By Topic

Fuzzy Information Processing Society, 2007. NAFIPS '07. Annual Meeting of the North American

Date 24-27 June 2007

Filter Results

Displaying Results 1 - 25 of 127
  • [Front cover]

    Page(s): C1
    Save to Project icon | Request Permissions | PDF file iconPDF (839 KB)  
    Freely Available from IEEE
  • [Breaker page]

    Page(s): 1
    Save to Project icon | Request Permissions | PDF file iconPDF (39 KB)  
    Freely Available from IEEE
  • Table of contents

    Page(s): iii - x
    Save to Project icon | Request Permissions | PDF file iconPDF (176 KB)  
    Freely Available from IEEE
  • [Opinion]

    Page(s): xii
    Save to Project icon | Request Permissions | PDF file iconPDF (56 KB)  
    Freely Available from IEEE
  • [Opinion]

    Page(s): xiii
    Save to Project icon | Request Permissions | PDF file iconPDF (54 KB)  
    Freely Available from IEEE
  • [Opinion]

    Page(s): xi
    Save to Project icon | Request Permissions | PDF file iconPDF (85 KB)  
    Freely Available from IEEE
  • A Fuzzy Inverse Model Construction Method for a General MISO System with a Monotonic Input-output Relationship

    Page(s): 1 - 6
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (408 KB) |  | HTML iconHTML  

    This paper presents a novel method of systematically constructing the fuzzy inverse model for a general multi-input single-output (MISO) system represented with triangular input membership functions, singleton output membership function and fuzzy-mean defuzzification. The fuzzy inverse model construction method has the ability of uniquely determining the inverse relationship for each input-output pair. It is derived in a straightforward way and the required input variables can be simultaneously obtained by the fuzzy inferencing calculation to realize the desired output value. Simulation examples are provided to demonstrate the effectiveness of the proposed method to find the inverse kinematics solutions for complex industrial robot manipulators. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Visualization of Item Features, Customer Preference and Associated Uncertainty using Fuzzy Sets

    Page(s): 7 - 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (760 KB) |  | HTML iconHTML  

    Some of the requirements during preferences discovery or preference modeling through machine learning and data mining are: (i) understanding of features of items, e.g. genres content of a movie; (ii) understanding of patterns in customer feedback on items to explore and identify customer preferences; (iii) understanding of discovered patterns in customer preference on features of items, e.g. preference to genres of movies; and (iv) understanding of similarity among customers' preference, e.g. to form similar cluster of customers in their genre preference. An attempt is made to satisfy these requirements using fuzzy set driven information visualization technique; and movie as the item and genre as the feature are used for illustration. Visualization of features of items, patterns of customer previous feedback to these items, and the relationship between the feedback and item features along with associated measure of uncertainty are presented. The uncertainty is non-stochastic type that is induced from subjectivity, vagueness and imprecision in item features and user preference; and it is modeled using fuzzy set. The visualization of the discovered preference along various demographic features of the users is also presented. This in turn can help forming clusters of users with similar preference to various kinds of items. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • EHSS Velocity Control by Fuzzy Neural Networks

    Page(s): 13 - 18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (555 KB) |  | HTML iconHTML  

    In this paper a fuzzy neural network (FNN) is presented for velocity control of an electro hydraulic servo system (EHSS) in presence of flow nonlinearties and internal friction. The system contains several major nonlinearties that limit the ability of simple controllers in achieving satisfactory performance. These nonlinearties include: valve dead zones, valve flow saturation, and cylinder seal friction. The performances achievable by classical linear controllers, e.g. PD, are usually limited due to highly nonlinear behavior of the hydraulic dynamics. It is shown that the fuzzy neural controller, which is employed in this paper, can be successfully used to stabilize any chosen operating point of the system. The EBP (error back propagation) method is employed in FNN and the advantaged are mentioned. The approach can be further extended to the control of hydraulically driven manipulators. All derived results are validated by computer simulation of a nonlinear mathematical model of the system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modern Data Visualization for Air Traffic Management

    Page(s): 19 - 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (787 KB) |  | HTML iconHTML  

    Air traffic at airports is affected by various factors. The capacity of an airport and the demand at a certain point in time are serious parameters that account for a big extent to aircraft delay and related variables. It has been proven that weather is another important impact in this regard. Although weather cannot be controlled, the knowledge of how weather affects the air traffic at an airport can be very helpful to optimize air traffic management. Data mining promises to gain that knowledge. Usually, the very first step in data mining is data visualization. In this paper we discuss two new visualization techniques that allow to visualize aviation data and weather data in order to contribute to the optimization process. These modern multi-dimensional scaling techniques provide mappings of high-dimensional data to low-dimensional feature spaces. We will show some results on a practical application of a major European airport. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Clustering without Use of Prototypes with Gradient Descent for Cluster Optimization

