Showing papers in "Expert Systems With Applications in 2006"

A GA-based feature selection and parameters optimizationfor support vector machines

Abstract: Support Vector Machines, one of the new techniques for pattern classification, have been widely used in many application areas. The kernel parameters setting for SVM in a training process impacts on the classification accuracy. Feature selection is another factor that impacts classification accuracy. The objective of this research is to simultaneously optimize the parameters and feature subset without degrading the SVM classification accuracy. We present a genetic algorithm approach for feature selection and parameters optimization to solve this kind of problem. We tried several real-world datasets using the proposed GA-based approach and the Grid algorithm, a traditional method of performing parameters searching. Compared with the Grid algorithm, our proposed GA-based approach significantly improves the classification accuracy and has fewer input features for support vector machines. q 2005 Elsevier Ltd. All rights reserved.

Fuzzy TOPSIS method based on alpha level sets with an application to bridge risk assessment

Abstract: This paper proposes a fuzzy TOPSIS method based on alpha level sets and presents a nonlinear programming (NLP) solution procedure. The relationship between the fuzzy TOPSIS method and fuzzy weighted average (FWA) is also discussed. Three numerical examples including an application to bridge risk assessment are investigated using the proposed fuzzy TOPSIS method to illustrate its applications and the differences from the other procedures. It is shown that the proposed fuzzy TOPSIS method performs better than the other fuzzy versions of the TOPSIS method.

Applying data mining to telecom churn management

Abstract: Taiwan deregulated its wireless telecommunication services in 1997. Fierce competition followed, and churn management becomes a major focus of mobile operators to retain subscribers via satisfying their needs under resource constraints. One of the challenges is churner prediction. Through empirical evaluation, this study compares various data mining techniques that can assign a ‘propensity-to-churn’ score periodically to each subscriber of a mobile operator. The results indicate that both decision tree and neural network techniques can deliver accurate churn prediction models by using customer demographics, billing information, contract/service status, call detail records, and service change log.

Hybrid genetic algorithms and support vector machines for bankruptcy prediction

Abstract: Bankruptcy prediction is an important and widely studied topic since it can have significant impact on bank lending decisions and profitability. Recently, the support vector machine (SVM) has been applied to the problem of bankruptcy prediction. The SVM-based method has been compared with other methods such as the neural network (NN) and logistic regression, and has shown good results. The genetic algorithm (GA) has been increasingly applied in conjunction with other AI techniques such as NN and Case-based reasoning (CBR). However, few studies have dealt with the integration of GA and SVM, though there is a great potential for useful applications in this area. This study proposes methods for improving SVM performance in two aspects: feature subset selection and parameter optimization. GA is used to optimize both a feature subset and parameters of SVM simultaneously for bankruptcy prediction.

Customer segmentation and strategy development based on customer lifetime value: A case study

Abstract: The more a marketing paradigm evolves, the more long-term relationship with customers gains its importance. CRM, a recent marketing paradigm, pursues long-term relationship with profitable customers. It can be a starting point of relationship management to understand and measure the true value of customers since marketing management as a whole is to be deployed toward the targeted customers and profitable customers, to foster customers' full profit potential. Corporate success depends on an organization's ability to build and maintain loyal and valued customer relationships. Therefore, it is essential to build refined strategies for customers based on their value. In this paper, we propose a framework for analyzing customer value and segmenting customers based on their value. After segmenting customers based on their value, strategies building according to customer segment will be illustrated through a case study on a wireless telecommunication company.

Using DEA-neural network approach to evaluate branch efficiency of a large Canadian bank

Abstract: In today's economy and society, the banking industry is of great importance to every one of us. We all depend on the efficiency and quality of services that the banking industry provides. With the improvement in technology, the competition in the banking industry has become increasingly intense. Therefore, performance analyses in the banking industry attract more and more attention. This paper integrates data envelopment analysis (DEA) and neural networks (NNs) to examine the relative branch efficiency of a big Canadian bank. The results are compared with the normal DEA results. On the whole they are comparable. Furthermore, the guidance on how to improve the branch performance is given. Neural networks are also applied to do short-term efficiency prediction. Finally, the comparison between these two approaches is presented.

