•An SNA based conflict detection and elimination decision making process is presented.•The impact of relationship strength on trust propagation efficiency is considered.•Multi-path trust propagation ...operator is presented to complete the social network.•Nonlinear optimization model guarantees a sufficient reduction of group conflict.•We promote the modification of the assessments by finding the optimal solution.
The paper proposes a Trust Relationship-based Conflict Detection and Elimination decision making (TR-CDE) model, applicable for Large-scale Group Decision Making (LSGDM) problems in social network contexts. The TR-CDE model comprises three processes: a trust propagation process; a conflict detection and elimination process; and a selection process. In the first process, we propose a new relationship strength-based trust propagation operator, which allows to construct a complete social network by considering the impact of relationship strength on propagation efficiency. In the second process, we define the concept of conflict degree and quantify the collective conflict degree by combining the assessment information and trust relationships among decision makers in the large group. We use social network analysis and a nonlinear optimization model to detect and eliminate conflicts among decision makers. By finding the optimal solution to the proposed nonlinear optimization model, we promote the modification of the assessments from the DM who exhibits the highest degree of conflict in the process, as well as guaranteeing that a sufficient reduction of the group conflict degree is achieved. In the third and last process, we propose a new selection method for LSGDM that determines decision makers’ weights based on their conflict degree. A numerical example and a practical scenario are implemented to show the feasibility of the proposed TR-CDE model.
•A comprehensive review on the integration of machine learning into meta-heuristics.•A unified taxonomy to provide a common terminology and classification.•Classification of numerous articles based ...on different characteristics.•Technical discussions on advantages, limitations, requirements, and challenges.•Proposition of promising future research directions.
In recent years, there has been a growing research interest in integrating machine learning techniques into meta-heuristics for solving combinatorial optimization problems. This integration aims to lead meta-heuristics toward an efficient, effective, and robust search and improve their performance in terms of solution quality, convergence rate, and robustness.
Since various integration methods with different purposes have been developed, there is a need to review the recent advances in using machine learning techniques to improve meta-heuristics. To the best of our knowledge, the literature is deprived of having a comprehensive yet technical review. To fill this gap, this paper provides such a review on the use of machine learning techniques in the design of different elements of meta-heuristics for different purposes including algorithm selection, fitness evaluation, initialization, evolution, parameter setting, and cooperation. First, we describe the key concepts and preliminaries of each of these ways of integration. Then, the recent advances in each way of integration are reviewed and classified based on a proposed unified taxonomy. Finally, we provide a technical discussion on the advantages, limitations, requirements, and challenges of implementing each of these integration ways, followed by promising future research directions.
Smartphones have become pervasive due to the availability of office applications, Internet, games, vehicle guidance using location-based services apart from conventional services such as voice calls, ...SMSes, and multimedia services. Android devices have gained huge market share due to the open architecture of Android and the popularity of its application programming interface (APIs) in the developer community. Increased popularity of the Android devices and associated monetary benefits attracted the malware developers, resulting in big rise of the Android malware apps between 2010 and 2014. Academic researchers and commercial antimalware companies have realized that the conventional signature-based and static analysis methods are vulnerable. In particular, the prevalent stealth techniques, such as encryption, code transformation, and environment-aware approaches, are capable of generating variants of known malware. This has led to the use of behavior-, anomaly-, and dynamic-analysis-based methods. Since a single approach may be ineffective against the advanced techniques, multiple complementary approaches can be used in tandem for effective malware detection. The existing reviews extensively cover the smartphone OS security. However, we believe that the security of Android, with particular focus on malware growth, study of antianalysis techniques, and existing detection methodologies, needs an extensive coverage. In this survey, we discuss the Android security enforcement mechanisms, threats to the existing security enforcements and related issues, malware growth timeline between 2010 and 2014, and stealth techniques employed by the malware authors, in addition to the existing detection methods. This review gives an insight into the strengths and shortcomings of the known research methodologies and provides a platform, to the researchers and practitioners, toward proposing the next-generation Android security, analysis, and malware detection techniques.
•A new hybrid algorithm (logit leaf model) is proposed for customer churn prediction.•It is designed to perform well in terms of both accuracy and interpretability.•Its competitive performance is ...apparent from an extensive benchmarking experiment.•Its ability to deliver actionable insights is demonstrated in a case study.
