Search

Analyzing and understanding the sentiments of social media documents on Twitter, Facebook, and Instagram has become a very important task at present. Analyzing the sentiment of these documents gives meaningful knowledge about the user opinions, which will help understand the overall view on these platforms. The problem of sentiment analysis (SA) can be regarded as a classification problem in which the text is classified as positive, negative, or neutral. This paper aims to give an intensive, but not exhaustive, review of the main concepts of SA and the state-of-the-art techniques; other aims are to make a comparative study of their performances, the main applications of SA as well as the limitations and the future directions for SA. Based on our analysis, researchers have utilized three main approaches for SA, namely lexicon/rules, machine learning (ML), and deep learning (DL). The performance of lexicon/rules-based models typically falls within the range of 55–85%. ML models, on the other hand, generally exhibit performance ranging from 55% to 90%, while DL models tend to achieve higher performance, ranging from 70% to 95%. These ranges are estimated and may be higher or lower depending on various factors, including the quality of the datasets, the chosen model architecture, the preprocessing techniques employed, as well as the quality and coverage of the lexicon utilized. Moreover, to further enhance models' performance, researchers have delved into the implementation of hybrid models and optimization techniques which have demonstrated an ability to enhance the overall performance of SA models. [ABSTRACT FROM AUTHOR]

The electroencephalogram (EEG) signal denoising problem has been considered a challenging task because of several artifact noises, such as eye blinking, eye movement, muscle activity, and power line interference, which can corrupt the original EEG signal during the recording time. Therefore, to remove these noises, the EEG signals must be processed to obtain efficient EEG features. Accordingly, several techniques have been proposed to reduce EEG noises, such as EEG signal denoising using wavelet transform (WT). The success of WT depends on the best configuration of its control parameters, which are often experimentally set. In this study, a multi-objective flower pollination algorithm (MOFPA) with WT (MOFPA-WT) is proposed to solve the EEG signal denoising problem. The novelty of this study is to find optimal EEG signal denoising parameters using MOFPA based on two measurement criteria for the denoised signals, namely minimum mean squared error (MSE) and maximum signal-to-noise ratio (SNR). The MOFPA-WT is tested using a standard EEG signal processing dataset, namely the EEG motor movement/imagery dataset. The performance of MOFPA-WT is evaluated using five criteria, namely SNR, SNR improvement, MSE, root mean squared error (RMSE), and percentage root mean square difference (PRD). Experiments are conducted using FPA with MSE, SNR, and MSE and SNR to show the effect of the multi-objective aspects on the performance of the proposed MOFPA-WT. Results show that FPA with MSE and SNR exhibits more subjective results than FPA with MSE and FPA with SNR. The convergence rate and Pareto front are also studied for the proposed MOFPA-WT. [ABSTRACT FROM AUTHOR]

To improve the optimization efficiency for different optimization problems and take advantage of the dynamic membrane computing framework, this paper proposes an improved bat algorithm, namely, Dynamic Membrane-driven Bat Algorithm (DMBA). The dynamic construction of the DMBA algorithm aims at enhancing population diversity by balancing the exploration-exploitation tradeoff. Unlike the static membrane algorithms, the membranes in DMBA will be dynamically evolved by using merging and separation rules which help in maintaining the diversity of the population. The experimental results on a set of well-known benchmark functions including CEC 2005, CEC 2011, and CEC 2017 clearly prove the effectiveness of the proposed DMBA algorithm in terms of maintaining the diversity and exploitation capabilities compared to that of the others. It is shown that the proposed DMBA algorithm is superior to recent variants of the bat algorithm and other well-known algorithms in terms of solution accuracy and convergence speed. [ABSTRACT FROM AUTHOR]