    Page(s): 25 - 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (275 KB) |  | HTML iconHTML  

    The first stage of organizing entities is to partition them into groups or clusters. The clustering is generally done on pattern vectors representing the entities. Clustering algorithms normally require a method of aggregating patterns and of measuring proximity between patterns. Because of the nature of the patterns it may not always be possible however to find a satisfactory method of aggregating patterns. Some of the features may not be numeric. Sometimes patterns may not even be available and only the proximities between patterns are known. This paper describes a method for finding a fuzzy membership matrix that provides cluster membership values for all the patterns based strictly on the proximity matrix. The method is based on the premise that the proximities between the membership vectors should be proportional to the proximities between the feature vectors. The membership matrix is found by applying gradient descent to an error function with the objective of reducing it to zero. Simulations show the method to be quite effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using Consensus to Measure Weighted Targeted Agreement

    Page(s): 31 - 35
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (216 KB) |  | HTML iconHTML  

    The information-theoretic measures of consensus, dissent and agreement are used to address the problem of the assignment of weights in recognition of expert opinions, and interval weights to reflect categorical weights. All measures are bounded in the 0 to 1 interval. Dissent is also interpreted as an indicator of dispersion. Thus, the values selected by a panel of experts is calculated for each targeted category and the category with the highest resulting value is the one chosen to represent the overall expert judgment. Further, the distances between threat levels can be calculated and the dispersion for the distribution may also be calculated. This is different from the standard statistical measures of variance for categorical values are based on an ordinal scale of ordered categories and the standard deviation requires the presence of an interval or ratio scale. Illustrations are shown to describe the functionality of the measures. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy PID Supervision for a Nonlinear, System: Design and Implementation

    Page(s): 36 - 41
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4472 KB) |  | HTML iconHTML  

    This paper presents an approach to design fuzzy supervisor for PID controllers used in the control of nonlinear systems. In the proposed approach, the design is based on an output performance criterion, namely a response with minimum settling time and without overshoot, whatever the operation conditions may be. The application used for a demonstration is a three-tank-system where the connection between the tanks and the leakage in each tank are simultaneously taken into account (MIMO system). According to the proposed approach, two fuzzy PI supervisors are designed in order to modify on-line the parameters of two PI controllers. The implementation of this type of control in the simulator of the three-tank-system model and in the laboratory installation using rapid protyping, confirms through the results that the output performances are reached whatever the set points, the configuration and the initial water levels in the three tanks may be. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Novel Granular Neural Network Architecture

    Page(s): 42 - 47
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (295 KB) |  | HTML iconHTML  

    We introduce a novel granular neural network (GNN) architecture based on the multi-layer perceptron architecture. The GNN uses linguistic terms as connection weights, and uses the operations of linguistic arithmetic to update those connection weights. The GNN has been implemented in a Java-based simulation environment, with support for both regression and classification learning tasks. We present the results of a preliminary experimental comparison between the GNN and the c4.5 decision tree algorithm on two benchmark datasets. Our results show that the GNN was slightly more accurate than c4.5 on both datasets. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • ECG Arrhythmia Detection Using Fuzzy Classifiers

    Page(s): 48 - 53
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (304 KB) |  | HTML iconHTML  

    An electrocardiogram (ECG) arrhythmia detection system has been developed. Piecewise continuous trapezoidal fuzzy membership functions and defuzzification schemes have been designed to be used in a fuzzy classifier. Fourteen types of arrhythmias and abnormalities can be detected implementing the classifier. We have evaluated the algorithm on MIT-BIH database. The classifier achieved a sensitivity of 99.18% plusmn 2.75 and a positive predictivity of 98.00% plusmn 4.45 in detecting twelve out of fourteen arrhythmias, but a sensitivity of 53.12% plusmn 34.04 and a positive predictivity of 36.80% plusmn 40.26 are designated to the other two. Due to the acceptable results, the novelty of the classification procedure and its fast application, the method is recommended for further study and practical implementation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Relevance Feedback for Association Rules using Fuzzy Score Aggregation

    Page(s): 54 - 59
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (238 KB) |  | HTML iconHTML  

    We propose a novel and more flexible relevance feedback for association rules which is based on a fuzzy notion of relevance. Our approach transforms association rules into a vector-based representation using some inspiration from document vectors in information retrieval. These vectors are used as the basis for a relevance feedback approach which builds a knowledge base of rules previously rated as (un)interesting by a user. Given an association rule the vector representation is used to obtain a fuzzy score of how much this rule contradicts a rule in the knowledge base. This yields a set of relevance scores for each assessed rule which still need to be aggregated. Rather than relying on a certain aggregation measure we utilize OWA operators for score aggregation to gain a high degree of flexibility and understandability. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy Parking Manoeuvres of Wheeled Mobile Robots

