Search

Information retrieval is a difficult process due to the overabundance of information on the web. Nowadays, search result responds to user queries with too many results although only a few are relevant. Therefore, the existing clustering methods that fail in clustering snippets (short texts) of web documents due to the low frequencies of document terms should be deeply investigated. One of the approaches that can be used to solve this problem is the expansion of document terms with semantically similar terms. Hence, a list of terms with their closest and accurate semantically similar words (word representation) must be built. This study aims to design and develop a new framework to enhance the performance of web search result clustering (WSRC). The research also presents a new unsupervised distributed word representation scheme where each word is represented by a vector of its semantically related words; such as scheme expands snippets and user queries. The proposed framework consists of several activities, such as (1) various standard datasets (Open Directory Project [ODP]-239 and MORESQUE) that are used for evaluating search result clustering algorithms for most cited dataset works, (2) text pre-processing, (3) document representation based on a new wiki-based k-nearest neighbors (KNN) representation method, (4) effect of the proposed model on the performance of traditional clustering methods (k-means, k-medoids, single-linkage, and complete-linkage) for WSRC, and (5) evaluation stage of the proposed method. Results indicate that enhanced clustering methods, according to the new wiki-KNN based representation method in comparison with the baseline methods, show a significant improvement in WSRC. Furthermore, the new data representation scheme has enhanced the overall performance of clustering methods.

The bird mating optimizer is a new metaheuristic algorithm that was originally proposed to solve continuous optimization problems with a very promising performance. However, the algorithm has not yet been applied for solving combinatorial optimization problems. Thus, the formulation may not be able to generate a discrete feasible solution. Many continuous algorithms used random-key representation to represent the discrete solution using real numbers or a discrete variant of the algorithm is used to deal with the discrete solution of the problem. However, there is no evidence which methodology is better for solving combinatorial optimization problems. Therefore, this work proposes two variants of bird mating optimizer (random-key bird mating optimizer and the discrete bird mating optimizer), to identify which one is more efficient in solving combinatorial optimization problems. In the first one, we use a random-key encoding scheme, whilst, in the later one, we use crossover (multi-parent) and mutation operators to combine the components of the selected parents to generate new broods. The performance of these algorithms is tested on the travelling salesman problem and berth allocation problem, and are compared with the results of two well-known optimization algorithms: Genetic Algorithm and Particle Swarm Optimization. Experimental results show that the discrete bird mating optimizer is more efficient than the others on all tested benchmark instances. Indeed, it is able to attain the best-known results in some of the BAP benchmark instances. These indicate the applicability and the effectiveness of the proposed discrete bird mating optimizer in solving combinatorial optimization problems.

Many multi-parent crossovers have been proposed to solve specific combinatorial optimization problem and not applicable to solve other problems (i.e. cannot produce feasible solution). Only multi-parent partially mapped crossover (MPPMX) and adjacency-based crossover (ABC) have been proposed to work over different combinatorial problems. However, both MPPMX and ABC suffered from a very high computational time or poor performance. Therefore, this work proposes a novel multi-parent order crossover (MPOX) for solving several combinatorial optimization problems with reasonable amount of time. The MPOX extends the two-parent order crossover by modifying the recombination operator to recombine more than two parents and generates a new offspring. MPOX at first selects the crossover points and divides the parents into n substrings based on these points (where n is the number of parents). Then, MPOX copies a predefined number of elements from each parent into the offspring based on their order while checking the feasibility of the offspring. The performance of MPOX is tested on the traveling salesman problems and berth allocation problems, which are widely studied in the literature. Experimental results demonstrated that the MPOX significantly improves the OX in both problem domains and outperforms both ABC and MPPMX over the travelling salesman problem and the berth allocation problem with less computational time. These results indicate the effectiveness of MPOX over OX, ABC and MPPMX, and its capability for solving both problems.