For text document clustering (TDC), a novel hybrid of the multi-verse optimizer (MVO) algorithm and k-means (also called H-MVO) are proposed in this work. Moreover, a new ensemble method for an automatic topic extraction (TE) has been proposed in this paper, from a set of scientific publications in the form of text documents with the purpose of extracting topics from clustered documents. Often, the existing TE methods draw upon the statistical theory. However, the results might be different when the same clustered document is utilized. Consequently, there can be imprecise results, which are related to the extracted topics from the clustered documents owing to the behavior of the TE methods. As a result, the vigorous characteristics of the TE methods are ensembled, thereby empowering the accuracy of the extracted topics. The results, which were yielded by H-MVO for TDC, were compared against 14 well-regarded methods, involving five clustering methods, in addition to seven metaheuristic algorithms, as well as two hybrid optimization algorithms. Also, the results, which were generated by the introduced ensembled TE method, were compared against those, which were produced by five established statistical methods in the literature. As a result, the findings revealed that the suggested ensembled TE method outperformed the entire comparative methods, thereby utilizing all the external measurements for almost the entire datasets. Moreover, the new method can complement the advantages of the five previously proposed methods. Accordingly, more advanced results were obtained. [ABSTRACT FROM AUTHOR]

The automatic topic extraction (TE) from scientific publications provides a very compact summary of the clusters' contents. This often helps in locating information easily. TE enables us to define the boundaries of the scientific fields. Text Document Clustering (TDC) represents, in general, the first step of topic identification to identify the documents, which address a related subject matter. Metaheuristics are typically used as efficient approaches for TDC. The multi-verse optimizer algorithm (MVO) involves a stochastic population-based algorithm. It has been recently proposed and successfully utilized to tackle many hard optimization problems. In the TE process, the focus of each statistical TE method is placed on various language feature space aspects. The aim of this paper is to design a novel ensemble method for an automatic TE from a collection of scientific publications based on MVO as the clustering algorithm. The automatic TE, which is used in our approach, is term frequency-inverse document frequency (TF-IDF), most frequent based keyword extraction (TF), co-occurrence statistical information-based keyword extraction (CSI), TextRank (TR), and mutual information (MI). A group of candidate topics can be provided by each automatic TE method for the proposed ensemble method. Next, the ensemble approach prunes the candidate topics' set via the application of a specific filtering heuristic. Then, their scores are recalculated based on the prescribed metrics. After that, for selecting a set of topics for certain scientific publications, dynamic threshold functions are applied. The findings emphasized the refined candidate set's efficiency, as well as effectiveness. The results also showed that the system's quality has been improved by new topics. The proposed method achieved better precision, as well as recall on a similar dataset compared to the state-of-the-art TE methods. [ABSTRACT FROM AUTHOR]

Text clustering has been widely utilized with the aim of partitioning specific document collection into different subsets using homogeneity/heterogeneity criteria. It has also become a very complicated area of research, including pattern recognition, information retrieval, and text mining. Metaheuristics are typically used as efficient approaches for the text clustering problem. The multi-verse optimizer algorithm (MVO) involves a stochastic population-based algorithm. It has been recently proposed and successfully utilized to tackle many hard optimization problems. However, a recently applied research trend involves hybridizing two or more algorithms with the aim of obtaining a superior solution regarding the problems of optimization. In this paper, a new hybrid of MVO algorithm with the K-means clustering algorithm is proposed, i.e., the H-MVO algorithm with the aims of enhancing the quality of initial candidate solutions, as well as the best solution, which is produced by MVO at each iteration. This hybrid algorithm aims at improving the global (diversification) ability of the search and finding a better cluster partition. The proposed H-MVO effectiveness was tested on five standard datasets, which are used in the domain of data clustering, as well as six standard text datasets, which are utilized in the domain of text document clustering, in addition to two scientific articles' datasets. The experiments showed that K-means hybridized MVO improves the results in terms of high convergence rate, accuracy, error rate, purity, entropy, recall, precision, and F-measure criteria. In general, H-MVO has outperformed or at least proven to be highly competitive compared to the original MVO algorithm and with well-known optimization algorithms like KHA, HS, PSO, GA, H-PSO, and H-GA and the clustering techniques like K-mean, K-mean++, DBSCAN, agglomerative, and spectral clustering techniques. [ABSTRACT FROM AUTHOR]

