Hybrid Genetic Fuzzy System Research Articles

Hybrid Genetic Fuzzy System for Modeling Consumer Behavior

Understanding consumer behavior is beneficial to a business in various aspects such as prediction of manufacturing quantity, new product launch, and aids in lock-in customers and lock-out competitors. The task is highly complex and traditional models do not help in absence of generalized decision making logic. Further such domains handle large amount of data in unstructured format. This article presents an intelligent system for modeling consumer behavior via a hybrid genetic fuzzy system from large source of data. The paper justifies and presents a literature survey with common observations. A four phase generic architecture of genetic fuzzy system presented for the modeling of consumer behavior. Detailed discussion on the architecture is also provided with an experiment. Technical details, fuzzy membership functions used in experiment, encoding strategy, genetic operators, and evaluation of rules using fitness function are also discussed in detail along with results. At end, applications of the research work in other domains are enlisted with possible future enhancements.

Hybrid Computational Intelligence System for Fashion Design

The domain of fashion design evolves continuously and is highly personalised, demanding intelligent and customised recommendation. The traditional artificial intelligence-based systems offer solutions based on stored knowledge; hence they can be quickly obsolete and require high effort. To meet the fashion designers’ needs and provide tailor-made recommendations effectively, a hybrid genetic-fuzzy system is proposed with interactive fitness functions. The system is based on generic hybrid architecture using fuzzy logic and genetic algorithm, which can be used to evolve various products in different domains and tested with interactive fuzzy fitness functions. The design of the generic architecture meets the research gap identified through an in-depth literature survey. To prove the utility of the architecture, an experiment is carried out showing encoding scheme, genetic operators, fuzzy membership functions, and fuzzy rules. The results are also discussed, along with the comparison, advantages, applications, and possible future enhancements.

Architecture for Hybrid genetic fuzzy system for text summarization

International Journal of Computer Sciences and Engineering (A UGC Approved and indexed with DOI, ICI and Approved, DPI Digital Library) is one of the leading and growing open access, peer-reviewed, monthly, and scientific research journal for scientists, engineers, research scholars, and academicians, which gains a foothold in Asia and opens to the world, aims to publish original, theoretical and practical advances in Computer Science,Information Technology, Engineering (Software, Mechanical, Civil, Electronics & Electrical), and all interdisciplinary streams of Computing Sciences. It intends to disseminate original, scientific, theoretical or applied research in the field of Computer Sciences and allied fields. It provides a platform for publishing results and research with a strong empirical component. It aims to bridge the significant gap between research and practice by promoting the publication of original, novel, industry-relevant research.

Comparative study on dimensionality reduction for disease diagnosis using fuzzy classifier

Machine learning is the worldwide recent research technique for various systems as they are intelligent enough to find the solution for classification and prediction problems. The proposed work is about a hybrid genetic fuzzy algorithm that performs an optimal search as well as classification upon uncertain data. The data which is uncertain is suitable for fuzzy classifiers to predict the disease. The hybrid genetic fuzzy system applied on the attributes selects relevant attributes. The selected attributes are fed into the fuzzy classifier. The fuzzy rules are again generated using genetic algorithms. This algorithm is applied on three of the important and bench marking data sets taken from the UCI machine learning repository. The heart disease, Wisconsin breast cancer and Pima Indian diabetes datasets produce classification accuracy as 89.65%, 99.5% and 88.93% respectively. In this article there is a comparative study on few of the feature selection and feature reduction techniques.

Energy-efficient allocation of computing node slots in HPC clusters through parameter learning and hybrid genetic fuzzy system modeling

Decision-making mechanisms for online allocation of computer node slots in HPC clusters are commonly based on simple knowledge-based systems comprised of individual sets of if---then rules. In contrast with previous works where these rules were designed using expert knowledge, two different types of evolutionary learning algorithms are compared in this paper. In the first case, some of the numerical parameters defining a human-designed knowledge base are tuned. In the second case, a genetic fuzzy system evolves a partial rule set that, after being combined with some expert rules, conforms the most appropriate knowledge base for a given load scenario. In both cases, the proposed approaches optimize the quality of service and the number of node reconfigurations along with the energy consumption. An experimental study has been made using actual workloads from the Scientific Modeling Cluster at Oviedo University, and statistical evidence was found supporting the adoption of the new learning system.

