Lean Yu

Verified

Lean verified their affiliation via an institutional email.

Learn more

Verified

Lean verified their affiliation via an institutional email.

Learn more

• PhD
• Distinguished Professor/PhD Supervisor at Sichuan University

Anchor-based methods are proposed to make use of anchors to produce an affinity matrix of objects to improve the scalability of traditional spectral clustering (SC). Nevertheless, the membership heterogeneity of objects inside a cluster, which would bring about low quality of anchors and hurt the clustering accuracy, is commonly neglected by existi...

Precise corporate credit risk (CCR) prediction empowers investors, banks and other financial institutions to build risk prevention and crisis evasion mechanisms. Currently, CCR prediction gradually trends towards integrating multi-source data for more accurate prediction, but the effective fusion of these data is still insufficient. For this purpos...

The acceptance of academic papers involves a complex peer-review process that requires substantial human and material resources and is susceptible to biases. With advancements in deep learning technologies, researchers have explored automated approaches for assessing paper acceptance. Existing automated academic paper rating methods primarily rely...

The prevalent challenge of class sparsity issues in credit risk classification commonly focuses on instance-view solutions, while feature-view solutions are overlooked. For this purpose, this paper designs a dual-view ensemble learning model to tackle class sparsity and its associated traits of overlap, noise, and irrelevance. The model comprises t...

To solve the high-dimensional issue in credit risk assessment, a hybrid clustering and boosting tree feature selection method is proposed. In the hybrid methodology, an improved minimum spanning tree model is first used to remove redundant and irrelevant features. Then three embedded feature selection approaches (i.e., Random Forest, XGBoost, and A...

This paper proposes a novel hybrid forecasting model, TDE-CNN, to model the complex dynamics of crude oil price movements. The model integrates Time-Delay Embedding (TDE) Method with a Convolutional Neural Network (CNN) to leverage both spatial and temporal information. The TDE-CNN model uses the TDE method to transform raw crude oil data into high...

Machine learning methods are widely used to evaluate the risk of small- and medium-sized enterprises (SMEs) in supply chain finance (SCF). However, there may be problems with data scarcity, feature redundancy, and poor predictive performance. Additionally, data collected over a long time span may cause differences in the data distribution, and clas...

In the current wave of technological innovation, the scientific and technological evaluation system not only plays a crucial role in assessing scientific research achievements but also serves as a driving force for advancing scientific research. To enhance the effectiveness and accuracy of peer review in the scientific and technological evaluation...

Accurate risk estimation formulates the essential foundation of risk management in the Bitcoin market. A new three-stage ensemble method is introduced to solve the inherent instability of single models when estimating Bitcoin risk. In the proposed method, single models are first employed to estimate risk using a training dataset. Second, a model se...

This paper investigates major railroad projects' emergency response decision-making process, considering timeliness, cost-effectiveness, and public involvement. It explores the optimal collaborative strategy among local governments, enterprises, and the public during construction. The study introduces an innovative tripartite collaborative decision...

Forecasting refined oil sales is essential in energy supply chain management. However, accurate forecasting is limited by several factors, including multiple influences of external features, heterogeneity of different gasoline stations, and difficulty in balancing linear and nonlinear forecasting. To address these issues, we propose a novel variabl...

In credit fraud detection practice, certain fraudulent transactions often evade detection because of the hidden nature of fraudulent behavior. To address this issue, an increasing number of positive-unlabeled (PU) learning techniques have been employed by more and more financial institutions. However, most of these methods are designed for single d...

This repository provides data for the problem and code for the method. The main folders are data and code.

Social media sentiment influences housing market trading and policy‐making in China. To explore the multiscale relationship between social media sentiment and house price index (HPI) and improve prediction performance, a sentiment‐based decomposition–ensemble approach is proposed for HPI forecasting. In this approach, five steps, that is, sentiment...

It is prone to overfitting and poor generalization ability for imbalanced small sample datasets in modeling. Auxiliary data is an effective solution. However, there may be data distribution differences between auxiliary data and small sample data, and the presence of noise samples affects the prediction performance. To address this issue, we propos...

The OPEC+, composed of the Organization of the Petroleum Exporting Countries (OPEC) and non-OPEC oil-producing countries, exerts considerable influence over the global crude oil market. However, existing literature lacks a comprehensive application of this factor in oil price forecasting, primarily due to the complexity of measuring such policy evo...