Decision trees and logistic regression are two very popular algorithms in customer churn prediction with strong predictive performance and good comprehensibility. Despite these strengths, decision trees tend to have problems to handle linear relations between variables and logistic regression has difficulties with interaction effects between variables. Therefore a new hybrid algorithm, the logit leaf model (LLM), is proposed to better classify data. The idea behind the LLM is that different models constructed on segments of the data rather than on the entire dataset lead to better predictive performance while maintaining the comprehensibility from the models constructed in the leaves. The LLM consists of two stages: a segmentation phase and a prediction phase. In the first stage customer segments are identified using decision rules and in the second stage a model is created for every leaf of this tree. This new hybrid approach is benchmarked against decision trees, logistic regression, random forests and logistic model trees with regards to the predictive performance and comprehensibility. The area under the receiver operating characteristics curve (AUC) and top decile lift (TDL) are used to measure the predictive performance for which LLM scores significantly better than its building blocks logistic regression and decision trees and performs at least as well as more advanced ensemble methods random forests and logistic model trees. Comprehensibility is addressed by a case study for which we observe some key benefits using the LLM compared to using decision trees or logistic regression.
With the rapid development of remote sensing technology, the monitoring of land surface ecological status (LSES) based on remote sensing has made remarkable progress, which has a positive ...contribution on improving the regional ecological environment and promoting the realization of Sustainable Development Goals (SDGs). Among them, the proposed Remote Sensing-based Ecological Index (RSEI) becomes the most widely used model in the current application of remote sensing-based LSES monitoring due to its complete derived from remote sensing images and no subjective intervention. RSEI is not flawless either, and it still suffers from some uncertainties in its application in multiple scenarios. However, compared to the extensive applied research, work on the instability assessment and improvement of RSEI is particularly scarce and urgently needed. Therefore, in this paper, we analyzed the possible instabilities in the RSEI calculation process and proposed various inversion models to evaluate their accuracy and stability in time-series LSES monitoring. The results indicated that the existing normalized RSEI is relatively stable for the characterization of single-phase LSES, however, there is a high risk in the time-series analysis or cross-regional comparison due to the interference of component extremes. The standard deviation discretized DRSEIs proposed in this paper perform better in both single-phase and long-term dynamics LSES assessments and are more consistent with the real land cover changes. Also, compared with the approach that measures LSES dynamics using time-series regional RSEI mean values, the DRSEIs change detection results can reveal the spatial heterogeneity of regional LSES dynamics more effectively and provide a finer reference for the formulation and implementation of ecological protection policies.
Display omitted
•The applicability of normalization and standardization in RSEI is well discussed.•The instability in the RSEI time-series assessment was found and replicated.•A discrete RSEI (DRSEIs) is proposed, which is more suitable for dynamic analysis.
Synchronization is a very important phenomenon in the nervous system, which is closely related to the encoding, integration and transmission of information. In this paper, synchronization and ...transition of a two-compartment respiratory neuron model under transcranial magnetic stimulation (TMS) are studied from the perspective of synchronization degree for the first time. We are established the correlation degree with synchronization, and discussed the firing mode and transition rule of the neurons in the two-compartment compartment pre-Bötzinger complex (PBC) by means of bifurcation theory and Lyapunov index. The results show that the synchronization of neurons has a great influence under TMS, which is embodied in the fact that the somatic will experience a peak firing and a transition from bursting to resting under the magnetic stimulation,which was a phenomenon never before shown in PBC neurons. These results fully reveal the dynamic behavior of PBC nervous system under TMS, and provide theoretical value for further understanding of respiratory rhythm.
•The dynamic behavior of PBC under TMS was studied.•The degree of synchronization was first used to study the respiratory rhythm of PBC.•Abundant bifurcation phenomenon is obtained.
•We present new operations of the probabilistic linguistic term sets.•We build a correlation measure-based consensus reaching process.•We propose a new outranking method: gained and lost dominance ...score method.•We give a probabilistic linguistic gained and lost dominance score method.•We validate the method with the green enterprise selection problem.
This paper proposes a comprehensive Multiple Criteria Group Decision Making (MCGDM) method with probabilistic linguistic information based on a new consensus measure and a novel outranking method, Gained and Lost Dominance Score (GLDS). Firstly, new operations of the probabilistic linguistic term sets are introduced based on the adjusted rules of probabilistic linguistic term sets and the linguistic scale functions for semantics of linguistic terms. After defining a new consensus measure based on the correlation degree between probabilistic linguistic term sets, we develop a consensus reaching method to improve the consensus degree of a group. To rank alternatives reasonably, we further propose the GLDS method which considers both the “group utility” and the “individual regret” values. The core of the GLDS is to calculate the gained and lost dominance scores that the optimal solution dominates all other alternatives in terms of the net gained dominance flow and the net lost dominance flow. Then, we integrate the GLDS ranking method with the consensus reaching process and develop a consensus-based PL-GLDS method to solve the MCGDM problems with probabilistic linguistic information. Finally, the proposed method is validated by a case study of selecting optimal green enterprises. Some comparative analyses are given to show the efficiency of the proposed method.