Predicting box-office success of motion pictures with neural networks

Abstract: Predicting box-office receipts of a particular motion picture has intrigued many scholars and industry leaders as a difficult and challenging problem. In this study, the use of neural networks in predicting the financial performance of a movie at the box-office before its theatrical release is explored. In our model, the forecasting problem is converted into a classification problem-rather than forecasting the point estimate of box-office receipts, a movie based on its box-office receipts in one of nine categories is classified, ranging from a 'flop' to a 'blockbuster.' Because our model is designed to predict the expected revenue range of a movie before its theatrical release, it can be used as a powerful decision aid by studios, distributors, and exhibitors. Our prediction results is presented using two performance measures: average percent success rate of classifying a movie's success exactly, or within one class of its actual performance. Comparison of our neural network to models proposed in the recent literature as well as other statistical techniques using a 10-fold cross validation methodology shows that the neural networks do a much better job of predicting in this setting.

A multilayer perceptron-based medical decision support system for heart disease diagnosis

Abstract: The medical diagnosis by nature is a complex and fuzzy cognitive process, and soft computing methods, such as neural networks, have shown great potential to be applied in the development of medical decision support systems (MDSS). In this paper, a multiplayer perceptron-based decision support system is developed to support the diagnosis of heart diseases. The input layer of the system includes 40 input variables, categorized into four groups and then encoded using the proposed coding schemes. The number of nodes in the hidden layer is determined through a cascade learning process. Each of the 5 nodes in the output layer corresponds to one heart disease of interest. In the system, the missing data of a patient are handled using the substituting mean method. Furthermore, an improved back propagation algorithm is used to train the system. A total of 352 medical records collected from the patients suffering from five heart diseases have been used to train and test the system. In particular, three assessment methods, cross validation, holdout and bootstrapping, are applied to assess the generalization of the system. The results show that the proposed MLP-based decision support system can achieve very high diagnosis accuracy (>90%) and comparably small intervals (<5%), proving its usefulness in support of clinic decision process of heart diseases.

Improved route planning and scheduling of waste collection and transport

Abstract: The collection of waste is a highly visible and important municipal service that involves large expenditures. Waste collection problems are, however, one of the most difficult operational problems to solve. This paper describes the optimization of vehicle routes and schedules for collecting municipal solid waste in Eastern Finland. The solutions are generated by a recently developed guided variable neighborhood thresholding metaheuristic that is adapted to solve real-life waste collection problems. Several implementation approaches to speed up the method and cut down the memory usage are discussed. A case study on the waste collection in two regions of Eastern Finland demonstrates that significant cost reductions can be obtained compared with the current practice.

A process-mining framework for the detection of healthcare fraud and abuse

Abstract: People rely on government-managed health insurance systems, private health insurance systems, or both to share the expensive healthcare costs. With such an intensive need for health insurances, however, health care service providers' fraudulent and abusive behavior has become a serious problem. In this research, we propose a data-mining framework that utilizes the concept of clinical pathways to facilitate automatic and systematic construction of an adaptable and extensible detection model. The proposed approaches have been evaluated objectively by a real-world data set gathered from the National Health Insurance (NHI) program in Taiwan. The empirical experiments show that our detection model is efficient and capable of identifying some fraudulent and abusive cases that are not detected by a manually constructed detection model.

An effective refinement strategy for KNN text classifier

Abstract: Due to the exponential growth of documents on the Internet and the emergent need to organize them, the automated categorization of documents into predefined labels has received an ever-increased attention in the recent years. A wide range of supervised learning algorithms has been introduced to deal with text classification. Among all these classifiers, K-Nearest Neighbors (KNN) is a widely used classifier in text categorization community because of its simplicity and efficiency. However, KNN still suffers from inductive biases or model misfits that result from its assumptions, such as the presumption that training data are evenly distributed among all categories. In this paper, we propose a new refinement strategy, which we called as DragPushing, for the KNN Classifier. The experiments on three benchmark evaluation collections show that DragPushing achieved a significant improvement on the performance of the KNN Classifier.

A Kansei mining system for affective design

Abstract: Affective design has received much attention from both academia and industries. It aims at incorporating customers' affective needs into design elements that deliver customers' affective satisfaction. The main challenge for affective design originates from difficulties in mapping customers' subjective impressions, namely Kansei, to perceptual design elements. This paper intends to develop an explicit decision support to improve the Kansei mapping process by reusing knowledge from past sales records and product specifications. As one of the important applications of data mining, association rule mining lends itself to the discovery of useful patterns associated with the mapping of affective needs. A Kansei mining system is developed to utilize valuable affect information latent in customers' impressions of existing affective designs. The goodness of association rules is evaluated according to their achievements of customers' expectations. Conjoint analysis is applied to measure the expected and achieved utilities of a Kansei mapping relationship. Based on goodness evaluation, mapping rules are further refined to empower the system with useful inference patterns. The system architecture and implementation issues are discussed in detail. An application of Kansei mining to mobile phone affective design is presented.