In this paper, the multi-objective grey wolf optimiser is utilised for the power scheduling problem (PSP). The grey wolf optimiser (GWO) is a recent swarm-based optimisation algorithm tailored for various optimisation problems. PSP is addressed by scheduling home appliances to a certain time horizon to minimise the electricity bill and peak-to-average ratio (PAR) and increase the comfort level of users. The multi-objective function is formalised and utilised in GWO to obtain an optimal schedule. Seven consumption profiles and seven real-time electricity prices with various characteristics are considered to evaluate the proposed multi-objective GWO. The performance of the proposed algorithm is tested against three factors, namely, electricity bill, PAR and user comfort level. The obtained schedule shows that all evaluation factors are optimally timetabled. For a comparative evaluation, the proposed method is firstly compared with the genetic algorithm. The proposed method exhibits and yield better performance than GA under the same consumption profiles. Secondly, the proposed method is compared with 19 state-of-the-art methods by using the recommended consumption profiles of these methods and their evaluation criteria. The proposed method nearly outperforms the compared methods in terms of minimisation of electricity bill and PAR. User comfort level is a criterion proposed in this study and has not been considered previously. It exerts a significant impact on the final schedule. [ABSTRACT FROM AUTHOR]

The cuckoo search algorithm (CSA) is a promising metaheuristic algorithm for solving numerous problems in different fields. It adopts the Levy flight to guide the search process. Nonetheless, CSA has drawbacks, such as the utilization of global search; in certain cases, this technique may surround local optima. Moreover, the results cannot be guaranteed if the step size is considerably large, thereby leading to a slow convergence rate. In this study, we introduce a new method for improving the search capability of CSA by combining it with the bat algorithm (BA) to solve numerical optimization problems. The proposed algorithm, called CSBA, begins by establishing the population of host nests in standard CSA and then obtains a solution through particular part to identify a new solution in BA (i.e., further exploitation). Therefore, CSBA overcomes the slow convergence of the standard CSA and avoids being trapped in local optima. The performance of CSBA is validated by applying it on a set of benchmark functions that are divided into unimodal and multimodal functions. Results indicate that CSBA performs better than the standard CSA and existing methods in the literature, particularly in terms of local search functions. [ABSTRACT FROM AUTHOR]

This paper proposed a new gene selection method based on modified Minimum Redundancy Maximum Relevancy (MRMR) as a filtering approach and hybrid bat algorithm with β-hill climbing as an efficient wrapper approach. The gene selection is a process of selecting the discriminative genes that aid in the development of efficient cancer diagnosis and classification. In general, the current filter-based approaches produced gene subset according to its discriminative power. However, one of the deficiencies of single filter approaches is that it has high variability of the classification results. Accordingly, this study aim to improve MRMR through incorporating its with ensemble of filters to increase the robustness and the stability of MRMR. The result of filtering-based approach is a set of discriminative genes. The wrapper-based approach considers the results from the filtering-based approach to formulate the gene selection search space. In wrapper approach, bat algorithm is tailored for gene selection problem and hybridized with a powerful local search method called beta hill climbing to further stress the deep learning side in the search space navigation and thus find a very robust and stable discriminative genes. Bat-inspired algorithm (BA) is a recent swarm-based optimization method while β-hill climbing is an exploratory local search. The proposed method is called Robust MRMR and Hybrid Bat-inspired Algorithm (rMRMR-HBA). To evaluate the proposed method, ten well-known microarray datasets are experimented with. These datasets are varies in terms of number of genes, samples, and classes. For performance evaluation, the proposed filtering-based approach (i.e., rMRMR) is initially tested against the standard MRMR and other well-regard filtering approaches. Thereafter, the wrapper-based approach (i.e., HBA) is evaluated by studying the convergence behavior of BA with and without β-hill climbing. For comparative evaluation, the results of the proposed rMRMR-HBA were compared with state-of-art methods using the same microarray datasets. The comparative results show that our proposed approach achieved outstanding results in two out of ten datasets in terms of clarification accuracy and minimum number of genes. [ABSTRACT FROM AUTHOR]

In this paper, a novel text clustering method, improved krill herd algorithm with a hybrid function, called MMKHA, is proposed as an efficient clustering way to obtain promising and precise results in this domain. Krill herd is a new swarm-based optimization algorithm that imitates the behavior of a group of live krill. The potential of this algorithm is high because it performs better than other optimization methods; it balances the process of exploration and exploitation by complementing the strength of local nearby searching and global wide-range searching. Text clustering is the process of grouping significant amounts of text documents into coherent clusters in which documents in the same cluster are relevant. For the purpose of the experiments, six versions are thoroughly investigated to determine the best version for solving the text clustering. Eight benchmark text datasets are used for the evaluation process available at the Laboratory of Computational Intelligence (LABIC). Seven evaluation measures are utilized to validate the proposed algorithms, namely, ASDC, accuracy, precision, recall, F-measure, purity, and entropy. The proposed algorithms are compared with the other successful algorithms published in the literature. The results proved that the proposed improved krill herd algorithm with hybrid function achieved almost all the best results for all datasets in comparison with the other comparative algorithms. [ABSTRACT FROM AUTHOR]