The diversity and quality of high-quality and diverse-solution external memory of the hybrid Elitist-Ant System is examined in this study. The Elitist-Ant System incorporates an external memory for preserving search diversity while exploiting the solution space. Using this procedure, the effectiveness and efficiency of the search may be guaranteed which could consequently improve the performance of the algorithm and it could be well generalized across diverse problems of combinatorial optimization. The generality of this algorithm through its consistency and efficiency is tested using a Nurse-Rostering Problem. The outcomes demonstrate the competitiveness of the hybrid Elitist-Ant System’s performance within numerous datasets as opposed to those by other systems. The effectiveness of the external memory usage in search diversification is evidenced in this work. Subsequently, such usage improves the performance of the hybrid Elitist-Ant System over diverse datasets and problems. Keywords: Metaheuristics, Elitist-Ant System, External memory, Diversification, Intensification, Nurse Rostering Problem

Having the best solution for Vehicle Routing Problem (VRP) is still in demand. Beside, Cuckoo Search (CS) is a popular metaheuristic based on the reproductive strategy of the Cuckoo species and has been successfully applied in various optimizations, including Capacitated Vehicle Routing Problem (CVRP). Although CS and hybrid CS have been proposed for CVRP, the performance of CS is far from the state-of-art. Therefore, this study proposes a hybrid CS with Simulated Annealing (SA) algorithm for the CVRP, consisting of three improvements&mdash, the investigation of 12 neighborhood structures, three selections strategy and hybrid it with SA. The experiment was conducted using 16 instances of the Augerat benchmark dataset. The results show that 6 out of 12 neighborhood structures were the best and the disruptive selection strategy is the best strategy. The experiments&rsquo, results showed that the proposed method could find optimal and near-optimal solutions compared with state-of-the-art algorithms.

The metaheuristic genetic algorithm (GA) is based on the natural selection process that falls under the umbrella category of evolutionary algorithms (EA). Genetic algorithms are typically utilized for generating high-quality solutions for search and optimization problems by depending on bio-oriented operators such as selection, crossover, and mutation. However, the GA still suffers from some downsides and needs to be improved so as to attain greater control of exploitation and exploration concerning creating a new population and randomness involvement happening in the population at the solution initialization. Furthermore, the mutation is imposed upon the new chromosomes and hence prevents the achievement of an optimal solution. Therefore, this study presents a new GA that is centered on the natural selection theory and it aims to improve the control of exploitation and exploration. The proposed algorithm is called genetic algorithm based on natural selection theory (GABONST). Two assessments of the GABONST are carried out via (i) application of fifteen renowned benchmark test functions and the comparison of the results with the conventional GA, enhanced ameliorated teaching learning-based optimization (EATLBO), Bat and Bee algorithms. (ii) Apply the GABONST in language identification (LID) through integrating the GABONST with extreme learning machine (ELM) and named (GABONST-ELM). The ELM is considered as one of the most useful learning models for carrying out classifications and regression analysis. The generation of results is carried out grounded upon the LID dataset, which is derived from eight separate languages. The GABONST algorithm has the capability of producing good quality solutions and it also has better control of the exploitation and exploration as compared to the conventional GA, EATLBO, Bat, and Bee algorithms in terms of the statistical assessment. Additionally, the obtained results indicate that (GABONST-ELM)-LID has an effective performance with accuracy reaching up to 99.38%.