Artificial neural networks with evolutionary instance selection for financial forecasting

Abstract: In this paper, I propose a genetic algorithm (GA) approach to instance selection in artificial neural networks (ANNs) for financial data mining. ANN has preeminent learning ability, but often exhibit inconsistent and unpredictable performance for noisy data. In addition, it may not be possible to train ANN or the training task cannot be effectively carried out without data reduction when the amount of data is so large. In this paper, the GA optimizes simultaneously the connection weights between layers and a selection task for relevant instances. The globally evolved weights mitigate the well-known limitations of gradient descent algorithm. In addition, genetically selected instances shorten the learning time and enhance prediction performance. This study applies the proposed model to stock market analysis. Experimental results show that the GA approach is a promising method for instance selection in ANN.

Fuzzy Delphi and back-propagation model for sales forecasting in PCB industry

Abstract: Reliable prediction of sales can improve the quality of business strategy. In this research, fuzzy logic and artificial neural network are integrated into the fuzzy back-propagation network (FBPN) for sales forecasting in Printed Circuit Board (PCB) industry. The fuzzy back propagation network is constructed to incorporate production-control expert judgments in enhancing the model's performance. Parameters chosen as inputs to the FBPN are no longer considered as of equal importance, but some sales managers and production control experts are requested to express their opinions about the importance of each input parameter in predicting the sales with linguistic terms, which can be converted into pre-specified fuzzy numbers. The proposed system is evaluated through the real world data provided by a printed circuit board company and experimental results indicate that the Fuzzy back-propagation approach outperforms other three different forecasting models in MAPE measures.

An automated FX trading system using adaptive reinforcement learning

Abstract: This paper introduces adaptive reinforcement learning (ARL) as the basis for a fully automated trading system application. The system is designed to trade foreign exchange (FX) markets and relies on a layered structure consisting of a machine learning algorithm, a risk management overlay and a dynamic utility optimization layer. An existing machine-learning method called recurrent reinforcement learning (RRL) was chosen as the underlying algorithm for ARL. One of the strengths of our approach is that the dynamic optimization layer makes a fixed choice of model tuning parameters unnecessary. It also allows for a risk-return trade-off to be made by the user within the system. The trading system is able to make consistent gains out-of-sample while avoiding large draw-downs.

Automatic detection of epileptic seizure using dynamic fuzzy neural networks

Abstract: In this study, a new approach based on neural network and fuzzy logic technologies was presented for detection of epileptic seizure to allow for the incorporation of both heuristics and deep knowledge to exploit the best characteristics of each. A dynamic fuzzy neural network (DFNN) that contains dynamical elements in their processing units is used in the classification of EEG signals. The detection of epileptiform discharges in the EEG is an important component in the diagnosis of epilepsy. EEG signals were decomposed into the frequency sub-bands using discrete wavelet transform (DWT). Then these sub-band frequencies were used as an input to a DFNN with two discrete outputs: normal and epileptic. Some conclusions concerning the impacts of features on epileptic seizure detection was obtained through analysis of the DFNN. The performance of the DFNN model was evaluated in terms of classification accuracies and the results confirmed that the proposed DFNN classifiers have some potential in detecting epileptic seizures. The DFNN model achieved accuracy rates, which were higher than that of neural network model.

Dynamic shopfloor scheduling in multi-agent manufacturing systems

Abstract: This paper is on the development of an agent-based approach for the dynamic integration of the process planning and scheduling functions. In consideration of the alternative processes and alternative machines for the production of each part, the actual selection of the schedule and allocation of manufacturing resources is achieved through negotiation among the part and machine agents which represent the parts and manufacturing resources, respectively. The agents are to negotiate on a fictitious cost with the adoption of a currency function. Two MAS architectures are evaluated in this paper. One is a simple MAS architecture comprises part agents and machine agents only; the other one involves the addition of a supervisor agent to establish a hybrid-based MAS architecture. A hybrid contract net protocol is developed in the paper to support both types of MAS architectures. This new negotiation protocol enables multi-task many-to-many negotiations, it also incorporates global control into the decentralized negotiation. Simulation runs are conducted to compare the performance of the proposed MAS-based IPPS approaches and that of an evolutionary algorithm. It also shows that the hybrid-based MAS, with the introduction of supervisory control, is able to provide integrated process plan and job shop scheduling solutions with a better global performance.

