By Topic

Intelligent and Advanced Systems, 2007. ICIAS 2007. International Conference on

Date 25-28 Nov. 2007

Filter Results

Displaying Results 1 - 25 of 281
  • Performance guarantees for agent-based Hierarchical Diff-EDF scheduler

    Page(s): 1 - 6
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1927 KB) |  | HTML iconHTML  

    Packet networks are currently enabling the integration of heterogeneous traffic with a wide range of characteristics that extend from video traffic with stringent QoS requirements to best-effort traffic requiring no guarantees. QoS guarantees can be provided in packet networks by the use of proper packet scheduling algorithms. Similar to the trends of computer revolution, many scheduling algorithms have been proposed to meet this goal. In this paper, we propose a new priority assignment scheduling algorithm, Hierarchical Diff-EDF, which can meet the real-time needs while continuing to provide best effort service over heterogeneous real-time network traffic. The Hierarchical Diff-EDF service meets the flow miss rate requirements through the combination of single step hierarchal scheduling for the different network flows, and the admission control mechanism that detects the overload conditions to modify packetspsila priorities. The implementation of this scheduler is based on the multi-agent simulation that takes the inspiration from object-oriented programming. The implementation itself is aimed to the construction of a set of elements which, when fully elaborated, define an agent system specification. When evaluating our proposed scheduler, it was extremely obvious that the Hierarchical Diff-EDF scheduler performs much better than both EDF and Diff-EDF schedulers. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Decoupled sliding-mode with fuzzy neural network controller for EHSS velocity control

    Page(s): 7 - 11
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1406 KB) |  | HTML iconHTML  

    In this paper a decoupled sliding-mode with fuzzy neural network controller for a nonlinear system is presented. To divided into two subsystems to achieve asymptotic stability by decoupled method for a class of three order nonlinear system. The fuzzy neural network (FNN) is the main regulator controller, which is used to approximate an ideal computational controller. The compensation controller is designed to compensate for the difference between the ideal computational controller and the FNN controller. A tuning methodology is derived to update weight parts of the FNN. Using Lyapunov law, we derive the decoupled sliding-mode control law and the related parameters adaptive law of FNN. The method can control one-input and multi-output nonlinear systems efficiently. Using this approach, the response of system will converge faster than that of previous reports. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive output feedback control of a class of underactuated systems using neural networks

    Page(s): 12 - 17
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1459 KB) |  | HTML iconHTML  

    This paper focus on adaptive output feedback control of a class of underactuated systems using neural networks. Through Lyapunov-like stability analysis, adaptive laws are obtained to drive neural networkpsilas free parameter adjustment and at the same time assure uniformly ultimate boundedness of the error signals. The approach permits to enhance the performance of an available linear controller by adding a neural network adaptive element which partially cancels the nonlinear modeling error and that does not lead to loss of stability. Simulation results, using a simplified mathematical model of a three degrees-of-freedom model helicopter, validate the proposed methodology. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A multi-class heartbeat classifier employing hybrid fuzzy -neural network

    Page(s): 18 - 21
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1536 KB) |  | HTML iconHTML  

    A major concern of the medical faculty is the onset of the rapidly increasing heart disease cases. Prevention of the heart diseases is one of the most trusted methods in curbing this problem. Electrocardiogram (ECG) diagnosis has proved to be a very effective means of studying the condition of the heart and thus its present state in addition to being inexpensive and non-invasive technique. A major problem of long term or ambulatory ECG is that large number of heartbeats is recorded and manually studying them and thus classifying them as belonging to certain cardiac problems is a time consuming task also prone to human errors. A still grievous situation exists where cardiac experts are not available easily, especially in remote areas (a typical scenario in developed and under-developed countries). In this paper the authors have proposed a novel strategy for automatic heartbeat classification to palliate the above mentioned problems. Ten types of heartbeats considered for automatic classification are atrial premature contraction (APC), fusion(F), left bundle branch block type I and type II (LBBBB I& LBBBB II), normal(N), paced(P), right bundle branch block type i and type II (RBBBB I & RBBBB II) , premature ventricular contraction type I and type II ( PVC I & PVC II). Fuzzy c-means clustering (FCM) is employed for feature extraction of the individual ECG cycles and these extracted features are then used for training multilayer perceptron. A detailed study has been undertaken to find the optimum number of clusters and optimal MLP configuration with the metric of overall percentage classification accuracy. The best FCM-MLP topology exhibited overall classification accuracy of 98.25%. This network was tested for performance in presence of additive white Gaussian noise and was found to be very robust. For comparison, a well-known method of principal component analysis (PCA) was also experimented with. FCM-MLP performs better than PCA-MLP in classifying the - - correct heartbeats. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extracting input features and fuzzy rules for detecting ECG arrhythmia based on NEWFM