Air pollution control in the U.S. has evolved into a comprehensive policy system spanning from the federal to the state level over time. A unified quantitative analysis of policy intensity can shed light on the policy evolution across different levels, the influence of partisan and regional factors on policy, and the relationships with emissions of...

To address the high‐dimensional issues in credit risk assessment, an improved multilayer restricted Boltzmann machine (RBM) based feature extraction method is proposed. In the improved multilayer RBM methodology, the reconstruction error method is first applied to ensure the number of RBM layers to construct an optimal model and then the weighted p...

Whether tax incentives can forecast the performance of tourism firms after restructuring remains an unsolved puzzle. Regarding restructurings as events, this work proposes an event-driven forecasting framework, in which an adaptive integrated model based on similar scenarios with optimal boundaries is designed to investigate the ability of tax ince...

Over the last 20 years, China’s infertility rate has risen from 3% to 12.5%–15%. Infertility has become the third largest disease following cancer and cardiovascular disease. Then, the in vitro fertilization and embryo transfer (IVF-ET) becomes more and more important in infertility treatment field. However, the reported success rate for IVT-ET is...

The security of credit card fraud detection (CCFD) models based on machine learning is important but rarely considered in the existing research. To this end, we propose a black-box attack-based security evaluation framework for CCFD models. Under this framework, the semisupervised learning technique and transfer-based black-box attack are combined...

The security of credit card fraud detection (CCFD) models based on machine learning is important but rarely considered in the existing research. To this end, we propose a black-box attack-based security evaluation framework for CCFD models. Under this framework, the semi-supervised learning technique and transfer-based black-box attack are combined...

This study examines the relationship and risk spillover between Bitcoin, crude oil, and six traditional markets (the US stock, Chinese stock, gold, bond, currency, and real estate markets) from 2019 to 2020, during which the coronavirus disease 2019 (COVID-19) outbreak occurred as well. We first discuss the static relationship between Bitcoin and t...

Knowledge embeddedness may exert a crucial role in improving synergetic innovation performance in the knowledge economy era. However, theoretical deduction and empirical investigation on the effects of knowledge embeddedness have not yet reached a consensus. One primary reason caused this result is that the different levels of knowledge embeddednes...

To fulfill the goals of sustainable development and emission reduction, China has established the consumption side renewable portfolio standards (RPS) and domestic nationwide carbon emission trading (CET) scheme. However, there are intricate interrelationships between the two mechanisms as they overlap in policy goals and implementation measures, w...

Based on the medical waste quantity and patient data during the corona virus disease 2019 (COVID-19) outbreak in China, this study used scenario analysis to quantitatively analyze the temporal and spatial evolution of medical waste generation during the pandemics. First, the results show that the estimated medical waste per capita reached 15.4 kg/d...

In order to deeply analyze the internal and external influencing factors of mutton price fluctuation, a multi-scale analytical framework was proposed in regards to the fluctuation mechanism analysis of mutton price in the Xinjiang region of China. By combining data decomposition and correlation analysis, this paper investigated the relationship bet...

Forecast combination, a well-established technique for improving forecasting accuracy, investigates the integration of competing forecasts to produce a composite superior to individual forecasts. In this study, we propose a novel forecast combination method that would reduce overfitting risk and improve forecast’s generalization ability. To capture...

The crude oil market is known to be subject to the influence of transient and extreme events. The rare and infrequent nature of these events leads to problems such as a lack of data for the estimation of reliable risk measures in the crude oil market. In this paper, an innovative MEMD-BiGAN risk forecasting methodology combining the power of multi...

To improve the prediction accuracy of short-term load series, this paper proposes a hybrid model based on a multi-trait-driven methodology and secondary decomposition. In detail, four steps were performed sequentially, i.e., data decomposition, secondary decomposition, individual prediction, and ensemble output, all of which were designed based on...

To improve the forecasting accuracy of short-term wind power, an intermittency-trait-driven methodology is proposed in this paper. In the proposed methodology, four main steps, i.e., intermittency-trait-driven data decomposition, intermittency-trait-driven mode reconstruction, intermittency-trait-driven component prediction and intermittency-trait-...

Bending strength of concrete is one of the significant indexes to measure the mechanical properties of concrete. A reliable prediction about the bending strength of concrete is of great importance to maintain the health state and service life of concrete. However, it is difficult to obtain reliable data of large samples due to the high cost, seriou...

Technological innovations in the power industry can help reduce electricity consumption but may also have a negative result due to rebound effects. Estimation and refinement of electricity demand rebound effects are important for assessing the impact of technological innovations. For this purpose, this paper first constructs a Log Mean Divisia Inde...