•A deep long short-term memory (LSTM) network is developed for nonlinear structural response modeling.•Two input-output schemes (LSTM-s and LSTM-f) are presented.•The deep learning model is capable ...of modeling both elastic and inelastic response of buildings.•An unsupervised learning algorithm is used to cluster the seismic inputs for training enhancement.•The approach was successfully verified by both numerical and experimental examples.
This paper presents a comprehensive study on developing advanced deep learning approaches for nonlinear structural response modeling and prediction. Two schemes of the long short-term memory (LSTM) network are proposed for data-driven structural seismic response modeling. The proposed deep learning model, trained on available datasets, is capable of accurately predicting both elastic and inelastic response of building structures in a data-driven fashion as opposed to the classical physics-based nonlinear time history analysis using numerical methods. In addition, an unsupervised learning algorithm based on a proposed dynamic K-means clustering approach is established to cluster the seismic inputs in order to (1) generate the least but the most informative datasets for training the LSTM and (2) improve the prediction accuracy and robustness of the model trained with limited data. The performance of the proposed approach is successfully demonstrated through three proof-of-concept studies that include a nonlinear hysteretic system, a real-world building with field sensing data, and a steel moment resisting frame. The results show that the proposed LSTM network is a promising, reliable and computationally efficient approach for nonlinear structural response prediction, and offers significant potential in seismic fragility analysis of buildings for reliability assessment.
•A nonlinear dynamic model is proposed to predict the skidding behavior.•Various effects are considered for roller bearings under time-variable load.•Local skidding for each roller could still be ...found and vary periodically.•Time-variable load lead to the frequency of time-varying slipping velocity change.•The results are useful for design and monitoring of rotating machinery.
Display omitted
Considering the radial clearance, roller crown profile and discontinuous contact between the roller and cage, a nonlinear dynamic model for skidding behavior of the cylindrical roller bearing is established based upon the Hertz contact theory and elastohydrodynamic lubrication. Through comparisons with both reference and experimental results, the proposed model is verified. Various load conditions are considered for their effects on single roller’s skidding behavior. It is shown that local skids exist and periodically vary with the roller revolution, especially for the roller entering and leaving the load region. Increasing the values of radial load, bending moment or amplitude of time-variable load all reduces the maximum roller slipping velocity, which means that the roller skids are attenuated. After considering the time-variable radial load, the frequency of time-varying slipping velocity is not the orbital speed but the combination of the inner race frequency and orbital speed.
Shear fuses are structural elements that protect surrounding members from damages by undergoing substantial yielding, and then are easily replaced after a major earthquake event. Butterfly-shaped ...shear fuse is a promising type of structural system, which can effectively align member's flexural capacity to the imposed moment demand due to its unique geometry. Although recent studies suggest that butterfly-shaped fuses exhibit substantial ductility and energy dissipation, their impact on the global performance of multi-story buildings requires further investigation. This study presents a comprehensive risk-based evaluation of a six-story eccentrically braced steel frame retrofitted with butterfly-shaped fuses. Two nonlinear finite element models of the original prototype building and retrofitted building with butterfly-shaped fuses are developed in OpenSees and incremental dynamic analysis is conducted. The results are used to derive global and story-based fragility and seismic demand hazard curves. Furthermore, earthquake-induced losses associated with structural and non-structural assemblies are quantified and the impact of butterfly-shaped fuses on the distribution of story acceleration and drift demands are evaluated.
The results show that butterfly-shaped fuses significantly improve the structure's performance in terms of all drift-related damage states and the improvement is more pronounced at severe damage states. In particular, the risk of exceeding complete damage state in the retrofitted building's lifetime is reduced to approximately one-fourth of the original building's values. Furthermore, shear fuses effectively mitigate weak story formation at lower stories due to their large energy dissipation and ductility. The improved drift-related performance reduces the drift-induced loss of structural and non-structural assemblies, resulting in 44.64% smaller total annual loss for the studied building. In addition, although butterfly-shaped fuses reduce the probability of exceeding slight damage state for the floor acceleration, their impact is negligible at higher acceleration-related damage states.
•Probabilistic seismic performance is conducted for rehabilitated EBF systems.•The results are used to derive global and story-based fragility and seismic demand hazard curves.•Loss evaluation of a multi-story steel building equipped with butterfly-shaped fuses is conducted.•Butterfly-shaped system reduces the induced loss by improving the drift-related performance.