Multi-label text categorization refers to the problem of assigning each document to a subset of categories by means of multi-label learning algorithms. Unlike English and most other languages, the unavailability of Arabic benchmark datasets prevents evaluating multi-label learning algorithms for Arabic text categorization. As a result, only a few recent studies have dealt with multi-label Arabic text categorization on non-benchmark and inaccessible datasets. Therefore, this work aims to promote multi-label Arabic text categorization through (a) introducing “RTAnews”, a new benchmark dataset of multi-label Arabic news articles for text categorization and other supervised learning tasks. The benchmark is publicly available in several formats compatible with the existing multi-label learning tools, such as MEKA and Mulan. (b) Conducting an extensive comparison of most of the well-known multi-label learning algorithms for Arabic text categorization in order to have baseline results and show the effectiveness of these algorithms for Arabic text categorization on RTAnews. The evaluation involves four multi-label transformation-based algorithms: Binary Relevance, Classifier Chains, Calibrated Ranking by Pairwise Comparison and Label Powerset, with three base learners (Support Vector Machine, k-Nearest-Neighbors and Random Forest); and four adaptation-based algorithms (Multi-label kNN, Instance-Based Learning by Logistic Regression Multi-label, Binary Relevance kNN and RFBoost). The reported baseline results show that both RFBoost and Label Powerset with Support Vector Machine as base learner outperformed other compared algorithms. Results also demonstrated that adaptation-based algorithms are faster than transformation-based algorithms.

The steganography is prone to number of attacks such as geomatical, salt & pepper, gaussian, median filtering, attacks. To overcome these problems, the cryptography and error correction codes are comes in the pictures and hybrid with steganography algorithms. The cryptography algorithms add one layer of security on steganography algorithms and error correction codes improves the robustness of steganography algorithms. On the other side, the hybridization of the algorithms, increase memory complexity and increase computation time for data embedding. Thus, in this paper a review on multi-layer algorithms for video steganography is done. This paper comprehensively reviews the steganography, spatial and frequency domain techniques hybrid with cryptography and error correction techniques. In-depth analysis on published literatures revealed that the spatial domain techniques require shorter computation time for data embedding and provide higher data hiding capacity than frequency domain methods. Performance of such techniques was evaluated in terms of visual quality and robustness parameters. Present challenges and future trends towards the improvement of the multi-layer data security in video steganography are discussed to provide the taxonomy for further navigation.

This paper delves into the capacity of enhanced Big Bang-Big Crunch (EBB-BC) metaheuristic to handle data clustering problems. BB-BC is a product of an evolution theory of the universe in physics and astronomy. Two main phases of BB-BC are big bang and big crunch. The big bang phase involves a creation of a population of random initial solutions, while in the big crunch phase these solutions are shrunk into one elite solution exhibited by a mass center. This study looks into enhancing the BB-BC’s effectiveness in clustering data. Where, the inclusion of an elite pool alongside implicit solution recombination and local search method, contribute to such enhancement. Such strategies resulted in a balanced search of good quality population that is also diverse. The proposed elite pool-based BB-BC was compared with the original BB-BC and other identical metaheuristics. Fourteen different clustering datasets were used to test BB-BC and the elite pool-based BB-BC showed better performance compared to the original BB-BC. BB-BC was impacted more by the incorporated strategies. The experiments outcomes demonstrate the high quality solutions generated by elite pool-based BB-BC. Its performance in fact supersedes that of identical metaheuristics such as swarm intelligence and evolutionary algorithms.

Boosting algorithms have been proved effective for multi-label learning. As ensemble learning algorithms, boosting algorithms build classifiers by composing a set of weak hypotheses. The high computational cost of boosting algorithms in learning from large volumes of data such as text categorization datasets is a real challenge. Most boosting algorithms, such as AdaBoost.MH, iteratively examine all training features to generate the weak hypotheses, which increases the learning time. RFBoost was introduced to manage this problem based on a rank-and-filter strategy in which it first ranks the training features and then, in each learning iteration, filters and uses only a subset of the highest-ranked features to construct the weak hypotheses. This step ensures accelerated learning time for RFBoost compared to AdaBoost.MH, as the weak hypotheses produced in each iteration are reduced to a very small number. As feature ranking is the core idea of RFBoost, this paper presents and investigates seven feature ranking methods (information gain, chi-square, GSS-coefficient, mutual information, odds ratio, F1 score, and accuracy) in order to improve RFBoost's performance. Moreover, an accelerated version of RFBoost, called RFBoost1, is also introduced. Rather than filtering a subset of the highest-ranked features, FBoost1 selects only one feature, based on its weight, to build a new weak hypothesis. Experimental results on four benchmark datasets for multi-label text categorization) Reuters-21578, 20-Newsgroups, OHSUMED, and TMC2007(demonstrate that among the methods evaluated for feature ranking, mutual information yields the best performance for RFBoost. In addition, the results prove that RFBoost statistically outperforms both RFBoost1 and AdaBoost.MH on all datasets. Finally, RFBoost1 proved more efficient than AdaBoost.MH, making it a better alternative for addressing classification problems in real-life applications and expert systems.