Data scarcity is a serious issue in credit risk assessment for some emerging financial institutions. As a typical category of data scarcity, small sample with high dimensionality often leads to the failure to build an effective credit risk assessment model. To solve this issue, a Wasserstein generative adversarial networks (WGAN)-based data augment...

There is a dilemma regarding the accuracy and reality of vehicle trajectory prediction. Balancing and predicting the effective trajectory is a topic of debate in autonomous driving. We investigated this issue using knowledge-driven and data-driven methods to estimate the performance of the two most common methods and found that improving the accura...

As a typical category of data scarcity, small sample often makes it difficult to build a reliable machine learning model in credit risk assessment, and thus many virtual sample generation (VSG) methods have been proposed for sample augmentation based on sample distribution. In particular, when small sample with low dimensionality exists in credit d...

A new framework is proposed to comprehensively explore the robustness (structure and function) of knowledge networks under different targeted attacks. Results show that (1) recalculated-based attacks cause greater damage than initial-based attacks. Meanwhile, a higher price is needed to destroy the structure than disrupt the function of the knowled...

In order to solve the problem of uncertain misclassification costs and class distributions in credit scoring tasks, an uncertainty-oriented credit scoring framework based on a multi-objective feature selection strategy is proposed in this study. This proposed framework searches for a pool of Pareto-optimal credit scoring models with different featu...

The enterprise default discriminant analysis provides a basis for decision making on bank loans and corporate bond investments. As such, this study attempts to address the following two questions: how should an optimal feature set with the highest default discriminant accuracy be selected, and how can a default discriminant model with the lowest to...

Purpose The purpose of this paper is to provide new insights for managing knowledge reuse in terms of the duality of innovator personality. Continuously developing new products is crucial for firms to maintain and enhance their competitive advantages. However, the limited and highly specialized knowledge can cause innovators of firms to face diffic...

A domestic electrical storage water heater (i.e., DESWH) is one of the 14 products listed in China’s Waste Electrical and Electronic Products Disposal Catalogue (Batch 2). Due to the lack of systematic quantitative analysis on the waste quantity and recovery value of a DESWH, a multi-data source-based hybrid methodology based on quarterly sales dat...

In order to improve the accuracy of the short-term wind power forecasting, a novel complexity-trait-driven rolling decomposition-reconstruction-ensemble forecasting model is proposed to predict short-term wind power. In this model, four steps are involved, i.e., data decomposition, mode reconstruction, component prediction and ensemble prediction,...

Emerging event-based social networks (EBSNs), such as Meetup, have grown rapidly and become popular in recent years. EBSNs differ from conventional social networks such as Facebook in that they not only involve online social interactions but also include offline, in-person interactions. Thus, EBSNs are naturally heterogeneous and possess more valua...

The traditional portfolio theory has relied heavily on historical asset returns while ignoring future information. Based on ensemble learning and maximum Sharpe ratio portfolio theory, this paper proposes a two-stage portfolio optimization method by considering asset forecast information, aming to improve the performance and robustness of a portfol...

To solve the high-dimensionality issue and improve its accuracy in credit risk assessment, a high-dimensionality-trait-driven learning paradigm is proposed for feature extraction and classifier selection. The proposed paradigm consists of three main stages: categorization of high dimensional data, high-dimensionality-trait-driven feature extraction...

The performance evaluation of scientific research institutions is significant to the development of universities. It can improve the needs of universities' own competitiveness, and keep the overall development of universities on a healthy track. In this paper, the production efficiency of research laboratories is evaluated by the combination of eco...

In order to predict the gasoline consumption in China, this paper propose a novel data-trait-driven rolling decomposition-ensemble model. This model consists of five steps: the data trait test, data decomposition, component trait analysis, component prediction and ensemble output. In the data trait test and component trait analysis, the original ti...

In order to improve the prediction performance in oil price forecasting, a novel memory-trait-driven decomposition-reconstruction-ensemble learning paradigm is proposed for oil price forecasting. The proposed methodology consists of four steps, i.e., data decomposition for original complex time series, component reconstruction for decomposed compon...

The shipping index has the characteristics of violent fluctuation, so its volatility is difficult to predict. To better predict the volatility of the shipping market, this paper proposes an AR-SVR-GARCH model, which combines traditional time series analysis and modern machine learning methods. This model overcomes linear limitations of traditional...

