Fuzzy-based Method Research Articles

Toward Mobile Telecommunication Recommendation System through Intelligent Customers Categorization

Now a day, usage of mobile devices is becoming indispensable. This is evident with current mobile penetration rates reaching 100% and even more in some countries. Customers across the world are enjoying competitive prices due to high competition among telecommunication companies. As a result of this, it is mandatory for mobile companies to provide high quality services to their customers to retain them. One aspect which will maximize customers’ trust and lead to high retention rate is to offer them a suitable plan that matches their usage. Mobile customer usage categorization is therefore an essential task to develop intelligent business plans. Personalized recommendation system is needed to dynamically adapt the different customer behaviours with the most appropriate plan for them. In this paper we propose a new automatic approach for costumers’ categorization. This will be the basis for the recommendation system. The proposed method is built using Fuzzy rule and aims at usage behaviour prediction. The rules was extracted from real customer data obtained from a leading provider. Comparison study with other categorization methods has been conducted and showed superior result and demonstrated the potential advantage of the proposed fuzzy based method.

Fuzzy-based interactive method for solution of bi- and multi-level programming problems

In this paper, a fuzzy-based interactive method for bi-level and multi-level, linear and non-linear, programming problems is proposed. In this approach product operator is used for aggregation of different fuzzy goals and MI-LXPM, a genetic algorithm, is used as optimisation technique. Performance of the proposed method is tested on a set of five examples taken from literature.

Fuzzy-Based Segmentation for Variable Font-Sized Text Extraction from Images/Videos

Textual information embedded in multimedia can provide a vital tool for indexing and retrieval. A lot of work is done in the field of text localization and detection because of its very fundamental importance. One of the biggest challenges of text detection is to deal with variation in font sizes and image resolution. This problem gets elevated due to the undersegmentation or oversegmentation of the regions in an image. The paper addresses this problem by proposing a solution using novel fuzzy-based method. This paper advocates postprocessing segmentation method that can solve the problem of variation in text sizes and image resolution. The methodology is tested on ICDAR 2011 Robust Reading Challenge dataset which amply proves the strength of the recommended method.

A new modified bat algorithm to solve optimal management of multi-objective reconfiguration problem

Nowadays, the Distribution Feeder Reconfiguration (DFR) is known as a famous reinforcement technique to enhance the radial distribution abilities. So that the optimal management of DFR problem can lead to several profits such as load balance improvement, power loss reduction, voltage profile intensification, etc. Also, this paper aims to take the positive effect of the DFR strategy on the network performance under investigation. In this regard, an evolutionary algorithm based on bat algorithm (BA) is suggested to study the DFR problem deeply. The total active power losses, voltage deviation and load balance are considered as objective functions. Thus, a new self-adaptive modification is implemented to search the multi-objective problem space globally. Other than that, in order to satisfy all the objective functions efficiently, a fuzzy-based interactive method is applied. This idea will give the operator the capability to decide the satisfying degree of each objective function. In the end, the proposed algorithm is tested on the 32-bus IEEE radial distribution system to demonstrate the feasibility and effectiveness of the method.

A multi-criteria fuzzy method for selecting the location of a solid waste disposal facility

Facility location is a multi-criteria decision process that has important operational and economic impacts and that typically involves uncertainty and vagueness of evaluations. A fuzzy-based method supporting preliminary decision-making about sitting solid waste incinerators is proposed building on a structured classification of criteria for location selection developed from the existing literature. The application to a case study revealed the advantages of the methodology. The work intends to provide a general and comprehensive taxonomy of decision criteria that may be adapted to various facility location problems together with a fuzzy inference process that is useful for companies and public administration institutions looking for rigorous but relatively simple decision-making tools in uncertain environments. Future research will compare the developed method with the most common tools for making location decisions. The approach will be then extended to different kinds of facilities.

OPTIMIZATION OF AN ULTRATHIN CENTRIFUGAL FAN BASED ON THE TAGUCHI METHOD WITH FUZZY LOGICS

This paper presents the use of fuzzy-based Taguchi method to optimize the design of the ultrathin centrifugal fan with multiple performance characteristics. An orthogonal array, the signal-to-noise (S/N) ratio, multiresponse performance index, analysis of variance (ANOVA), and computational-fluid-dynamics were used to study the multiple-objectives in the ultrathin centrifugal fan design. The design parameters, outlet dimensions, inlet dimensions, blade angle, and impeller diameter were optimized with considerations of the performance characteristics, including volume flow ratio, static pressure, and noise. The results demonstrate that volume flow rate of the new design fan was almost 29% larger than that of the original design. This study also identified the optimized design parameters that affect the cooling performance of the centrifugal fan.