The harmony search algorithm has been proven to be an effective optimization method for solving diverse optimization problems. However, due to its slow convergence, the performance of HSA over constrained optimization problems is not very competitive. Therefore, many researchers have hybridized HSA with local search algorithms. However, it’s very difficult to known in advance which local search should be hybridized with HSA as it depends heavily on the problem characteristics. The question is how to design an effective selection mechanism to adaptively select a suitable local search to be combined with HSA during the search process. Therefore, this work proposes an adaptive HSA that embeds an adaptive selection mechanism to adaptively select a suitable local search algorithm to be applied. This work hybridizes HSA with five local search algorithms: hill climbing, simulated annealing, record to record, reactive tabu search and great deluge. We use the Solomon’s vehicle routing problem with time windows benchmark to examine the effectiveness of the proposed algorithm. The obtained results are compared with basic HSA, the local search algorithms and existing methods. The results demonstrate that the proposed adaptive HSA achieves very good results compared other methods. This demonstrates that the selection mechanism can effectively assist HSA to adaptively select a suitable local search during the problem solving process.

Multi-label text classification deals with the issue that arises from each sample being related to multiple labels. The text data suffers from high dimensionality. In order to resolve this issue, a feature selection (FS) method can be implemented for efficiently removing the noisy, irrelevant, and redundant features. Multi-label FS is a powerful tool for solving the high-dimension problem. With regards to handling correlation and high dimensionality problems in multi-label text classification, this paper investigates the various heterogeneous FS ensemble schemes. In addition, this paper proposes an enhanced FS method called dynamic multi-label two-layers MI and clustering-based ensemble feature selection algorithm (DMMC-EFS). The proposed method considers the: 1) dynamic global weight of feature, 2) heterogeneous ensemble, and 3) maximum dependency and relevancy and minimum redundancy of features. This method aims to overcome the high dimensionality of multi-label datasets and acquire improved multi-label text classification. We have conducted experiments based on three benchmark datasets: Reuters-21578, Bibtex, and Enron. The experimental results show that DMMC-EFS has significantly outperformed other state-of-the-art conventional and ensemble multi-label FS methods.

Harmony search algorithm (HSA) is one of the relatively new metaheuristic algorithms that classified under population-based search algorithms. Based on literature, hybridizing local-based searching algorithms with population-based algorithms can improve the performance of hybridized algorithms. This research is an extension to our previous work that focus on solving Nurse Rostering Problems (NRP) using hybrid metaheuristic algorithms. One of the improved version of HSA is enhanced harmony search algorithm (EHSA) where it overcomes some of the weaknesses of basic HSA. Slow convergence is noticed in EHSA which encourage us to hybridize it with other metaheuristic algorithms to improve its performance. In this research, EHSA is hybridized with great deluge algorithm (GD) and called Deluged harmony search algorithm (DHSA). DHSA then compared to CHSA (the hybridization of EHSA with Hill climbing (HC)) which developed earlier. To strike the balance between exploration and exploitation, the exploration stage run using EHSA and the exploitation stage used GD. DHSA is tested to solve a real world NRP problem at National University Malaysia Medical Center (UKMMC). The results show that, DHSA performed much better than CHSA in all instances in terms of solution quality with slightly higher execution time.