    Page(s): 22 - 25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1801 KB) |  | HTML iconHTML  

    Fuzzy neural networks have been successfully applied to generate predictive rules for medical or diagnostic data. This paper presents an approach to automatically detect ECG arrhythmias using the neural network with weighted fuzzy membership functions (NEWFM). NEWFM classifies normal and abnormal beats by the trained bounded sum of weighted fuzzy membership functions (BSWFMs) using normalized features in the range of [0, 1] from UCI repository of machine learning. The generalized 4 features, locally related to the time signal, are extracted by the non-overlap area measurement method. The total numbers of samples are 452 data. The 80% of the data are used for training and 20% for testing. The result of accuracy rate is 81.32%. The BSWFMs of the 4 features trained by NEWFM are shown visually, which makes the features interpret explicitly. Since each BSWFM combines multiple weighted fuzzy membership functions into one using bounded sum, the four small-sized BSWFMs can realize real-time ECG arrhythmias detection in mobile environment. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intelligent mininig for capturing processes through event logs to represent workflows using FP tree

    Page(s): 26 - 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1565 KB)  

    Data mining applications require an ability to understand unfiltered data embedded in event logs. The scalability of the data, end-user comprehensibility of the results, non-presumption of any canonical data distribution, and insensitivity to the order of input records will determine efficiency of data mining. Contemporary workflow management systems are driven by explicit process models based on completely specified workflow designs. Creating a workflow design is a complicated time-consuming process and typically there are discrepancies between the actual workflow processes and the processes as perceived by the management. In this paper, we propose a Process Mining Architecture (PROARCH) model which involves capturing processes in a system through event logs containing information about the different processes under execution. We assume that events in logs bear timestamps. But these logs will also contain log of unformatted data which may be dirty data for our model. Hence this information needs to be filtered before further processing. After filtering, the clean data is represented in MXML format and will serve as input to our model. This MXML data is parsed into a Petri net representation. The nodes and transitions, are connected to form a workflow representation. Since the initial input logs are dirty we use FP tree approach to build our workflow model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Ant Colony System algorithm for path planning in sparse graphs

    Page(s): 31 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2282 KB) |  | HTML iconHTML  

    The general problem of path planning can be modeled as a travelling salesman problem which assumes a graph is fully connected. Full connectivity is however not realistic in many practical path planning problems. The graphs are typically sparse graphs such as for Unmanned Reconnaissance Aerial Vehicles (URAV). This paper describes an Ant Colony System algorithm proposed for path planning in sparse graphs. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intelligent credit risk evaluation system using evolutionary-neuro-fuzzy scheme

    Page(s): 37 - 42
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1355 KB) |  | HTML iconHTML  

    Building an accurate credit scoring model is very important to predict effectively the creditworthiness of new customers. Neural networks and genetic algorithm are suitable for building highly predictive credit scoring model, but the lack of transparency of these methods is a major drawback. On the other hand the main advantage of fuzzy models is their ability to describe the behavior of systems with a series of linguistic humanly understandable rules. In this paper we develop an accurate as well as transparent credit scoring model based on the evolutionary-neuro-fuzzy method. Two datasets from the UCI machine learning repository are selected to evaluate the proposed method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improved Dynamic Ant Colony System (DACS) on symmetric Traveling Salesman Problem (TSP)

    Page(s): 43 - 48
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1338 KB) |  | HTML iconHTML  

    Ants are a fascinating creature that demonstrates a capability of finding food and bring it back to their nest. Their ability as a colony to find paths or routes to the food sources has inspired the development of an algorithm namely ant colony system (ACS). The principle of cooperation has been the backbone in these algorithmic developments. However, observing the behavior of a single ant can provide an added value to the principle. Ants communicate to each other through a chemical substance called pheromone. Manipulating and empowering this substance is the trivial factor in finding the best solution. However, without considering the experiences of individuals would contribute a complete waste of available knowledge. Having the concepts of a single ant trying to reconstruct or reconnect the paths that was previously laid by its colony when a certain obstacle placed on its normal paths has added another level of pheromone updates. Thus, this new level of pheromone updates which manipulating and empowering the searching experiences of individual ants can improve the current ACS algorithm. Traveling salesman problem (TSP) was used as a case study to show the capability of the algorithm in order to find the best solution in terms of the shortest distance. At the end of this paper, we presented an experimental result on a benchmark data to show how it could improve the fundamental of ACS algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mining sequence of event from vector-based spatiotemporal database

