Influence maximization problem is trying to identify a set of nodes by which the spread of influence, diseases or information is maximized. The optimization of influence by finding such a set is NP-hard problem and a key issue in analyzing complex networks. In this paper, a new greedy and hybrid approach based on a community detection algorithm and an MADM technique (TOPSIS) is proposed to cope with the problem, called, 'Greedy TOPSIS and Community-Based' (GTaCB) algorithm. The paper concisely introduces community detection and TOPSIS technique, then it presents the pseudo-code of the proposed algorithm. Afterwards, it compares the performance of the solution which is found by GTaCB with some well-known greedy algorithms, based on Degree Centrality, Closeness Centrality, Betweenness Centrality, PageRank as well as TOPSIS, from two aspects: diffusion quality and diffusion speed. In order to evaluate the performance of GTaCB, computational experiments on nine different types of real-world networks are provided. The tests are conducted via one of the renowned epidemic diffusion models, namely, Susceptible-Infected-Recovered (SIR) model. The simulations exhibit that in most of the cases the proposed algorithm significantly outperforms the others, chiefly as number of initial nodes or probability of infection increases.
Influence maximization is a well-known problem in the social network analysis literature which is to find a small subset of seed nodes to maximize the diffusion or spread of information. The main application of this problem in the real-world is in viral marketing. However, the classic influence maximization is disabled to model the real-world viral marketing problem, since the effect of the marketing message content and nodes’ opinions have not been considered. In this paper, a modified version of influence maximization which is named as “opinion-aware influence maximization” (OAIM) problem is proposed to make the model more realistic. In this problem, the main objective is to maximize the spread of a desired opinion, by optimizing the message content, rather than the number of infected nodes, which leads to selection of the best set of seed nodes. A nonlinear bi-objective mathematical programming model is developed to model the considered problem. Some transformation techniques are applied to convert the proposed model to a linear single-objective mathematical programming model. The exact solution of the model in small datasets can be obtained by CPLEX algorithm. For the medium and large-scale datasets, a new genetic algorithm is proposed to cope with the size of the problem. Experimental results on some of the well-known datasets show the efficiency and applicability of the proposed OAIM model. In addition, the proposed genetic algorithm overcomes state-of-the-art algorithms.
Fault diagnosis plays an essential role in reducing the maintenance costs of rotating machinery manufacturing systems. In many real applications of fault detection and diagnosis, data tend to be imbalanced, meaning that the number of samples for some fault classes is much less than the normal data samples. At the same time, in an industrial condition, accelerometers encounter high levels of disruptive signals and the collected samples turn out to be heavily noisy. As a consequence, many traditional Fault Detection and Diagnosis (FDD) frameworks get poor classification performances when dealing with real-world circumstances. Three main solutions have been proposed in the literature to cope with this problem: (1) the implementation of generative algorithms to increase the amount of under-represented input samples, (2) the employment of a classifier being powerful to learn from imbalanced and noisy data, (3) the development of an efficient data preprocessing including feature extraction and data augmentation. This paper proposes a hybrid framework which uses the three aforementioned components to achieve an effective signal based FDD system for imbalanced conditions. Specifically, it first extracts the fault features, using Fourier and wavelet transforms to make full use of the signals. Then, it employs Wasserstein Generative Adversarial with Gradient Penalty Networks (WGAN-GP) to generate synthetic samples to populate the rare fault class and enrich the training set. Moreover, to achieve a higher performance a novel combination of Convolutional Long Short-term Memory (CLSTM) and Weighted Extreme Learning Machine (WELM) is also proposed. To verify the effectiveness of the developed framework, different bearing datasets settings on different imbalance severities and noise degrees were used. The comparative results demonstrate that in different scenarios GAN-CLSTM-ELM significantly outperforms the other state-of-the-art FDD frameworks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.