This work investigates the effect of elite pool that has high-quality and diverse solutions in three hybrid population-based meta-heuristics with an elite pool of a hybrid Elitist-Ant System, a hybrid Big Bang-Big Crunch optimization, and a hybrid scatter search. The purpose of incorporating an elite pool in population-based meta-heuristics is to maintain the diversity of the search while exploiting the solution space as in the reference set of the scatter search. This may guarantee the effectiveness and efficiency of the search, which could enhance the performance of the algorithms and generalized well across different datasets. To test the generality of these meta-heuristics via their consistency and efficiency, we use three classes of well-known combinatorial optimization problems as follows: symmetric traveling salesman problem, 0-1 multidimensional knapsack problem, and capacitated vehicle routing problem. Experimental results showed that the performance of our hybrid population-based meta-heuristics, compared to the best known results, is competitive in many instances. This finding indicates the effectiveness of utilizing an elite pool in our hybrid meta-heuristics in diversifying the search and subsequently enhances their performance over different instances and problems.

A standard honey-bee mating optimization algorithm (HBMO) utilizes the steepest descent local search algorithm as a worker. The steepest descent algorithm has the advantage of being simple to understand, fast and is easy to implement. However, it can easily trapped in a local optimum and subsequently restrict the performance of HBMO. Furthermore, the type of neighborhood structures that are used within the local search algorithm might impact on the performance of algorithm. This work aimed to enhance the performance of HBMO by using an adaptive guided variable neighborhood search (AGVNS) as a worker. The AGVNS algorithm is a variant of variable neighborhood search algorithm that incorporates some problem-specific knowledge and utilizes an adaptive learning mechanism to find the most suitable neighborhood structure during the searching process. In order to evaluate the effectiveness of the proposed algorithm, the Socha course timetabling dataset has been chosen as the tested domain problem. The results demonstrated that the performance of the proposed algorithm is comparable to other approaches in the literature. Indeed, the proposed algorithm obtained the best results as compared to other approaches on some instances. These results indicate the effectiveness of combining HBMO and AGVNS for solving course timetabling problems, hence demonstrated that the AGVNS can enhance the performance of HBMO.

The coronavirus disease (COVID-19), is an ongoing global pandemic caused by severe acute respiratory syndrome. Chest Computed Tomography (CT) is an effective method for detecting lung illnesses, including COVID-19. However, the CT scan is expensive and time-consuming. Therefore, this work focus on detecting COVID-19 using chest X-ray images because it is widely available, faster, and cheaper than CT scan. Many machine learning approaches such as Deep Learning, Neural Network, and Support Vector Machine; have used X-ray for detecting the COVID-19. Although the performance of those approaches is acceptable in terms of accuracy, however, they require high computational time and more memory space. Therefore, this work employs an Optimised Genetic Algorithm-Extreme Learning Machine (OGA-ELM) with three selection criteria (i.e., random, K-tournament, and roulette wheel) to detect COVID-19 using X-ray images. The most crucial strength factors of the Extreme Learning Machine (ELM) are: (i) high capability of the ELM in avoiding overfitting; (ii) its usability on binary and multi-type classifiers; and (iii) ELM could work as a kernel-based support vector machine with a structure of a neural network. These advantages make the ELM efficient in achieving an excellent learning performance. ELMs have successfully been applied in many domains, including medical domains such as breast cancer detection, pathological brain detection, and ductal carcinoma in situ detection, but not yet tested on detecting COVID-19. Hence, this work aims to identify the effectiveness of employing OGA-ELM in detecting COVID-19 using chest X-ray images. In order to reduce the dimensionality of a histogram oriented gradient features, we use principal component analysis. The performance of OGA-ELM is evaluated on a benchmark dataset containing 188 chest X-ray images with two classes: a healthy and a COVID-19 infected. The experimental result shows that the OGA-ELM achieves 100.00% accuracy with fast computation time. This demonstrates that OGA-ELM is an efficient method for COVID-19 detecting using chest X-ray images.