    Page(s): 49 - 54
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (705 KB) |  | HTML iconHTML  

    In this paper, we propose VectoFlowMiner, an extension to FlowMiner, a spatiotemporal data mining algorithm. FlowMiner, are targeted for mining from raster-based spatial data. Most GIS (geographic information system) however, requires, process, and stores spatial data in vector representation. We propose Neighborhood graph as dasiadrop-inpsila replacement for relationship operator in FlowMiner. We present here the result from experimenting with real life dataset consisting of road traffic data. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hybrid method for optimization (discrete PSO + CLA)

    Page(s): 55 - 60
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1715 KB) |  | HTML iconHTML  

    PSO is an evolutionary algorithm that is inspired from collective behavior of animals such as fish schooling or bird flocking. One of the drawbacks of this model is premature convergence and trapping in local optima. In this paper we propose a solution to this problem in discrete version of PSO that uses Learning Automata and introduce a cellular learning automata (CLA) based discrete PSO. Experimental results on five optimization problems show the superiority of the proposed algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonlinear model predictive control based on lexicographic multi-objective genetic algorithm

    Page(s): 61 - 65
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1143 KB) |  | HTML iconHTML  

    By using a series of dynamic coefficients in fitness function, a modified genetic algorithm is proposed. It can solve the lexicographic multi-objective optimization problem stemmed from multivariable nonlinear model predictive control directly. A control problem of a two-tank control system is then given as an example. Stair-like control strategy and feedback compensation are also used to develop a better performance of the controller. Simulation results verify the efficiency of the algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Forecasting KOSPI by weighted average defuzzification based on NEWFM

    Page(s): 66 - 70
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2532 KB) |  | HTML iconHTML  

    Fuzzy neural networks have been successfully applied to generate predictive rules for stocks forecasting. This paper presents a methodology for forecasting the daily Korea composite stock price index (KOSPI) based on the neural network with weighted fuzzy membership functions (NEWFM) and time series of KOSPI based on the defuzzyfication of weighted average method (The fuzzy model suggested by Takagi and Sugeno in 1985). NEWFM is a new model of neural networks to improve forecasting accuracy by using self adaptive weighted fuzzy membership functions. The degree of classification intensity is obtained by bounded sum of weighted fuzzy membership functions extracted by NEWFM, and then weighted average defuzzification is used for forecasting KOSPI. In this paper, the Haar wavelet function is used as a mother wavelet. A set of five extracted coefficient features of the Haar WT are presented to forecast KOSPI. The total number of samples is 2928 trading days, from January 1989 to December 1998. About 80% of the data is used for training and 20% for testing. The result of classification rate is 58.0034%. The implementation of the NEWFM demonstrates an excellent capability in the field of stocks forecasting. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using genetic algorithm to evolves algebraic rule-based classifiers for NPC prognosis

    Page(s): 71 - 74
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1553 KB) |  | HTML iconHTML  

    Over the last twenty years, soft computing has developed rapidly as a discipline and method for the diagnosis and prognosis in medical informatics. In this paper we try to introduce a new method for the prognosis in the cancer subdomain of nasopharyngeal carcinoma. An overview of the new techniques is presented, and instances of these in NPC cancer research are discussed. An insight into the plausible makeup of NPC data, for current and future systems is presented. The absence of any previous occurrence of the use of genetic algorithms influences the choice as an alternative means for determining NPC prognosis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Solving quadratic assignment problems by a tabu based simulated annealing algorithm

    Page(s): 75 - 80
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1455 KB) |  | HTML iconHTML  

    The quadratic assignment problem (QAP) is a hard and classical combinatorial optimization problem. In this paper a hybrid algorithm that combines the simulated annealing (SA) and tabu search (TS) is proposed to solve the QAP. The search of simulated annealing may stuck at a local optimum due to the low acceptable moves, particularly as the barrier is high and the temperature is low. A guided restart strategy is incorporated into SA to escape from a local optimum and re-annealing from a promising point more efficiently. The tabu list, as a short-term memory, is used in the move generation procedure to prohibit the highly frequent moves and diversify the search. Performance evaluation has been carried out on several problem instances in the QAPLIB. Experimental results indicate that these two strategies significantly improve the performance of simulated annealing algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural network multi layer perceptron modeling of surface quality in laser machining