Cuckoo search algorithm is considered one of the promising metaheuristic algorithms applied to solve numerous problems in different fields. However, it undergoes the premature convergence problem for high dimensional problems because the algorithm converges rapidly. Therefore, we proposed a robust approach to solve this issue by hybridizing optimization algorithm, which is a combination of Cuckoo search algorithmand Hill climbing called CSAHC discovers many local optimum traps by using local and global searches, although the local search method is trapped at the local minimum point. In other words, CSAHC has the ability to balance between the global exploration of the CSA and the deep exploitation of the HC method. The validation of the performance is determined by applying 13 benchmarks. The results of experimental simulations prove the improvement in the efficiency and the effect of the cooperation strategy and the promising of CSAHC. [ABSTRACT FROM AUTHOR]

In this paper, the update process of harmony search (HS) algorithm is modified to improve its concept of diversity. The update process in HS is based on a greedy mechanism in which the new harmony solution, created in each generation, replaces the worst individual in the population, if better. This greedy process could be improved with other updates mechanisms in order to control the diversity perfectly. Three versions of HS have been proposed: (1) Natural Proportional HS ; (2) Natural Tournament HS; (3) Natural Rank HS. These three HS versions employed the natural selection principle of the “survival of the fittest”. Instead of replacing the worst individual in population, any individual can be replaced based on certain criteria. Four versions of economic loading dispatch (ELD) problems with valve point have been used to measure the effect of the newly proposed HS versions. The results show that the new HS versions are very promising for ELD domain. This claim is proved based on the comparative evaluation process where the new HS versions are able to excel the state-of-the-art methods in almost ELD problems used. [ABSTRACT FROM AUTHOR]

Krill herd (KH) is a stochastic nature-inspired optimization algorithm, it has been successfully used to solve many involved optimization problems. Occasionally, poor exploration (diversification) capability affects the performance of krill herd algorithm (KHA). In this paper, we proposed a new hybridization strategy, namely, hybrid the krill herd algorithm with the harmony search (HS) algorithm (Harmony-KHA), to improve the data clustering technique. This hybridization strategy seeks to enhance the global (diversification) search capability of the KH algorithm to obtain the best global optima. The proposed algorithm are conducted through the addition of the global search operator from the HS algorithm in order to improve the exploration around the optimal solution in KH and thus kill individuals move towards the global best solution. The proposed algorithm is applied to keep the best krill individuals during the updating positions of the krill individuals. Experiments were conducted using four standard datasets from the UCI Machine Learning Repository, which is commonly used in the domain of data clustering. The results showed that the proposed hybrid the KH algorithm with the HS algorithm (Harmony-KHA) is produced very accurate clusters especially in the large dataset. Furthermore, the Harmony-KHA obtained a high convergence rate and it can overcome the other comparative algorithms. The proposed algorithm is compared with other well-known based on data clustering algorithms including the original KH algorithm. [ABSTRACT FROM AUTHOR]

The job shop scheduling problem (JSSP) is regarded as one of the problems which is widely studied. This paper presents a bee colony optimization (BCO) algorithm with a local search named the modified two-enhancement scheme with neighbourhood N5 perturbation (BCO+mTESN5) to solve the JSSP. In previous research, the BCO algorithm with TESN5 local search (BCO+TESN5) is applied to solve the JSSP. The BCO+TESN5 algorithm uses a brute force strategy in performing the N5 neighbourhood local search. However, the brute force local search strategy incurs expensive overhead. To address the high overhead issue, the BCO+mTESN5 algorithm is proposed where the local search is performed on a targeted bottleneck machine within the N5 neighbourhood structure. The selection of the targeted bottleneck machine is done based on a list of bottleneck machines identified by the shifting bottleneck heuristic (SBH). Two selection strategies are tested to select the targeted bottleneck machine, namely: the greedy selection and the linear ranking selection. The proposed algorithms are examined using a set of benchmark problems obtained from the OR-library. The results show that the proposed BCO+mTESN5 with linear ranking selection successfully solves 54% of the 82 OR-library benchmark dataset to ⩽ 1% from known optimum and it is comparable to the BCO+TESN5 algorithm. In terms of computational time to obtain the best makespan, the proposed BCO+mTESN5 with linear ranking selection outperforms the BCO+TESN5 algorithm by 25%. [ABSTRACT FROM AUTHOR]