In this paper, a novel heuristic search algorithm called Smart Root Search (SRS) is proposed. SRS employs intelligent foraging behavior of immature, mature and hair roots of plants to explore and exploit the problem search space simultaneously. SRS divides the search space into several subspaces. It thereupon utilizes the branching and drought operations to focus on richer areas of promising subspaces while extraneous ones are not thoroughly ignored. To achieve this, the smart reactions of the SRS model are designed to act based on analyzing the heterogeneous conditions of various sections of different search spaces. In order to evaluate the performance of the SRS, it was tested on a set of known unimodal and multimodal test functions. The results were then compared with those obtained using genetic algorithms, particle swarm optimization, differential evolution and imperialist competitive algorithms and then analyzed statistically. The results demonstrated that the SRS outperformed comparative algorithms for 92% and 82% of the investigated unimodal and multimodal test functions, respectively. Therefore, the SRS is a promising nature-inspired optimization algorithm.

The text clustering is considered as one of the most effective text document analysis methods, which is applied to cluster documents as a consequence of the expanded big data and online information. Based on the review of the related work of the text clustering algorithms, these algorithms achieved reasonable clustering results for some datasets, while they failed on a wide variety of benchmark datasets. Furthermore, the performance of these algorithms was not robust due to the inefficient balance between the exploitation and exploration capabilities of the clustering algorithm. Accordingly, this research proposes a Memetic Differential Evolution algorithm (MDETC) to solve the text clustering problem, which aims to address the effect of the hybridization between the differential evolution (DE) mutation strategy with the memetic algorithm (MA). This hybridization intends to enhance the quality of text clustering and improve the exploitation and exploration capabilities of the algorithm. Our experimental results based on six standard text clustering benchmark datasets (i.e. the Laboratory of Computational Intelligence (LABIC)) have shown that the MDETC algorithm outperformed other compared clustering algorithms based on AUC metric, F-measure, and the statistical analysis. Furthermore, the MDETC is compared with the state of art text clustering algorithms and obtained almost the best results for the standard benchmark datasets.

Teenagers can easily expose to criminal activities through media social, surrounding, and peers. They tend to imitate some of these criminal activities such as cyber bullying, fighting, cyber grooming and cyber harassment without thinking of his or communities’ side-effect. Therefore, we develop a self-crime prevention modules encompassing faith, positivity, social relationship, role model and reflection tailored for school teens. Then, we make a list of survey questions using 4 Likert scale to measure their acceptance and adaptation after participating our self-crime module. About 105 school children aged 13-17 participated from Sekolah Menengah Titiwangsa and they responded positively with approximately 3.63 of average score.

Optimal match in the cover media was found to be effective for accurate data hiding and improving the visual quality of steganography algorithms. Lately, several optimized steganography algorithms including least significant bit (LSB) match, genetic algorithm (GA), particle swarm optimization (PSO) were developed for cloud data security against possible attacks. Although these algorithms can search optimal bits match in the cover media to hide the secret data precisely with less variability, the techniques consume computational time. Therefore, this paper proposed a steganography technique which can achieve almost zero variability and very low computational time. The data hiding of secret data bits in complemented or non-complemented forms were achieved and optimized by using video steganography. Furthermore, the indexes for the complemented or non-complemented form were hidden in the covered frame. This allowed information to be conveyed efficiently to the receiver when extracting the secret messages. The performance of the proposed algorithm was assessed using various parameters such as peak signal to noise ratio (PSNR), embedding capacity, normalized cross-correlation, average difference and normalized absolute error. The Results revealed that the proposed steganography system is superior compared to the existing state of the art techniques. The developed algorithm was established to be greatly effective for video data security management in cloud computing.

