates a proactive replacement policy for a stochastically deteriorating system concurrently subject to two types of shocks. Firstly, the closed form representation of system reliability function suffering from both a degradation process and environmental shocks is derived based on the degradation-threshold failure (DTS) modelling framework. An age- and state-dependent competing risks model with mutual dependence between the two failure processes is embedded into system reliability modelling, where two types of shocks are taken into consideration upon arrival of an external shock including a minor one and a major one. Based on which, a bivariate maintenance policy is put forward for the deteriorating system, where the system is proactively replaced before failure at a planned time, or at an appropriate number of minimal repairs, whichever takes place first. The expected long-run cost rate (ELRCR) is formulated, and optimal solutions are evaluated analytically for two special cases. Finally, an illustrative example is redesigned to validate the theoretical results, exploring the significance of two types of shocks and mutual dependence in system reliability modelling, and illustrating the potential applications in maintenance decisions in various manufacturing systems.
Metadata only
Development of grey system research
(Springer, 2023-02-05) Yang, Yingjie; Liu, Sifeng
The first chapter presents an introduction to grey system theory. We begin with some background information on the appearance and growth of grey system theory and founder “Professor Julong Deng”, who established the new branch of science. Then we examine the development history and current state, followed by a comparison of grey system theory with other uncertain systems theories such as probability and statistics, fuzzy mathematics, and rough set theory. Finally, the elementary concepts, the fundamental principles and the framework of grey system theory are presented. This introductory chapter vividly presents the reader with an overall picture of grey system theory and its frontier research.
Embargo
Do not try to evaluate research results in a hurry
(Emerald, 2019-01-28) Liu, Sifeng; Yang, Yingjie
We analysed the problems of the current research evaluation, and concluded that research results should be evaluated after their impacts (academic or non-academic) are fully released, and not immediately after publication. Many of the problems associated with mismanagement in research could be eradicated if people did not try to evaluate research results immediately after publication.
Metadata only
Emerging studies and applications of grey systems
(Springer, 2023-02-06) Yang, Yingjie; Liu, Sifeng
At a time when most researchers and businesses are talking about Big Data, small and incomplete data are never far away from us. In fact, most of our decisions are made under incomplete and small data. There are many areas where small and incomplete data are inevitable, such as health care, traffic management, economic analysis, etc. The theory of grey systems emerged as a novel solution to deal with small and incomplete data, and it is attracting more and more attention from researchers and professionals in different disciplines. It has the ability to establish models from extremely limited data under uncertainty, which is usually difficult for other models. Although grey systems provide powerful capability in data analysis, its impacts in applications have mainly materialized in China. I have frequently received questions regarding the differences between Grey systems and other related models, and to provide examples of their real world use. To promote grey systems and its applications, an international research network was formed under the support of the Leverhulme Trust in 2015. The leading experts from UK, China, Canada, Romania and Spain formed the core of the network. A number of activities were conducted to foster the collaboration between members of the network. As a result, the international association of grey systems and uncertainty analysis was established in 2016. Through our discussion in the network and the association, it was deemed necessary to assemble the latest advance in this field to clarify confusions between grey systems and grey sets with other models, and provide a collection of successful application cases as guidance for people interested in applying grey systems to their problems. Therefore, a collection of contributions from our network partners and leading experts in the field is put together to form this book. The contributors include world leading experts as well as a new generation of research leaders from both Europe and China. The application covers many different disciplines, ranging from social economics to engineering, energy and management. The book focuses mainly on two themes: 1. The connection between grey systems and other related models, such as fuzzy sets and rough sets (chapter 3-5). 