Developing a bioaerosol detector using hybrid genetic fuzzy systems

The aim of this work is to develop a model, which works as a reasoning mechanism in a bioaerosol detector. Ability to distinguish between safe and harmful aerosols is one of its main requirements. Instead of commonly used misclassification rate as a metric of accuracy, true positive (TP) and false positive (FP) rates are used because of the uneven misclassification costs and class distributions of the collected data. Interpretability of the model builds up the confidence for the developed model and enables its adjustment in cases when bioaerosol detector is further developed. Thus, it is another crucial requirement for the model. Clearly, the objectives are contradicting and therefore multiobjective evolutionary algorithms (MOEAs) are applied to find tradeoff models. Fuzzy classifiers (FCs) are selected as a model type because their linguistic rules are intuitive to human beings. FCs are identified by hybrid genetic fuzzy system (GFS) which initializes the population adequately using decision trees (DTs) and simplification operations. During MOEA optimization transparency of fuzzy partition is used as a metric of interpretability and TP and FP rates as metrics of accuracy. Heuristic rule and rule condition removal is applied to offspring population in order to keep the rule base consistent. The identified FCs are highly comprehensible yet accurate and their linguistic rules provide valuable insights for further development of bioaerosol detector.

Detection of Safe and Harmful Bioaerosols by Means of Fuzzy Classifiers

This paper aims to create a fuzzy classifier (FC) to be used in a recently developed bioaerosol detector. The main requirements for FC are high true positive (TP) rate, low false positive (FP) rate, and interpretability, which is measured by transparency of fuzzy partition. Due to the contradicting nature of the above requirements, FCs are identified by hybrid genetic fuzzy system (GFS), which initializes the population using decision trees (DTs) and simplification operations. Then, a multiobjective evolutionary algorithm (MOEA) is run in order to find a Pareto-optimal set of FCs. During MOEA optimization, heuristic rule and rule condition removal is applied to keep the rule base consistent. Real-world bioaerosol data, collected from Umeå trial field, Sweden, and from laboratory of Finnish Defense Forces Technical Research Center, were used to validate the proposed GFS. By means of it, a widely spread set of interpretable and accurate FCs was obtained. Moreover, an FC based on this project was installed into the bioaerosol detector and the preliminary tests proved its capability in distinguishing between safe and harmful bioaerosols.

Learning fuzzy partitions in FIR methodology

The main goal of this research is the development of a hybrid genetic fuzzy system (GFS), composed by the fuzzy inductive reasoning (FIR) methodology and a genetic algorithm (GA) that is responsible of learning the fuzzy partitions needed in the recode process of FIR. A partition includes the number of fuzzy sets (classes) per variable and the membership function of each class. The resulting GFS is applied to two real problems, i.e. the estimation of the maintenance cost of medium voltage lines in Spanish towns and the prediction of ozone levels in Austria. The results obtained in each application are compared with some of the most popular classical statistical modeling methods, neural networks and other hybrid evolutionary data analysis techniques.

Assessment and improvement of quality of service in wireless networks using fuzzy and hybrid genetic-fuzzy approaches

Fuzzy and hybrid genetic-fuzzy approaches were used to assess and improve quality of service (QoS) in simulated wireless networks. Three real-time audio and video applications were transmitted over the networks. The QoS provided by the networks for each application was quantitatively assessed using a fuzzy inference system (FIS). Two methods to improve the networks' QoS were developed. One method was based on a FIS mechanism and the other used a hybrid genetic-fuzzy system. Both methods determined an optimised value for the minimum contention window (CW min) in IEEE 802.11 medium access control (MAC) protocol. CW min affects the time period a wireless station waits before it transmits a packet and thus its value influences QoS. The average QoS for the audio and video applications improved by 42.8% and 14.5% respectively by using the FIS method. The hybrid genetic-fuzzy system improved the average QoS for the audio and video applications by 35.7% and 16.5% respectively. The study indicated that the devised methods were effective in assessing and significantly improving QoS in wireless networks.