    Page(s): 60 - 65
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (432 KB) |  | HTML iconHTML  

    This work deals with the parking manoeuvres problem for a wheeled mobile robot (WMR). The robot has the same non-holonomic kinematic constraint that has a car vehicle. This constraint makes the robot having its direction always tangent to the trajectory. Two sub-cases of parking problems are considered. These are forward and backward maneuvers, aiming to stabilize the robot at a pre-specified pose. The environment is assumed to be known, obstacle-free and a local map of the area is already done by prior processing the information obtained from ultrasonic sensors mounted on the robot. A linguistic fuzzy model to represent the robot and its environment is developed. From this model, the parking manoeuvres are carried out by mimicking a human car driver using a fuzzy control system. Upon simulation tests this approach has been proved efficient giving very encouraging results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy Clustering of Baseball Statistics

    Page(s): 66 - 68
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (298 KB) |  | HTML iconHTML  

    A previous investigation into the ability of fuzzy clustering to be a sound method for comparing Major League Baseball players' batting averages yielded promising results. Yet, the study involved a rather small sample of 90 batting averages, which were fuzzy clustered into three categories. Furthermore, the primary study focused on batting averages, a statistic that is incapable of reflecting a player's hitting ability on its own, and it certainly does not account for the defensive skill of a player. While the original work highlighted some of the inherent advantages to fuzzy clustering, the small sample size and number of groups did not allow for a complete spectrum to be generated. Without an uncondensed continuum from which to compare the relative overall skills of all players, the original results limited the practical applications of fuzzy clustering in Major League Baseball. The current research aims to greatly improve upon the first study and emphasize the potential gains from the implementation of fuzzy clustering in the practices of Major League Baseball. In an effort to provide a more comprehensive analysis of baseball statistics, this investigation includes two additional hitting statistics, on base percentage and slugging percentage, and incorporates fielding percentage. The three added statistics reflect a player's bat control, power, and defensive reliability, respectively, all of which teams use to gauge a player's skills. All three offensive statistics are averaged to generate an inclusive measure of a player's offensive capabilities, and the corresponding fielding percentage was added as a second dimension into the fuzzy clustering program. The new four-input model is a more developed and more applicable version of the one produced in the original research. Fuzzy clustering of batting averages, on base percentages, slugging percentages, and fielding percentages is an innovative way for teams to compare an individual's skills to that of all pro- fessional players simultaneously, since fuzzy clustering is ideal for establishing relationships between data that would not normally be associated. Baseball statisticians will no longer be forced to merely note the numerical difference between players' three key hitting statistics and a critical defensive measure. Instead, players can be grouped according to their relative production, providing organizations with a more comprehensive view of players' capabilities. In this investigation, 968 Major League Baseball players' selected statistics were fuzzy clustered into nine groups, in an effort to better express the range of baseball skills. The results of the research offer insight into an amount of data that cannot efficiently be processed by an individual, which would make fuzzy clustering of batting averages an invaluable tool for Major League Baseball. Motivational resources are greatly needed in such a mentally and emotionally draining sport, and fuzzy-clustered statistics would provide organizations with such a resource. Players can be greatly uplifted when shown that they rank relatively well among their peers. Successful players can also use their relative groupings to secure better contracts. Owners, however, can save money in contract negotiations and retain desired players that are currently not performing well, according to the fuzzy cluster to which they belong. An additional way for owners to conserve funds is by utilizing fuzzy clustering in the scouting process. Instead of large travel budgets, fuzzy clustering can be used to compare the skill of a prospect among his peers. Finally, the teams that are the quickest to apply fuzzy clustering of baseball data to player trades will gain a competitive edge. Teams will seek trades for players of the same relative skill level, and possibly receive two or more players for one that has slightly better nominal numbers. Major League Baseball is filled with countless amounts of data. Fuzzy clustering of statis View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Applying Novel Resampling Strategies To Software Defect Prediction

    Page(s): 69 - 72
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (267 KB) |  | HTML iconHTML  

    Due to the tremendous complexity and sophistication of software, improving software reliability is an enormously difficult task. We study the software defect prediction problem, which focuses on predicting which modules will experience a failure during operation. Numerous studies have applied machine learning to software defect prediction; however, skewness in defect-prediction datasets usually undermines the learning algorithms. The resulting classifiers will often never predict the faulty minority class. This problem is well known in machine learning and is often referred to as learning from unbalanced datasets. We examine stratification, a widely used technique for learning unbalanced data that has received little attention in software defect prediction. Our experiments are focused on the SMOTE technique, which is a method of over-sampling minority-class examples. Our goal is to determine if SMOTE can improve recognition of defect-prone modules, and at what cost. Our experiments demonstrate that after SMOTE resampling, we have a more balanced classification. We found an improvement of at least 23% in the average geometric mean classification accuracy on four benchmark datasets. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Methodology for Statistical Matching with Fuzzy Logic