2. Successful real world applications in different disciplines (chapter 6-12). To introduce to the readers who have no background in grey systems, a brief introduction is also included (chapter 1-2). Chapter 1 is a general overview of the whole field of grey systems and its recent development. It helps the readers to see the development and the state of the art of this field. Chapter 2 then gives a brief introduction to the fundamental concepts in grey systems, especially grey models so as to enable the readers to understand other chapters. These two chapters lay the essential backgrounds for the contents in other chapters. Chapter 3 then moves into the connection between grey systems, especially grey sets with various fuzzy sets and rough sets. It helps to clarify the confusions between grey systems and fuzzy sets and rough sets. Chapter 4 focuses on the hybridization of grey systems with neural-fuzzy systems and illustrates how they can be integrated together. Chapter 5 puts grey systems into a more general level and puts forward the novel concept of grey knowledge. These three chapters demonstrate the distinctive and collaborative features of grey systems in uncertainty modelling. Starting from Chapter 6, the remaining chapters cover various real world application studies. Chapter 6 reveals how to apply grey systems in economic studies using agent-based systems. Chapter 7 demonstrates how grey systems help with the short term forecasting of traffic flow. Chapter 8 illustrates how grey models are applied to predict and manage Yellow River ice disasters in China. Chapter 9 showcases the application of grey systems to social network data analysis. Chapter 10 gives some case studies in applying grey systems into the energy-economic system. Chapter 11 explains how to select business strategy using grey stratified decisions model. Finally, Chapter 12 illustrates a cost level analysis for the components of the smartphones using greyness based quality function deployment. The cases studies in Chapter 6-12 covers a wider range of disciplines, geographical areas and different culture backgrounds, and it can serve as a useful reference for practitioners challenged by small and incomplete data. Combined with the connections covered in Chapter 3-5 and the basics in Chapter 1-2, this book will provide a convenient handbook for the practical application of grey systems.
Open Access
Explanation of terms of grey clustering evaluation models
(Emerald, 2017) Yang, Yingjie; Liu, Sifeng
Purpose – The purpose of this paper is to present the terms of grey clustering evaluation models. Design/methodology/approach –The definitions of basic terms about grey clustering evaluation models are presented one by one. Findings –The reader could know the basic explanation about the important terms about various grey clustering evaluation models from this paper. Practical implications –Many of our colleagues thought that unified definitions of key terms would be beneficial for both the readers and the authors . Originality/value – It’s a fundamental work to standardize all the definitions of terms for a new discipline. It’s also propitious to spread and universal of grey system theory.
Open Access
Explanation of terms of grey forecasting models
(Emerald, 2017) Liu, Sifeng; Yang, Yingjie
Purpose – The purpose of this paper is to present the terms of grey forecasting models and techniques. Design/methodology/approach –The definitions of basic terms about grey forecasting models and techniques are presented one by one. Findings –The reader could know the basic explanation about the important terms about various grey forecasting models and techniques from this paper. Practical implications –Many of our colleagues thought that unified definitions of key terms would be beneficial for both the readers and the authors. Originality/value – It’s a fundamental work to standardize all the definitions of terms for a new discipline. It’s also propitious to spread and universal of grey system theory.
Open Access
Explanation of terms of grey incidence analysis models
(Emerald, 2017) Liu, Sifeng; Zhang, Hongyang; Yang, Yingjie
Purpose – The purpose of this paper is to present the terms of grey incidence analysis models. Design/methodology/approach –The definitions of basic terms about various grey incidence analysis models are presented one by one. Findings –The reader could know the basic explanation about the important terms about various grey incidence analysis models from this paper. Practical implications –Many of our colleagues thought that unified definitions of key terms would be beneficial for both the readers and the authors . Originality/value – It’s a fundamental work to standardize all the definitions of terms for a new discipline. It’s also propitious to spread and universal of grey system theory.
Open Access
Grey Models and Their Roles in Data Analytics
(United Kingdom Simulation Society, 2018-06) Yang, Yingjie; Liu, Sifeng
This paper gives an introduction to the basic concepts of grey systems, grey numbers and grey models, and discusses their roles in data analysis of data science. The necessity of small data models is demonstrated, and their complementary functions to Big Data models are investigated. Based on these investigations, a novel framework for combining Big Data models with grey models is proposed.