    Page(s): 81 - 86
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3166 KB) |  | HTML iconHTML  

    Uncertainty is inevitable in problem solving and decision making. One way to reduce it is by seeking the advice of an expert in related field. On the other hand, when we use computers to reduce uncertainty, the computer itself can become an expert in a specific field through a variety of methods. One such method is machine learning, which involves using a computer algorithm to capture hidden knowledge from data. The researchers conducted the prediction of laser machining quality, namely surface roughness with seven significant parameters to obtain singleton output using machine learning techniques based on Quick Back Propagation Algorithm. In this research, we investigated a problem solving scenario for a metal cutting industry which faces some problems in determining the end product quality of Manganese Molybdenum (Mn-Mo) pressure vessel plates. We considered several real life machining scenarios with some expert knowledge input and machine technology features. The input variables are the design parameters which have been selected after a critical parametric investigation of 14 process parameters available on the machine. The elimination of non-significant parameters out of 14 total parameters were carried out by single factor and interaction factor investigation through design of experiment (DOE) analysis. Total number of 128 experiments was conducted based on 2k factorial design. This large search space poses a challenge for both human experts and machine learning algorithms in achieving the objectives of the industry to reduce the cost of manufacturing by enabling the off hand prediction of laser cut quality and further increase the production rate and quality. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using learning automata for tuning fuzzy membership functions in learning driver preferences

    Page(s): 87 - 92
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1612 KB) |  | HTML iconHTML  

    With the growth of car navigation systems technology come a variety of enhancements aiming to increase user comfort and satisfaction. One such application is the appearance of methods for learning a driverpsilas preferences in making a choice between several routes. A driver may know his/her basic and most important factors in making such a decision, but may have these factors weighing in differently. Hence, machine learning methods can be applied to model the driverpsilas preferences, thus predicting the result of the decision process. This paper proposes a new method which combines a fuzzy expert system approach with learning automata. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • PLC-based fuzzy logic controller for induction-motor drive with constant V/Hz ratio

    Page(s): 93 - 98
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2745 KB) |  | HTML iconHTML  

    This paper presents the design and implementation of a PLC-based fuzzy logic controller for an induction motor speed control at constant V/Hz ratio. The PLC has arithmetic and logic operations instructions set that was utilized in the implementation of the fuzzy control of induction motor speed control. Fuzzy logic algorithm applies rules obtained from human expert of a system. The input signals to fuzzy controller are the linguistic variable of speed error and change of speed error, while the output signal is frequency of PWM inverter. The supply voltage to the induction motor is ascertained from frequency output requirement using constant V/Hz ratio. The objective of the controller is to provide stability in responds to disturbance and sudden changes in reference speed. Results of the proposed system show satisfactory performance. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Anti-overshoot control of model helicopter’s yaw angle with combination of fuzzy controller and fuzzy brake

    Page(s): 99 - 103
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1428 KB) |  | HTML iconHTML  

    In this research, the fuzzy control of the yaw angle of a model helicopter is studied, particularly, in order to reduce the overshoot which can be a serious problem in high inertia systems. Initially, a Sugeno-type controller is designed. This controller provides quick convergence and keeps the control input in a permitted range .Moreover, a good stability is offered by this fuzzy controller. But, a significant and repeating overshoot is observed in controlled system behaviour that is not desirable. In order to solve this problem and improve the control system, another fuzzy inference system, namely ldquofuzzy brakerdquo, is added to the closed loop circuit. Fuzzy brakepsilas task is to reduce the control input when the error is low. The proposed Sugeno-type fuzzy controller with brake (SFCB) not only vanishes the overshoot practically but also causes a considerable reduction in energy consumption, at the same time, SFCB improves the performance. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Interactive Evolutionary Computation and density-based clustering for data analysis

    Page(s): 104 - 108
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1834 KB) |  | HTML iconHTML  