A novel unsupervised fuzzy clustering method for preprocessing of quantitative attributes in association rule mining

In Association rule mining, the quantitative attribute values are converted into Boolean values using fixed intervals. Conventional association rule mining algorithms are then applied to find relations among the attribute values. These intervals may not be concise and meaningful enough for human users to easily obtain non trivial knowledge from those rules discovered. Clustering techniques can be used for segmenting quantitative values into meaningful groups instead of fixed intervals. But the conventional clustering techniques like k-means and c-means require the user to specify the number of clusters and initial cluster centres. This initialization is one of the major challenges of clustering. A novel fuzzy based unsupervised clustering algorithm proposed by the authors is extended to segment quantitative values into fuzzy clusters in this paper. Membership values of quantitative items in the partitioning fuzzy clusters are used with weighted fuzzy rule mining techniques to find natural association rules. This fuzzy based method for handling quantitative attributes is compared with that of fixed intervals and segmenting using conventional k-means clustering method along with Apriori algorithm.

Adaptive multimodal vibration suppression using fuzzy-based control with limited structural data

We propose a novel fuzzy-based method of adaptive multimodal vibration suppression with limited structural data. The adaptive control consists of fuzzy inference and a semi-active switching approach. We demonstrate it to be applicable to multimodal vibration suppression for vibrating structures, where a single piezoelectric actuator suppresses two modal vibrations simultaneously. Our fuzzy-based semi-active control requires only the structural information of natural frequencies for real-time adaptive feedback, whereas common adaptive controls require highly precise structural models or complete equations of motion. We conduct experiments in semi-active vibration suppression using the proposed fuzzy-based control, and compare the suppression performance of our fuzzy-based approach with conventional controls. The experiments indicate that the proposed fuzzy-based control demonstrates good adaptability when experiencing sudden changes in disturbance excitation, and also demonstrates high suppression performance. The fuzzy-based control can adapt to a wide range of disturbance conditions, both within and outside the range of vibration excitations assumed when the controller is designed.

Use of the Fuzzy-Based Taguchi Method for Improving the Yield of BGA Packaging

This study presents the use of fuzzy-based Taguchi method for improving of the yield of the floating device of the ball grid array (BGA) mold chase in the molding process with multiple performance characteristics. The objective of this study was to obtain the maximal pressure of the floating device, minimal processing volume, and minimal processing projected area. The results indicate that the frequency of replacing the appropriate spacer of the mold chase decreased by 75 %, the tolerance of the substrate increased twofold, the utilization rate of the auto-mold machine increased by 12 %, and the production capacity was upgraded to 1.7 kpcs in one day.

New Fuzzy-Based Retinex Method for the Illumination Normalization of Face Recognition

We propose a new illumination normalization for face recognition which robust in relation to the illumination variations on mobile devices. This research is novel in the following five ways when compared to previous works: (i) a new fuzzy-based Retinex method is proposed for illumination normalization; (ii) the performance of face recognition is enhanced by determining the optimal parameter of Retinex filtering based on fuzzy logic; (iii) the output of the fuzzy membership function is adaptively determined based on the mean and standard deviations of the grey values of the detected face region; (iv) through the comparison of various defuzzification methods in terms of the accuracy of face recognition, one optimal method was selected; (v) we proved the validations of the proposed method by testing it with various face recognition methods. Experimental results showed that the accuracy of the face recognition with the proposed method was enhanced compared to previous ones.

Seismic induced damageability evaluation of steel buildings: a Fuzzy-TOPSIS method

Seismic resiliency of new buildings has improved over the years due to better seismic codes and design practices. However, there is still large number of vulnerable and seismically deficient buildings. It is not economically feasible to retrofit and upgrade all vulnerable buildings, thus there is a need for rapid screening tool. Many factors contribute to the damageability of buildings; this makes seismic evaluation a complex multi-criteria decision making problem. Many of these factors are noncommensurable and involve subjectivity in evaluation that highlights the use of fuzzy-based method. In this paper, a risk-based framework earlier proposed by Tesfamariam and Saatcioglu (2008a) is extended using Fuzzy-TOPSIS method and applied to develop an evaluation and ranking scheme for steel buildings. The ranking is based on damageability that can help decision makers interpret the results and take appropriate decision actions. Finally, the application of conceptual model is demonstrated through a case study of 1994 Northridge earthquake data on seismic damage of steel buildings.

A Fuzzy Based Solution for Allocation and Sizing of Multiple Active Power Filters

Active power filters (APF) can be employed for harmonic compensation in power systems. In this paper, a fuzzy based method is proposed for identification of probable APF nodes of a radial distribution system. The modified adaptive particle swarm optimization (MAPSO) technique is used for final selection of the APFs size. A combination of Fuzzy-MAPSO method is implemented to determine the optimal allocation and size of APFs. New fuzzy membership functions are formulated where the harmonic current membership is an exponential function of the nodal injecting harmonic current. Harmonic voltage membership has been formulated as a function of the node harmonic voltage. The product operator shows better performance than the AND operator because all harmonics are considered in computing membership function. For evaluating the proposed method, it has been applied to the 5-bus and 18-bus test systems, respectively, which the results appear satisfactorily. The proposed membership functions are new at the APF placement problem so that weighting factors can be changed proportional to objective function.