    Page(s): 73 - 78
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB) |  | HTML iconHTML  

    The Analysis of data often requires information that is not available from a single source, but from multiple sources. Statistical matching procedures are methods that help to merge information from different sources into a single data set. Traditionally, statistical matching is done on the basis of computed distances between selected variables found in all data sets. Situations where no decision can be made in traditional statistical matching, e.g., in the case of identical distances, cause problems. We present a methodology for statistical matching with fuzzy logic which solves these problems. After a short introduction, the basics of traditional statistical matching are presented. The description of the theory of statistical fuzzy matching follows thereafter. The paper concludes with a short example. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ontology-based Fuzzy Inference Agent for Diabetes Classification

    Page(s): 79 - 83
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2582 KB) |  | HTML iconHTML  

    Diabetes is a chronic illness that requires continuing medical care and patient self-management to prevent acute complications and to reduce the risk of long-term complications. This paper presents an ontology-based fuzzy inference agent, including a fuzzy inference engine, and a fuzzy rule base, for diabetes classification. The diabetes disease dataset used in our study is retrieved from the UCI Machine Learning Database. The experimental results indicate that the proposed approach can work effectively for classifying the diabetes. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Conceptualized Query for Information Retrieval

    Page(s): 84 - 88
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (380 KB) |  | HTML iconHTML  

    Many search engines are term-based information retrieval models. The disadvantage of this type of model is that it does not consider word sense. If we can represent the meanings of the terms that a user inputs, the IR system can retrieve the information the user really wants; not simply match the terms. To represent word sense, we proposed conceptual fuzzy sets (CFSs). A CFS is a framework that represents word concepts and that changes dynamically with fuzzy sets. In this paper, we experiment with concept retrieval for documents using conceptualized queries using CFSs. In our experiment, we evaluated our system on a large-scale corpus consisting of 1 million newswire text data. The experimental results showed that the performance of the IR system was improved. It also indicated that generating conceptualized queries is effective in an IR system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • About the Division Operator in a Possibilistic Database Framework

    Page(s): 89 - 94
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2872 KB) |  | HTML iconHTML  

    This paper is situated in the area of possibilistic databases. Any possibilistic database has a canonical interpretation as a set of more or less possible regular databases, also called worlds. In order to manipulate such databases in a safe and efficient way, a constrained framework has been previously proposed, where a restricted number of querying operations are permitted (selection, union, projection and foreign key join which may handle attributes taking imprecise values). The key for efficiency resides in the fact that these operators do not require to make computations explicitly over all the more or less possible worlds. The division operation is dealt with in this paper. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Agent Model for the Control of Smart Appliances

    Page(s): 95 - 99
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (415 KB) |  | HTML iconHTML  

    This paper proposed an agent model for the control of smart appliances. A smart appliance here refers to a home appliance with limited computation and network capabilities that can communicate with other network devices. The simple control protocol (SCP) technology was used to connect smart appliances over power lines to provide agents with an integrated network environment. The proposed agent model consists of four software units: behavior learner, control operator, event handler, and message carrier. A hybrid soft computing approach was used to analyze the appliance usage patterns for learning users' behaviors. Through the collaboration of software agents, the control of home appliances can be achieved to meet users' expectation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Control Performance Comparison between a Type-2 Fuzzy Controller and a Comparable Conventional Mamdani Fuzzy Controller

    Page(s): 100 - 105
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (380 KB) |  | HTML iconHTML  

    Control performance comparison between a type-1 fuzzy controller (Tl-FC) and a comparable type-2 fuzzy controller (T2-FC) was carried out using computer simulation. Our objective was to study whether T2 fuzzy control always had a control performance advantage over its Tl counterpart as claimed in some simulation-based reports. We used a genetic algorithm to optimize the Tl-FC and the T2-FCs that control process models of three different types (i.e., linear, linear with a time-delay, and nonlinear). Controllers' robustness against model parameter variation and capabilities of dealing with random noise were compared as well. The simulation results show that different settings result in different comparison outcomes: (1) the Tl-FC and the T2-FC performed (almost) identically, and (2) the T2-FC outperformed its Tl counterpart, and (3) the T1-FC was superior. These results are theoretically sensible because from the controllers' input-output mapping standpoint, their ability to produce continuous nonlinear control functions should be similar and no inherent advantage likely exists. Thus, one controller can appear to be better than, worse than, or equal to its counterpart depending on the specific configuration of the whole control system. Consequently, no one should claim that T2 fuzzy control is generally better than T1 fuzzy control. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.