To address the drawback of single machine learning prediction model which cannot capture the complex hidden factors of crude oil price, ensemble learning method has been widely verified as an excellent solution for crude oil price forecasting. In ensemble learning, diversity strategy is one of the most important determinants to obtain good performa...

The power sector has tremendous technological potential for decarbonization, hence, the decarbonization of China's power sector is crucial to the successful implementation of national carbon emissions reduction plan. In this study, a decarbonization model with consideration of both technological progress and cross-regional power transmission for Ch...

Developed by the World Economic Forum in collaboration with Accenture, the Energy Architecture Performance Index (EAPI) looks at trends and the real performance of countries' energy systems, and provides the latest available global energy data, aiding policy formation by offering a reliable indicator of strengths and target areas for improvement. I...

Accurate demand forecast can help improve teleconsultation efficiency. But teleconsultation demand forecast has not been reported in existing literature. For this purpose, the study proposes a novel model based on deep learning algorithm for daily teleconsultation demand forecast to fill in the research gap. Because of the significant effect of hol...

In this paper, an effective rolling decomposition-ensemble model is proposed for quarterly gasoline consumption forecasting in China. In this model, three steps, data decomposition, component prediction and ensemble output, are involved. In the data decomposition, wavelet decomposition and ensemble empirical mode decomposition are used due to few a...

We propose an efficient method for pricing foreign currency options given that foreign currency returns have heavy-tailed distributions. In our approach, the heavy tail of the distribution are modeled using Student-t distribution rather than normal distribution, and the parameters of Student-t distribution are estimated using the Method of Moments...

In the original publication of the article, the corresponding author name has been misspelled and now the same has been corrected here.

Waste electrical and electronic equipment (viz., WEEE or e-waste) is the fastest-growing type of hazardous solid waste in the worldwide. The accurate prediction of the amount of e-waste might help improve the efficiency of e-waste disposal. In this study, a novel decomposition-ensemble-based hybrid forecasting methodology that integrates variationa...

This paper is concerned with proposing a new mechanism to re-construct the published composite indicators that are conventionally aggregated in terms of equal weighting scheme, by means of taking all possible preferences among the indicators into account. Regarding to each preference, we apply a sophisticated mathematical transformation to formulat...

In this study, a comprehensive model for suitable carrying capacity of resources and environment was proposed based on ecological footprint method. Using the spatiotemporal distribution data of land use in Chongqing Section of Three Gorges Reservoir Area from 2001 to 2016, the response changes of carrying capacity of resources and environment under...

Missing data has become an increasingly serious problem in credit risk classification. A one-hot encoding-based data preprocessing method is proposed to solve the missing data problem in credit classification. In this paradigm, the proposed missing-data preprocessing method is first used to deal with missing values to fill in the incomplete dataset...

In this paper, a novel dual-weighted fuzzy proximal support vector machine (FPSVM) model hybridizing fuzzy set theory (FST) and proximal support vector machine (PSVM) is proposed for credit risk analysis. In the proposed model, the fuzzy memberships are introduced into both objective function and constraint conditions of PSVM model to make full use...

This paper formulates a novel integrated measure for energy market efficiency, by investigating different aspects of the market performance. Different from most existing models focusing on one certain aspect, the novel measure especially takes into consideration the self-similarity (or system memo ability or long-term persistence) via fractality, t...

In this paper, a dual-voting-based learning paradigm is proposed to solve attribute noise problem in credit risk classification. In the proposed learning paradigm, three stages are involved. In the first stage, four indexes are introduced to evaluate the noise level of attributes. In the second stage, attributes with different noise levels are divi...

Blockchain has attracted much attention in recent years with the development of cryptocurrency and digital assets. As the underlying technology of cryptocurrency, blockchain has numerous benefits, such as decentralization, collective maintenance, tamper-resistance, traceability, and anonymity. The potential of the block-chain technology (BT) is wid...

Aiming at the complex and restrictive characteristics of human resource allocation in multiple scientific university research projects, an improved pigeon-inspired optimization (IPIO) algorithm is proposed wherein loss minimization and the shortest project delay time are considered as optimization goals. Firstly, mathematical modelling of the probl...

The emerging online peer to peer (P2P) lending platforms have only a small number of samples in the early stage, it is thus unable to conduct an efficient credit risk assessment on internet loan applicants. In order to solve the sample shortage issue, a virtual sample generation (VSG) methodology integrating multi-distribution mega-trend-diffusion...