Fuzzy Approach for Semantic Face Image Retrieval

Our aim is to realize semantic face image retrieval (FIR)\ by bridging the semantic-gap between low-level visual features of images and the high-level labels that describe facial features. This allows a user to retrieve face images based on a description of face features rather than an example image. We approach the semantic-gap problem by developing a fuzzy-based method of finding a mapping between the low-level features and the vocabulary that describes the semantics of the face features. The contribution of this paper is a new method of using fuzzy clustering and fuzzy inference methods to derive the degree of membership for each semantic label to a new image. Our experiments show that our approach has good results for annotating images, and provides a sound foundation for local face feature-based FIR systems. We have made available a demonstration system online. Further, our system is not domain specific and can be generalized and applied to other problems in the field of image retrieval.

Rules modification on a Fuzzy-based modular architecture for medical device design and development

Medical devices have a very high failure rate in their first prototype tests. According to the international testing body Intertek, out of every ten medical devices, nine fail in their first prototype tests—a 90% failure rate. In addition to the cost implication, quality is a key issue. To address this, we present an integrated, collaborative modular architecture method for medical device design and development. The methodology focuses on analyzing the input of stakeholder data from existing products and components to achieve an optimal number of modules. The objective of this research is to investigate the effect of rules modification on the final number of product modules. The methodology starts by defining a product's functional and physical decompositions. Next, product parameters are selected and prioritized using an analytical hierarchy process (AHP) to determine the medical device manufacturers’ focus area(s). Candidate modules are evaluated by acquiring stakeholder data and converting them to crisp values by applying the fuzzy-based Sugeno method. Optimal module values are then determined using a multi-optimization goal programming model. Finally, we analyse the effect of changing the number of fuzzy rules on the optimal number of modules and minimum deviation, ‘d’. A typical glucometer is used for a proof of concept. The implication of this work is the determination that the optimal number of product modules is affected by the rules changes.

Automatic Metadata Mining from Multilingual Enterprise Content

Personalization is increasingly vital especially for enterprises to be able to reach their customers. The key challenge in supporting personalization is the need for rich metadata, such as metadata about structural relationships, subject/concept relations between documents and cognitive metadata about documents (e.g. difficulty of a document). Manual annotation of large knowledge bases with such rich metadata is not scalable. As well as, automatic mining of cognitive metadata is challenging since it is very difficult to understand underlying intellectual knowledge about document automatically. On the other hand, the Web content is increasing becoming multilingual since growing amount of data generated on the Web is non-English. Current metadata extraction systems are generally based on English content and this requires to be revolutionized in order to adapt to the changing dynamics of the Web. To alleviate these problems, we introduce a novel automatic metadata extraction framework, which is based on a novel fuzzy based method for automatic cognitive metadata generation and uses different document parsing algorithms to extract rich metadata from multilingual enterprise content using the newly developed DocBook, Resource Type and Topic ontologies. Since the metadata generation process is based upon DocBook structured enterprise content, our framework is focused on enterprise documents and content which is loosely based on the DocBook type of formatting. DocBook is a common documentation formatting to formally produce corporate data and it is adopted by many enterprises. The proposed framework is illustrated and evaluated on English, German and French versions of the Symantec Norton 360 knowledge bases. The user study showed that the proposed fuzzy-based method generates reasonably accurate values with an average precision of 89.39% on the metadata values of document difficulty, document interactivity level and document interactivity type. The proposed fuzzy inference system achieves improved results compared to a rulebased reasoner for difficulty metadata extraction (~11% enhancement). In addition, user perceived metadata quality scores (mean of 5.57 out of 6) found to be high and automated metadata analysis showed that the extracted metadata is high quality and can be suitable for personalized information retrieval.