Metadata only
Grey Number Sequence Forecasting Approach for Interval Analysis: A case of China's Gross Domestic Product Prediction
(Research Information Ltd, 2014) Xie, Naiming; Liu, Sifeng; Yuan, Chaoqing; Yang, Yingjie
Embargo
Grey relational analysis and natural language Processing
(IEEE, 2015-08-18) Khuman, A. S.; Yang, Yingjie; Liu, Sifeng
This paper investigates validity of using grey relational analysis (GRA) for natural language processing (NLP). The domain of NLP is one associated with inherent vagueness and abstraction, with many sub-domains all invoking their own associated uncertainties. Regardless of the particularisation, the main objective is understanding and making sense of linguistic lexicon. The inferencing and understanding of sentiment from natural language has been investigated thoroughly, however, the use of grey system theory in conjunction with NLP has yet to be explored in any great detail. Ergo, an introductory investigation into the effectiveness of using GRA on and with regards to NLP. This paper describes the feasibility of using grey system methodologies and tools, specifically the use of grey incidence, to provide a means for analysis of a sequence's geometric curve. The use of GRA provides one with the ability to inspect and infer sequences of data. Using this notion and by having a sequence represented as an input stream, it can be correlated against possible output commands. The use of grey incidence for quantifying and evaluating the correlation between what is inputted, against what output it is most similar to, is novel and should provide an additional facet to grey system theory.
Open Access
Grey Relational Analysis and Natural Language Processing to: Grey Language Processing
(RESEARCH INFORMATION LTD, 2016-01-01) Khuman, A. S.; Yang, Yingjie; Liu, Sifeng
This paper investigates the validity of using grey relational analysis (GRA) for natural language processing (NLP). The domain of NLP is one associated with inherent vagueness and abstraction; with many sub-domains, all invoking their own associated uncertainties. Regardless of the particularisation, the main objective is understanding and making sense of linguistic lexicon. The inferencing and understanding of sentiment from natural language has been investigated thoroughly, however, the use of grey system theory in conjunction with NLP has yet to be explored in any great detail. Ergo, an introductory investigation into the effectiveness of using GRA on and with regards to NLP. This paper describes the feasibility of using grey system methodologies and tools, specifically the use of grey incidence, to provide a means for analysis of a sequence's geometric curve. The use of GRA provides one with the ability to inspect and infer sequences of data. Using this notion and by having a sequence represented as an input stream, it can be correlated against possible output commands. The use of grey incidence for quantifying and evaluating the correlation between what is inputted, against what output it is most similar to, is novel and should provide an additional facet to grey system theory.
Open Access
Grey relational analysis model with cross-sequences and its application in evaluating air quality index
(Elsevier, 2023-06-26) Lu, Ningning; Liu, Sifeng; Du, Junliang; Fang, Zhigeng; Dong, Wenjie; Tao, Liangyan; Yang, Yingjie
It is important to detect the internal operating regularity in system developing with poor information. To identify the real relationship among multi factors, we propose a grey relational analysis (GRA) method inspired by the characteristics of sequences variation. The proposed model considers the changes of fluctuating sequences like cross-sequences both in domain time and between time intervals. To obtain the quantitative change about sequences, relative angle change is employed to determine the variation in each interval, and the relative angle oscillation change is utilized for measuring variations between intervals. To find the optimal time lag or time intervals, the corresponding cycles are extracted by time-delay models. The reliability of the proposed models will be verified through cases in identifying crucial factors for air quality, and the final detection will then be made. To compare with existing representative GRA models clearly, the relation between two fluctuating sequences shaped in cross-sequences is examined by the proposed model. The empirical results show that the relation degree between pollutants and air quality is reasonable. The compared experiment shows that the GRA for cross-sequences can effectively identify the relationship among fluctuating sequences and the impact of time-delay is small for the proposed model with similar shapes.