A pedestrian evacuation under fire-spreading conditions is simulated by using a two-dimensional cellular automaton model. The proposed model presents a non-static fire-spreading behavior to avoid considerable discrepancies between reality and simulation. The proposed model adopts a circular fire front shape based on spiral fire movement. Moreover, four dynamic parameters are introduced to simplify the decision-making process of a pedestrian's movement inside the layout during fire spreading. In addition, the proposed model includes the number of victims (i.e., caught in the fire) and the number of pedestrians who were evacuated safely. By analyzing these variables, a suitable evacuation plan enabling the control of crowd movements in different situations such as fire disasters can be consequently designed. [ABSTRACT FROM AUTHOR]

Crowd and evacuation management have been active areas of research and study in the recent past. Various developments continue to take place in the process of efficient evacuation of crowds in mass gatherings. This article is intended to provide a review of intelligent evacuation management systems covering the aspects of crowd monitoring, crowd disaster prediction, evacuation modelling, and evacuation path guidelines. Soft computing approaches play a vital role in the design and deployment of intelligent evacuation applications pertaining to crowd control management. While the review deals with video and nonvideo based aspects of crowd monitoring and crowd disaster prediction, evacuation techniques are reviewed via the theme of soft computing, along with a brief review on the evacuation navigation path. We believe that this review will assist researchers in developing reliable automated evacuation systems that will help in ensuring the safety of the evacuees especially during emergency evacuation scenarios. [ABSTRACT FROM AUTHOR]

It is known that the Common Algorithmic Problem (CAP) has the nice property that several other NP-complete problems can be reduced to it in linear time. The decision version of this problem is known to be efficiently solved by a family of recognizer P systems with active membranes with three electrical charges working in the maximally parallel way. We here work with a variant of P systems with active membranes without polarizations and present a uniform solution to CAP in the minimally parallel mode. [ABSTRACT FROM AUTHOR]

The selection methods of population-based metaheuristics provide the driving force to generate good solutions. These selection methods select the individuals with a higher fitness to be members of the population in the next iteration correspond to the natural rule of Darwin's principle survival-of-the-fittest. Harmony search algorithm is a population-based metaheuristic, which mimicking the musical improvisation process where a group of musicians play the pitches of their musical instruments seeking for a pleasing harmony. It improvises the new harmony based on three rules: memory consideration, random consideration, and pitch adjustment. In this paper, we investigate the replacement of the original random selection of memory consideration with a set of selection methods in order to speed-up the convergence. These selection methods include tournament, proportional, and liner rank of Genetic Algorithm, and Global-best of Particle Swarm Optimization. The proposed harmony search with the different memory consideration selection methods evaluated using standard dataset published in the first International Nurse Rostering Competition INRC2010. Nurse rostering problem is a combinatorial optimization problem tackled by assigning a set of nurses with different skills to a set of shifts over predefined scheduling period. Experimentally, the tournament memory consideration selection method achieved the best rate of convergence as well as the best results in comparison with the other memory consideration selection methods. [ABSTRACT FROM AUTHOR]

Structured population in evolutionary algorithms (EAs) is an important research track where an individual only interacts with its neighboring individuals in the breeding step. The main rationale behind this is to provide a high level of diversity to overcome the genetic drift. Cellular automata concepts have been embedded to the process of EA in order to provide a decentralized method in order to preserve the population structure. Harmony search (HS) is a recent EA that considers the whole individuals in the breeding step. In this paper, the cellular automata concepts are embedded into the HS algorithm to come up with a new version called cellular harmony search (cHS). In cHS, the population is arranged as a two-dimensional toroidal grid, where each individual in the grid is a cell and only interacts with its neighbors. The memory consideration and population update are modified according to cellular EA theory. The experimental results using benchmark functions show that embedding the cellular automata concepts with HS processes directly affects the performance. Finally, a parameter sensitivity analysis of the cHS variation is analyzed and a comparative evaluation shows the success of cHS. [ABSTRACT FROM AUTHOR]