Automatic metadata mining from multilingual enterprise content

Personalization is increasingly vital especially for enterprises to be able to reach their customers. The key challenge in supporting personalization is the need for rich metadata, such as metadata about structural relationships, subject/concept relations between documents and cognitive metadata about documents (e.g. difficulty of a document). Manual annotation of large knowledge bases with such rich metadata is not scalable. As well as, automatic mining of cognitive metadata is challenging since it is very difficult to understand underlying intellectual knowledge about document automatically. On the other hand, the Web content is increasing becoming multilingual since growing amount of data generated on the Web is non-English. Current metadata extraction systems are generally based on English content and this requires to be revolutionized in order to adapt to the changing dynamics of the Web. To alleviate these problems, we introduce a novel automatic metadata extraction framework, which is based on a novel fuzzy based method for automatic cognitive metadata generation and uses different document parsing algorithms to extract rich metadata from multilingual enterprise content using the newly developed DocBook, Resource Type and Topic ontologies. Since the metadata generation process is based upon DocBook structured enterprise content, our framework is focused on enterprise documents and content which is loosely based on the DocBook type of formatting. DocBook is a common documentation formatting to formally produce corporate data and it is adopted by many enterprises. The proposed framework is illustrated and evaluated on English, German and French versions of the Symantec Norton 360 knowledge bases. The user study showed that the proposed fuzzy-based method generates reasonably accurate values with an average precision of 89.39% on the metadata values of document difficulty, document interactivity level and document interactivity type. The proposed fuzzy inference system achieves improved results compared to a rule-based reasoner for difficulty metadata extraction (∼11% enhancement). In addition, user perceived metadata quality scores (mean of 5.57 out of 6) found to be high and automated metadata analysis showed that the extracted metadata is high quality and can be suitable for personalized information retrieval.

AN ADAPTIVE DIGITAL IMAGE WATERMARKING USING FUZZY GRADIENT ON DCT DOMAIN

In this paper, an adaptive digital image watermarking technique using fuzzy gradient on DCT domain is presented. In our approach, the image is divided into separate blocks and the DCT is applied on each block individually. Then, the watermark is inserted in the transform domain and the inverse transform is carried out. We increase the robustness of the watermark by increasing the watermark strength. However, this reduces the fidelity of the watermarking scheme. This is because the fidelity and robustness of watermarking are generally in conflict with each other. To improve the fidelity, a new fuzzy-based method is introduced. In this method, a fuzzy gradient-based mask is generated from the host image. Then, as a post-processing stage, the generated mask is combined with the watermarked image. Experimental results show that the proposed technique has high fidelity as well as high robustness against a variety of attacks.

A fuzzy based method for leveling output power fluctuations of photovoltaic-diesel hybrid power system

A Photovoltaic system’s output power fluctuates as insolation varies with weather condition. Fluctuating PV power causes frequency deviations when large PV power is penetrated in the isolated utility. In this paper, a fuzzy based method for leveling the fluctuations of PV power in a PV-diesel hybrid power system is proposed. By means of the proposed method, output power control of PV system becomes possible considering power utility conditions and the conflicting objective of output power leveling and maximizing energy capture is achieved. Here, fuzzy control is used to generate the output leveling power command. The fuzzy control has three inputs of average insolation, variance of insolation, and absolute average of frequency deviation. First, the proposed method is compared with the method where captured maximum power is given to the utility without leveling. Second, the proposed method is compared with a conventional method where captured maximum power is leveled by using an energy storage system and is given to the isolated utility. Simulation results show that the proposed method is effective in leveling PV power fluctuations and is feasible to reduce the frequency deviations of the isolated power utility.

WF-MSB: A weighted fuzzy-based biclustering method for gene expression data

Biclustering is an important analysis method on gene expression data for finding a subset of genes sharing compatible expression patterns. Although some biclustering algorithms have been proposed, few provided a query-driven approach for biologists to search the biclusters, which contain a certain gene of interest. In this paper, we proposed a generalised fuzzy-based approach, namely Weighted Fuzzy-based Maximum Similarity Biclustering (WF-MSB), for extracting a query-driven bicluster based on the user-defined reference gene. A fuzzy-based similarity measurement and condition weighting approach are used to extract significant biclusters in expression levels. Both of the most similar bicluster and the most dissimilar bicluster to the reference gene are discovered by WF-MSB. The proposed WF-MSB method was evaluated in comparison with MSBE on a real yeast microarray data and synthetic data sets. The experimental results show that WF-MSB can effectively find the biclusters with significant GO-based functional meanings.

A fuzzy-based method for the regional validation of global maps: the case of MODIS-derived phenological classes in a mega-diverse zone

Regional validation studies of global land cover products are usually based on medium-resolution imagery and auxiliary high-resolution information. Because of the heterogeneity of landscapes at 500–1000 m resolution and the characteristics of the reference sites/data, the accuracy of the results may depend on the methodology used. This work proposes an integration scheme for the regional validation of global products. The assessment produces an accuracy index with possibility margins, representing various levels of positional, thematic and temporal tolerance. The methodology is used to evaluate MOD12Q1 on 1 million km2 of the land cover in Mexico, with emphasis on the diversity of forest types/phenologies. The area of higher confusion is reported in deciduous tropical dry forests, which tend to be classified as evergreen forests, and in oak forests, which tend to be classified as grassland. The index could be used as a tool for the consistent comparison between validation results among regions.