Metadata only
Grey Systems Analysis: Methods, Models and Applications
(Springer, 2022-12-08) Liu, Sifeng; Yang, Yingjie; Forrest, Jeffrey Yi-Lin
The book covers up-to-date theoretical and applied advances in grey systems theory from across the world and vividly presents the reader with the overall picture of this new theory and its frontier research. Many of the concepts, models and methods in the book are original by the authors, including simplified form of grey number, general grey number and the operations of grey numbers; the axiomatic system of buffer operators and a series of weakening and strengthening operators; a series of grey relational analysis models, including grey absolute, relative, synthetic, similarity, closeness, negative and three dimension degree, etc.; grey fixed weight clustering model, grey evaluation models based on center-point and end-point mixed possibility functions; original difference grey model (ODGM), even difference grey model (EDGM), discrete grey model (DGM), fractional grey models, self-memory grey models; multi-attribute intelligent grey target decision models, weight vector group with kernel and the weighted comprehensive clustering coefficient vector, and spectrum analysis of sequence operators, etc. This book will be appropriate as a reference and/or professional book for courses of grey system theory for graduate students or high-level undergraduate students, majoring in areas of science, technology, agriculture, medicine, astronomy, earth science, economics, and management. It can also be utilized by researchers and practitioners in research institutions, business entities, and government agencies.
Metadata only
Grey systems and uncertainty modelling
(Springer, 2023-02-05) Yang, Yingjie; Khuman, A. S.; Liu, Sifeng
Information can, and often is, rather uncertain; with only partial information initially being made available, from which one would be able to hopefully provide for a solution. The information itself may contain conflicts that have arisen from the possible different sources used to acquire it. In addition, the information may be viewed and interpreted differently by different cohorts, this in itself can be the cause of extenuating circumstances. These are just some of the issues that one can face with uncertain information. These issues can understandably create problems when considering the deployment of applications. Being able to cater for the volatility that is inherently present in uncertainty, becomes an objective with high importance and precedence.
Open Access
A greyness reduction framework for prediction of grey heterogeneous data
(Springer, 2020-05-23) Li, Chong; Yang, Yingjie; Liu, Sifeng
Existing operational rules of interval grey numbers do not make full use of possible background information when determining the interval boundaries, and this may result in inconsistent results if applying different logical operations. This paper finds that multiplication and division rules of interval grey numbers do not meet the calculation rule of inverse operators. Direct solution and inverse solution of the same interval grey number object may differ not only in numerical ranges but also in greyness degrees. To improve the accuracy of grey number calculation, new operational rules for multiplication and division of interval grey numbers are proposed. Then the traditional prediction modeling method of grey heterogeneous data is refined and expanded by integrating a greyness reduction preprocessing, which is based on the proposed calculation rules. Application of the expanded heterogeneous interval grey number prediction model to a stock replenishment scheduling problem in emergency rescue scenarios is included to illustrate the new operational rules of grey numbers and their application in prediction algorithm, and the proposed approach is compared with other existing methods to demonstrate its effectiveness.
Open Access
Micro-macro dynamics of the online opinion evolution: an asynchronous network model approach
(John Wiley & Sons Ltd, 2020-09-02) Li, Chong; Yang, Yingjie; Liu, Sifeng
This paper investigates the complex relationship between endogenous and exogenous, deterministic and stochastic stimulating factors in public opinion dynamics. An asynchronous multi-agent network model is proposed to explore the interaction mechanism between individual opinions and the public opinion in online multi-agent network community, including both the micro and the macro patterns of opinion evolution. In addition, based on random network models, a novel algorithm is provided for opinion evolution prediction. The model property analysis and numerical experiments show that the proposed asynchronous multi-agent network model can assimilate and explain some interesting phenomena that are observed in the real world. Further case studies with numerical simulation and real-world applications confirm the feasibility and flexibility of the proposed model in public opinion analysis. The results challenge the common perception that mass media or opinion facilitators play the fundamental role in controlling the development trends of public opinion. This study shows that the formation and evolution of public opinion in the presence of opinion leaders depend also on an individual’s emotional inertia and conformity pressures from peers in the same topic group.

Browsing by Author "Liu, Sifeng"

Results Per Page

Sort Options