The ever-escalating attacks on the internet network are due to rapid technological growth. In order to surmount such challenges, multi-layer security algorithms were developed by hybridizing cryptography and steganography techniques. Consequently, the overall memory size became enormous while hybridizing these techniques. On the other side, the least significant bit (LSB) and modified LSB replacing approaches could provide the variability as detected by steganalysis technique, most found to be susceptible to attack too due to numerous reasons. To overcome these issues, in this paper a lightweight and optimized data hiding algorithm is proposed which consume less memory, provide less variability, and robust against histogram attacks. The proposed steganography system was achieved in two stages. First, data was encrypted using lightweight BORON cipher that only consumed less memory as compared to conventional algorithm such as 3DES, AES. Second, the encrypted data was hidden in the complemented or non-complemented form to obtain minimal variability. The performance of the proposed technique was evaluated in terms of avalanche effect, visual quality, embedding capacity and peak signal to noise ratio (PSNR). The results revealed that the lightweight BORON cipher could produce approximate same avalanche effect as the AES algorithm produced. Furthermore, the value of PSNR had shown much improvement in comparison to optimization algorithm GA.

University Course Timetabling Problem (UCTP) is an educational timetabling problem that deals with the task of assigning educational events (lectures, tutorials, and laboratories) and lecturers to timeslots and classrooms or lab; to generate a weekly schedule. The UCTP has gathered great research interests for decades. Many solution approaches and models have been published in the literature. However, different institutions may have different requirement, which results in a gap between benchmark UCTP and practical problems. Therefore, in this work, we have conducted a survey to investigate the satisfaction factors of a course timetable for a Practical University Course Timetabling Problem at Fakulti Teknologi dan Sains Maklumat, Universiti Kebangsaan Malaysia.

The text representation model Bag-Of-Words is a simple and typical model which uses the single words as elements to represent the texts in the feature space. However, using the single words as features will produce a high dimensional feature space, which result in the learning computational cost, particularly for ensemble learning algorithms, such as the boosting algorithm AdaBoost.MH. The straightforward solution of this matter can be managed by using a feature selection method capable of reducing the features space effectively. This work describes how to utilize the supervised topic model Labeled Latent Dirichlet Allocation for feature selection, as well accelerating AdaBoost.MH learning for multi-label text categorization. The experimental results on three benchmarks demonstrated that using Labeled Latent Dirichlet Allocation for feature selection improves and accelerates AdaBoost.MH and exceeds the performance of three existing methods.

The performance of data clustering algorithms is mainly dependent on their ability to balance between the exploration and exploitation of the search process. Although some data clustering algorithms have achieved reasonable quality solutions for some datasets, their performance across real-life datasets could be improved. This paper proposes an adaptive memetic differential evolution optimisation algorithm (AMADE) for addressing data clustering problems. The memetic algorithm (MA) employs an adaptive differential evolution (DE) mutation strategy, which can offer superior mutation performance across many combinatorial and continuous problem domains. By hybridising an adaptive DE mutation operator with the MA, we propose that it can lead to faster convergence and better balance the exploration and exploitation of the search. We would also expect that the performance of AMADE to be better than MA and DE if executed separately. Our experimental results, based on several real-life benchmark datasets, shows that AMADE outperformed other compared clustering algorithms when compared using statistical analysis. We conclude that the hybridisation of MA and the adaptive DE is a suitable approach for addressing data clustering problems and can improve the balance between global exploration and local exploitation of the optimisation algorithm.

The heuristic method is a well-known constructive method for initialize trail quality solutions in capacitated vehicle routing problem. Cheapest insertion heuristic is a popular construction heuristic known for being fast, producing decent solutions, simple to implement and easy to extend handling complicated constraints. However, in previous work, there was less focus on diverse initial quality solutions. Therefore, this study proposed an extension to the cheapest insertion heuristic which consider various combinations of seed customer criteria (the first customer inserted on a route) to preserve solutions diversification. Three seed customer criteria proposed which based on the combination of two criteria based on (farthest, nearest and random criteria). The best performing criteria selected and tested on benchmark dataset, later compared with Clarke and Wright saving heuristic. The results shown that the combination of (farthest and random) criteria obtained the best initial solution which preserve balance between the quality and diversity, with less time when compared to Clarke and wright saving heuristic. This approach is for generating diverse and quality starting solutions for the capacitated vehicle routing problem.