Search

In recent years, gender detection has been an important problem in speech analysis, with many different applications. Different features of sound data such as pitch, median, and frequency are used for gender detection. In this study, a feature selection method based on metaheuristic optimization algorithms is proposed for gender detection from voice data. In the proposed method, the feature set representing the voice data in the most appropriate way is selected with optimization algorithms, and gender detection has been implemented with artificial intelligence algorithms using the obtained features. Nature-inspired metaheuristic optimization algorithms, which are capable of solving complex problems, are used to select features from voice data. Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Salp Swarm Algorithm (SSA), and Whale Optimization Algorithm (WOA) have been modeled for the first time for feature selection from voice data. A publicly accessible data set has been used to measure the efficiency of metaheuristic optimization algorithms. The performances of PSO, ACO, SSA, and WOA for feature selection have been compared in terms of three different criteria: fitness function value, accuracy value, and the number of selected features. After feature selection with metaheuristic optimization algorithms, Naive Bayes and Decision Tree algorithms have been applied to the new data sets obtained and the original data set. As a result of the analysis, it was observed that this method which uses metaheuristic optimization algorithms for feature selection increased the success rate in the results obtained with Naive Bayes and Decision Tree algorithms. [ABSTRACT FROM AUTHOR]

Classification rules mining is one of the most studied data mining problems. In this article, chemistry based Artificial Chemical Reaction Optimization Algorithm (ACROA) has been for the first time used for mining of generalized classification rules, which is a complex and no well researched generalized variant of classification rules mining where there is more than one goal attribute to be predicted. Furthermore, interestingness measure has been added by performing the adaptations to the algorithm in order to make the rules mined by the algorithm not only accurate and comprehensible but also interesting, surprising, and unexpected. Different rule sets within different databases satisfying different objectives have been flexibly mined by adapting the representation scheme and objective function. Performance of ACROA in classification rules mining within different real public databases have been compared that of genetic algorithm, particle swarm optimization algorithm, and ant colony optimization algorithm. It has shown that performance of ACROA within this special task of data mining is promising. ACROA can be an efficient solution method for different data mining tasks such as association rules mining, clustering rules mining, sequential pattern mining, and etc. [ABSTRACT FROM AUTHOR]

Optimization is the process of finding the best solution of a problem. There are many optimization algorithms proposed for optimization problems. The metaheuristic algorithm can give solutions close to optimum in an acceptable period of time for large-scaled optimization problems. The metaheuristic optimization algorithms are evaluated in seven different groups which are biology-based, physics-based, swarm-based, social-based, music-based, sport-based, and chemistry-based. The swarm-based optimization algorithms have been developed by observing the behaviors of creatures e.g. birds, fishes, cats, bees etc. Data mining is the process of discovery of meaningful and useful data within huge databases. Classification rules mining is one of the most commonly studied data mining problems and with methods for this problem, users can easily understand the rules extracted from databases. In this work, one of the most recent swarm based optimization algorithm, Cat Swarm Optimization (CSO), has been firstly used for classification rules mining within databases composed of numerical or mixed data types automatically. There is not any preprocess for finding true ranges for appropriate attributes of the rules, this has been automatically done by CSO. Furthermore, the used objective function is very flexible and many different objectives can easily be added to. For this purpose, four numerical databases obtained from UCI data warehouse have been used and accurate and comprehensible classification rules have been mined. The results have been compared with the results obtained from Weka program. Although CSO has not been embedded with any improvement and has firstly implemented in this research area, the obtained results seem promising. [ABSTRACT FROM AUTHOR]