    Data clustering is useful in solving many pattern recognition and decision support tasks. This work has empirically demonstrated the effectiveness of a hybrid neural network model for density-based clustering. The cluster regions formed were then evaluated based on visualisation of clustering information on the map. The visual inspection of the map revealed the number of clusters as well as their spatial relationships. By analysing the clustering information in this way, the cluster (or density) structures of the data were obtained. In this paper, a case study of pen-based handwritten digits recognition was chosen to demonstrate how, in this by using the interactive evolutionary computational (IEC), both the computer system and the user work together in the cluster analysis process and subsequently, shown that this approach is suitable for exploratory data analysis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An off-line fuzzy backstepping controller for rotary inverted pendulum system

    Page(s): 109 - 113
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (935 KB) |  | HTML iconHTML  

    In this study a new combination of nonlinear backstepping scheme with off-line fuzzy system is presented for controlling a rotary inverted pendulum system to achieve better performance in nonlinear controller. The inverted pendulum, a popular mechatronic application, exists in many different forms. The common thread among these systems is their goal: to balance a link on end using feedback control. The purpose of this study is to design a stabilizing controller that balances the inverted pendulum in the up-right position. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Development of multilayer neural network for real time temperature control system

    Page(s): 114 - 119
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1716 KB) |  | HTML iconHTML  

    Since last two decade a substantial amount of research has been reported in the field of identification and control of nonlinear dynamical systems using artificial neural networks. The work presented in this paper comprises the study of the development of multilayer neural network for real time temperature control system. Inverse dynamic modelling approach is used as a basic control strategy. The primary aim is to construct a network that allows for modeling of a global dynamics of a system. The study devoted to efficient identification and control design methods using neural network based nonlinear control for water bath temperature, and to implement them in a real world environment. The performance of the control systems was verified by simulations and with real-time experiments using pilot process. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modeling heat exchanger using neural networks

    Page(s): 120 - 124
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1655 KB) |  | HTML iconHTML  

    Tools to predict the effects caused by frequent changes in the feedstock and in the operating condition in crude preheat train (CPT) in a refinery are essential to maintain optimal operating conditions in the heat exchanger. Currently, no such tools are used in industries. In this paper, an approach based on nonlinear auto regressive with exogenous input (NARX) type multi layer perceptron neural network model is proposed. This model serves as the prediction tool in order to determine the optimal operating conditions. The neural network model was developed using data collected from CPT in a refinery. In addition to the data on flow rates and temperatures of the streams in the heat exchanger, data on physico-chemical properties and crude blend were also included as input variables to the model. It was observed that the root mean square error (RMSE) during training and validation phases are less than 0.3degC proving that the modeling approach employed in this research is suitable to capture the complex and nonlinear characteristics of the heat exchanger. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Frontal obstacle avoidance of an autonomous subsurface vehicle (ASV) using fuzzy logic method

    Page(s): 125 - 128
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1922 KB) |  | HTML iconHTML  

    Utilization of robots and autonomous vehicles to replace human being in executing remote, difficult, highly risk and dangerous tasks is the niche issue among robotic engineers. In order to develop such application, the main challenge is to integrate the advances of various engineering disciplines. This project proposes the design of frontal obstacle avoidance for ASV using fuzzy logic. This project also compares analyses done between Mamdani and Sugeno methods to control the maneuvering of ASV. This project uses MATLAB as the main designing tool. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Missing data estimation on heart disease using Artificial Neural Network and Rough Set Theory

    Page(s): 129 - 133
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1245 KB) |  | HTML iconHTML  

    The objective of this research is to implement a method for estimating the real missing data in heart disease datasets and to show how it affects the resulting knowledge. Missing data is common problem in knowledge discovery from database (KDD) processes that can lead significant error in extracted knowledge. We use hybridization of artificial neural network and rough set theory (ANNRST) to estimate the real missing data on heart disease from UCI (University of California, Irvine) datasets. ANN with reduced input features is used to estimate the missing data. RST is used to reduce the dimensionality of input features and to extract the knowledge as reducts and rules from heart disease datasets with estimated missing data. RST, decomposition tree, local transfer function classifier (LTF-C) and k-nearest neighbor (k-NN) classifier are used to calculate the accuracy. Comparative study with k-NN estimation, most common attribute value filling and deletion of missing data are made to evaluate the extracted knowledge. ANNRST can be considered as the appropriate estimation method when strong relationship between original complete datasets and estimated datasets is important (the estimated datasets really represent the nature of original complete datasets) as it gives the best accuracy and coverage for almost all the classifiers. View full abstract»

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