Research Review: Harmony Search Hyper-heuristic for Scheduling Problems
1000 until 1130
Meeting Room 7th Floor
Prof. Dr. Ahamad Tajudin Khader
Dr. Mohammed Azmi Al-Betar
This thesis is concerned with the investigation of hyper-heuristic methods. Hyper-heuristic is a new trend in optimization methods. Basically hyper-heuristic can be referred as a method to find the best heuristic to solve an optimization problem on hand. The main motivation of using hyper-heuristic is to produce a general method that can be used to solve different hard optimisation problems. In this thesis we proposed a new hyper-heuristic framework named as Harmony Search-based Hyper-heuristic (HSHH). The original idea was to apply a sequence of low-level heuristics to a selected solution in order to produce good quality solutions to given problem. Therefore, to achieve this goal, we combine three main operators in harmony search algorithm: memory consideration, random consideration and pitch adjustment as a high level heuristic in order to select and generate a sequence of improvement low-level heuristics. To demonstrate the effectiveness of the method, we apply the proposed method to three timetabling and scheduling problems, taken from the real world and our results are compared with those of other heuristic methods in the literature. Experimentally, the HSHH approach can achieve comparable results with other methods and in several instances, HSHH are able to produces competitive results with those obtained using other sophisticated methods.
Research Review: A Study on Ontology-based and Hybrid Genetic Algorithms Approach in Menu Planning Model for Malaysian Old Folks Home
1500 until 1630
Meeting Room 7th Floor
Ngo Hea Choon
Assoc. Prof. Dr. Cheah Yu-N
The number of elderly in Malaysia is not only rising rapidly but also in their life expectancies. Increasing number of old age group presents a real challenge to nutritionists and health professionals. Thus, proper nutrition for the elderly is important to maintain the health and well-being of older people that can leads fulfilling and independent lives. This research presents a study on menu planning using ontology-based and hybrid genetic algorithms approach for Malaysian old folks home in general and cancer disease in particular. Nowadays, there are many diet recommendation systems in the market that provide general advice to the clients. These systems are still insufficient to provide customized diet plan based on the older people who might be at risk of malnutrition. We are attempted to consider elderly with certain chronic disease such as cancer, diabetes, chronic kidney disease, hypertension and hyperlipidemia. In this work, we take into account the elderly with cancer disease to support their nutrition plan. In the aims to discover diets and food products that deliver health to elderly, ontology is used to classify nutrients, food groups and meal structure. Following that, Hybrid genetic algorithms are employed to ensure that the constructed menu is satisfied all the objectives and predefined constraints. Instead of Boolean logic, a fuzzy logic control system was applied in modeling of membership functions of fuzzy sets for estimating nutrition needs in elderly with cancer disease. Fuzzy membership functions are constructed to describe the range of nutrients intake in the range from deficient to excess amounts. It is important to guide dietitians towards a standardized dietary management along the nutrition care process for cancer patients in order to improve patients’ outcome. The proposed work aims to (i) produce a diet plan representation based on diet plan ontology; (ii) design a planning engine by integrating genetic algorithms with local search technique to enhance menu plan; (iii) develop a special menu plan to cater patients with cancer disease using fuzzy reasoning mechanism. The evidence from this study showed that the proposed methods yield significant improvement in menu planning model.
This work presents a comprehensive investigation on the concept of pair bonds (monogamous pairs) for the mating phase of genetic algorithms (GAs). GA is a heuristic search technique based on the principles and mechanisms of natural selection grounded on the theory of survival of the fittest. Traditionally, parents are selected at every generation to reproduce offspring through crossover and mutation operations. The process reiterates until some termination conditions are met. However, Nature sometimes exhibits the formation of enduring relationships between mating partners. In modern human society, some avian model, fish, rodents, and even lizards, pair bonds are integral aspects of their social behaviour. These species usually share the same mating partners throughout their lifetime - socially monogamous. Taking the cue from Nature, this thesis studies the feasibilities of pair bonds in GA. Consequently, two methodologies are proposed: In the first proposed methodology, coined as Monogamous Pairs Genetic Algorithm (MopGA), parents are bonded and mated consistently over several predefined generations. Selection of new parents pairs will only take place at the end of pair bond tenure. Meanwhile, competition occurs between siblings to ensure only the best are retained. Occasional infidelity generates variety and spreads genetic information through the population. Next, a parameter-free version, known as adaptive MopGA (AMopGA) is introduced. Algorithm sensitive-parameters will be tuned adaptively throughout the evolutionary process - further improving the performance and relieving from the burden of parameter-tuning. Rigorous performance investigation of both methodologies are carried out on different notable benchmarks. The preliminary tests reveal that they can be executed in less processing time without trading-off in solution quality when compare with the standard genetic algorithm.
Research Review: Utilizing Website Structure, Content and Ontologies for Web Usage Mining Preprocessing
1500 until 1630
Meeting Room 7th Floor
Mohammad Hani Nayel Al-Majali
Assoc. Prof. Dr. Cheah Yu-N
Web Usage Mining (WUM) is the process of incorporation data mining techniques over user accesses to web servers in order to discover useful patterns about users’ behavior and enhance the web surfing experience. The former technique has long been established to be significant in today’s social-web world, in which a huge number of content is being added daily to the corpus of the world wide web repository. Users’ behavior over the internet is now shaping the structure of the internet corpus, where accuracy of related content’s retrieval is becoming a troubling issue for search engines and websites’ engineers. Among the full processing of WUM, a large gap exists in investigating the tasks involved in the data preprocessing phase. Usage mining is replete with preprocessing techniques, yet most of which still process the log files based on its limited content. The need for understanding proper associations for users’ browsing behavior facilitated intensive research in the field of semantic-based web mining. Most WUM existing preprocessing techniques produce only “user Sessions”. This study investigates the current state-of-the-art WUM preprocessing techniques to highlight key concepts behind advanced techniques and present their advantages in an enhanced model of log file preprocessing able to produce enriched output in the form of “user episodes”. This research aims to include ontologies as an interpretation layer within the WUM system processes to enhance preprocessing outcomes and consider the content of web documents for episode identification and web structure for an enhanced quality of output episodes. The proposed model consists of four major tasks to mine a website’s access logs. The first is structure data preprocessing which crawls the website and presents the structure in the form of a matrix using Petri Nets maps. Discovery of reachability paths between webpages and pageview identification are achieved through the resulting matrix. The second is content data preprocessing which scrapes the documents’ content and associates it to a domain-specific ontology to find relatedness between documents. Ontology concepts are extended by labels representing hypernyms, hyponyms and synonyms from a lexical thesaurus, and with semantically related concepts from within the ontology. The third task involves preparing the log through fusion, cleaning and user-session identification before identifying user-episodes by combining all previous data. The last task is to integrate all preprocessed data together for other WUM phases. The proposed model was tested against a real-life dataset. The results are promising, confirming that the developed model successfully extracted user-episodes in a mineable form. Future work should include the integration of the proposed model into other WUM tools. However, these techniques require further investigation in terms of associating documents’ semantics into different mining categories. In addition, the proposed model could be extended to support episode discovery based on the purposes behind the mining process.