Showing papers in "Applied Soft Computing in 2020"

TL;DR: A comprehensive literature review on DL studies for financial time series forecasting implementations and grouped them based on their DL model choices, such as Convolutional Neural Networks (CNNs), Deep Belief Networks (DBNs), Long-Short Term Memory (LSTM).

TL;DR: This study aims to investigate the impact of fourteen data normalization methods on classification performance considering full feature set, feature selection, and feature weighting and suggests a set of the best and the worst methods combining the normalization procedure and empirical analysis of results.

TL;DR: To perform parameter optimization and feature selection simultaneously for SVM, an improved whale optimization algorithm (CMWOA), which combines chaotic and multi-swarm strategies is proposed, which significantly outperformed all the other competitors in terms of classification performance and feature subset size.

TL;DR: A cheap, fast, and reliable intelligence tool has been provided for COVID-19 infection detection, and the developed model can be used to assist field specialists, physicians, and radiologists in the decision-making process.

TL;DR: The use of ensembles is recommended to forecast agricultural commodities prices one month ahead, since a more assertive performance is observed, which allows to increase the accuracy of the constructed model and reduce decision-making risk.

TL;DR: Two different deep architectures for detecting the type of infection in tomato leaves are presented and the first architecture applies residual learning to learn significant features for classification and the second architecture applies attention mechanism on top of the residual deep network.

TL;DR: An experimental study is designed to compare five MCDM methods to validate the proposed approach with 10 feature selection methods, nine evaluation measures for binary classification, seven Evaluation measures for multi-class classification, and three classifiers with 10 small datasets, and the results demonstrate the effectiveness of the used M CDM-based method in evaluating feature selection method.

TL;DR: The experimental results show that the proposed CCGBFO significantly outperforms the original BFO in terms of both convergence speed and solution accuracy.

TL;DR: A recurrent neural network based on encoder–decoder framework with attention mechanism is proposed to predict HI values, which are designed closely related with the RUL values in this paper.

TL;DR: This paper addresses the dynamic flexible job shop scheduling problem (DFJSP) under new job insertions aiming at minimizing the total tardiness and confirms both the superiority and generality of DQN compared to each composite rule, other well-known dispatching rules as well as the stand Q-learning-based agent.

TL;DR: The application of the DBNLP algorithm model to collaborative robots can significantly improve its accuracy and safety, providing an experimental basis for the performance improvement of later collaborative robots.

TL;DR: This paper tried to provide a state-of-the-art snapshot of the developed DL models for financial applications, as of today, and categorized the works according to their intended subfield in finance but also analyzed them based on their DL models.

TL;DR: The proposed hHHO-SCA optimization algorithm is much better than standard sine–cosine optimization algorithm, Harris Hawks Optimizer, Ant Lion Optimizer algorithm, Moth Flame Optimization algorithm, grey wolf optimizer algorithm and others recently described meta-heuristics, heuristics and hybrid type optimization search algorithm and endorses its effectiveness in multi-disciplinary design and engineering optimization problems.

TL;DR: A deep neural network model using the desensitized transaction records and public market information to predict stock price trend is proposed and achieves the best performance in comparison with other prediction baselines.

TL;DR: A novel reinforcement learning based grey wolf optimizer algorithm called RLGWO has been presented for solving the problem of feasible and effective route acquisition in three-dimensional complex flight environment.

TL;DR: A method of combining (Convolutional neural network) CNN and ELM (extreme learning machine) improves the accuracy of ECG automatic classification and has good generalization ability.

TL;DR: This article employs a systematic literature survey approach to systematically review statistical and machine learning models in credit scoring, to identify limitations in literature, to propose a guiding machine learning framework, and to point to emerging directions.

TL;DR: It is concluded that the fuzzy neural network models and their derivations are efficient in constructing a system with a high degree of accuracy and an appropriate level of interpretability working in a wide range of areas of economics and science.

TL;DR: A multi- scale deep convolutional neural network (MS-DCNN) which have powerful feature extraction capability due to its multi-scale structure is proposed in this paper and achieves good prognostics performance compared with other network architectures and state-of-the-art methods.

TL;DR: The proposed method combines the strength of the fuzzy set in handling internal uncertainty and the advantages of the rough set in manipulating external uncertainty in a hybrid rough-fuzzy DEMATEL-TOPSIS approach to sustainable supplier selection for a smart supply chain.

TL;DR: An elite learning operator that is based on social comparison theory to improve the upper bound of the whole population’s quality and the accuracy and the convergence speed of the multimodal medical registration can be greatly enhanced.

TL;DR: A multi-objective optimization problem suite consisting of 16 bound-constrained real-world problems, which includes various problems in terms of the number of objectives, the shape of the Pareto front, and the type of design variables, is presented.

TL;DR: A novel q-rung orthopair fuzzy DTRS (q-ROFDTRS) model is established and some fundamental properties of the expected losses are explored and two methods to handle q-ROFNs and obtain 3WDs are proposed.

TL;DR: An end-to-end two-stream attention based LSTM network that can selectively focus on the effective features for the original input images and pay different levels of attentions to the outputs of each deep feature maps is proposed.

TL;DR: This paper implements the compact cuckoo search algorithm, and then, a new parallel communication strategy is proposed that can effectively save the memory of the unmanned robot and increase the accuracy and achieve faster convergence.

TL;DR: A contour detection based image processing algorithm based on Mamdani (Type-2) fuzzy rules for detection of blood vessels in retinal fundus images that offers an improved dynamics and flexibility in formulation of the linguistic threshold criteria.

TL;DR: A hybrid forecasting model that combines random forest, improved grey ideal value approximation, complementary ensemble empirical mode decomposition, and particle swarm optimization algorithm is constructed, which proved that the RF-CEEMD-DIFPSO-BPNN is a promising approach in terms of PV power generation forecasting.

TL;DR: This study presents a new multi-objective multi-echelon multi-product multi-period pharmaceutical supply chain network (PSCN) along with the production–distribution–purchasing–ordering–inventory holding-allocation-routing problem under uncertainty and develops a novel robust fuzzy programming method to cope with uncertainty parameters.

TL;DR: A way to combine interval type-2 fuzzy sets (IT2FSs) with evidential reasoning (ER) method, which is able to overcome some disadvantages of the conventional FMEA approach and deal with uncertainties more efficiently is presented.

TL;DR: Evaluation results indicate that the best and worst selected machine tool of the proposed method keeps high conformance with the actual ranking in real factory and that the presented hybrid model has advantages in granting flexibility to the preferences of decision makers.