Model-Based Method for Optimisation of an Adaptive System

Bachir, Magagi Ali; Ismail, Jelloulil; Said, El Garouani; Amjad, Souad

doi:10.4018/ijssci.301269

Cited by 4 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The weakness of meta‐heuristic wrapper techniques resulted in the idea of combining different meta‐heuristic methods to take the advantage of the strength side of each method. Many research proposed many different methods 39–41 . Sarhani et al 42 is an example of hybrid method where gravitational search algorithm is added to particle swarm optimization for the objective of feature selection.…”

Section: Related Workmentioning

confidence: 99%

“…Many research proposed many different methods. [39][40][41] Sarhani et al 42 is an example of hybrid method where gravitational search algorithm is added to particle swarm optimization for the objective of feature selection. In this study a multiobjective fitness function is defined where the error rate of KNN classification and the number of selected features are considered as the objectives.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multiobjective whale optimization algorithm‐based feature selection for intelligent systems

Riyahi

Rafsanjani

Gupta

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

With regard to large dimensions of contemporary data sets and restricted computational time of intelligent systems, reducing the dimensions of data sets is necessary. Feature selection is a practical way to remove a set of redundant, irrelevant, and noisy features. In this way, the speed of decision-making procedure will be increased while the accuracy of decisions will be retained. To this end, numerous attentions have been attracted to the topic and consequently, extensive range of methods has been proposed. Regarding the goals of the feature selection concept, the proposed algorithms in this field must be fast and accurate. Therefore, this paper proposes a light meanwhile accurate algorithm to fulfill the mentioned goals. The presented algorithm takes the speed advantage of Whale Optimization Algorithm (WOA) to propose a novel feature selection method for intelligent systems. Moreover, to reach the goal of accuracy, the proposed strategy considers three important fitness objectives, namely, the number of selected features, the accuracy of classification, and information gain. The proposed scheme considers an accurate multiobjective fitness function instead of manipulating the basic algorithm. The reason is that improving the basic algorithms, WOA in our case, may lead to loading

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Multiobjective whale optimization algorithm‐based feature selection for intelligent systems

Riyahi

Rafsanjani

Gupta

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

show abstract

“…Due to existing challenging problems, like multi‐label classifications, power management, 4 the human ability is not enough to solve them, so it shows the significant capability of intelligent systems and collaborative systems in high challenging issues 5–9 . In addition, the devices are become more complex 10 and the data volume is increased exponentially, 11–13 especially in text format, and their dimensions also have grown, which are super‐high‐dimensional textual data, that is, each instance or some instances has many features and needs adapative system 14 . Text data sets contain many terms, which simultaneously put documents into various classes, in which multi‐label text classification is utilized to recognize the various classes of documents 15–17 .…”

Section: Introductionmentioning

confidence: 99%

“…[5][6][7][8][9] In addition, the devices are become more complex 10 and the data volume is increased exponentially, [11][12][13] especially in text format, and their dimensions also have grown, which are super-high-dimensional textual data, that is, each instance or some instances has many features and needs adapative system. 14 Text data sets contain many terms, which simultaneously put documents into various classes, in which multi-label text classification is utilized to recognize the various classes of documents. [15][16][17] Generally, irrelevant and redundant features in text data affect negatively or do not affect data analysis output.…”

Section: Introductionmentioning

confidence: 99%

Ensemble feature selection for multi‐label text classification: An intelligent order statistics approach

Miri

Dowlatshahi

Hashemi

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Because of the overgrowth of data, especially in text format, the value and importance of multi‐label text classification have increased. Aside from this, preprocessing and particularly intelligent feature selection (FS) are the most important step in classification. Each FS finds the best features based on its approach, but we try to use a multi‐strategy approach to find more useful features. Evaluating and comparing features’ importance and relevance makes using multiple strategy and methods more suitable than conventional approaches because each feature is measured based on several perspectives. Nevertheless, the ensemble FS merges the final performance results of various methods to take advantage of different methods’ strengths and better classify. In this article, we have proposed an ensemble FS method for multi‐label text data (MLTD) for the first time using the order statistics (EMFS) approach. We have utilized four multi‐label FS (MLFS) algorithms with various particular performances to achieve a good result. In this method, as one of the most important statistics methods, Order Statistics was used to aggregate the ranks of different algorithms, which is robust against noise, redundant and inessential features. In the end, the performance of EMFS, executing six MLTDs, was evaluated according to six performance criteria (ranking‐based and classification‐based). Surprisingly, the proposed method was more accurate than others among all used MLTDs. The proposed method has improved by 1.5% compared to other methods. This value is based on the results obtained based on six evaluation criteria and all tested data sets.

show abstract

“…Additionally, it can address the issue of some knowledge graphs lacking entity context data. These neural-network-based entity linking models [17] [18] [19] can use the surrounding context information to address the linking issue. The global consistency information is also introduced together with the local model for better disambiguation [20][21] [22][23] [24][25] [26][27] [28] These models improve the performance of term representation and linking.…”

Section: Introductionmentioning

confidence: 99%