Web page classification based on a support vector machine using a weighted vote schema

Abstract: Traditional information retrieval method use keywords occurring in documents to determine the class of the documents, but usually retrieves unrelated web pages. In order to effectively classify web pages solving the synonymous keyword problem, we propose a web page classification based on support vector machine using a weighted vote schema for various features. The system uses both latent semantic analysis and web page feature selection training and recognition by the SVM model. Latent semantic analysis is used to find the semantic relations between keywords, and between documents. The latent semantic analysis method projects terms and a document into a vector space to find latent information in the document. At the same time, we also extract text features from web page content. Through text features, web pages are classified into a suitable category. These two features are sent to the SVM for training and testing respectively. Based on the output of the SVM, a voting schema is used to determine the category of the web page. Experimental results indicate our method is more effective than traditional methods.

Failure prediction in the Russian bank sector with logit and trait recognition models

Abstract: The Russian banking sector experienced considerable turmoil in the late 1990s, especially around the Russian banking crisis in 1998. The question is what types of banks are vulnerable to shocks and whether or not bank-specific characteristics can be used to predict vulnerability to failures. In this study we employ a parametric logit model and a nonparametric trait recognition approach to predict failures among Russian commercial banks. We modify the trait recognition approach such that the default probabilities are calculated directly without preliminary classification of cells in the voting matrix as safe or unsafe. We test the predictive power of the models based on their prediction accuracy using holdout samples. All models performed better than the benchmark; the modified trait recognition approach outperformed logit and the traditional trait recognition approach in both the original and the holdout samples. As expected liquidity plays an important role in bank failure prediction, but also asset quality and capital adequacy turn out to be important determinants of failure.

The evaluation of consumer loans using support vector machines

Abstract: The commencement of the Basel II requirement, popularization of consumer loans and the intense competition in financial market has increased the awareness of the critical delinquency issue for financial institutions in granting loans to potential applicants. In the past few decades, the scheme of artificial neural networks has been successfully applied to the financial field. Recently, the Support Vector Machine (SVM) has emerged as the better neural network in dealing with classification and forecasting problems due to its superior features of generalization performance and global optimum. This study develops a loan evaluation model using SVM to identify potential applicants for consumer loans. In addition to conducting experiments on performance comparison via cross-validation and paired t test, we analyze misclassification errors in terms of Type I and Type II and their effect on selecting network parameters of SVM. The analysis findings facilitate the development of a useful visual decision-support tool. The experimental results using a real-world data set reveal that SVM surpasses traditional neural network models in generalization performance and visualization via the visual tool, which helps decision makers determine appropriate loan evaluation strategies.

Development of a knowledge management initiative and system: A case study

Abstract: As knowledge emerges as the primary strategic resource in the 21st century, many firms in the manufacturing and service sectors alike are beginning to introduce and implement Knowledge Management (KM). Organisations can certainly benefit from its application for enhanced decision support, efficiency and innovation, thus helping them to realise their strategic mission. However, KM is an emerging paradigm, and not many organisations have a clear idea of how to proceed with it. This paper presents the results of a case study conducted in one company in the United Kingdom (UK), the major aim being to identify how it has developed a KM initiative and system. Hopefully, the information extracted from this study will be beneficial to other organisations that are attempting to implement KM or to those that are in the throes of adopting it.

A cardiac sound characteristic waveform method for in-home heart disorder monitoring with electric stethoscope

Abstract: An analytical model based on a single-DOF is proposed for extracting the characteristic waveforms (CSCW) from the cardiac sounds recorded by an electric stethoscope. Also, the diagnostic parameters [T1, T2, T11, T12], the time intervals between the crossed points of the CSCW and an adaptive threshold line (THV), were verified useful for identification of heart disorders. The easy-understanding graphical representation of the parameters was considered, in advance, even for an inexperienced user able to monitor his or her pathology progress. Since the diagnostic parameters were influenced much by a THV, the FCM clustering algorithm was introduced for determination of an adaptive THV in order to extract reliable diagnostic parameters. Further, the minimized J m and [ v 1 , v 2 , v 3 , v 4 ] could be also efficient indicators for identifying the heart disorders. Finally, a case study on the abnormal/normal cardiac sounds is demonstrated to validate the usefulness and efficiency of the cardiac sound characteristic waveform method with FCM clustering algorithm. NM1 and NM2 as the normal case have very small value in J m ( v 1 , v 2 , v 3 , v 4 ] are about [0.1, 0.1, 0.8, 0.4]. For abnormal cases, in case of AR, its J m is very small and the values of [ v 1 , v 3 , v 4 ] are very high comparing to the normal cases. However, in cases of AF and MS have very big values in J m (>0.38).

Personalized curriculum sequencing utilizing modified item response theory for web-based instruction

Abstract: Curriculum sequencing is an important research issue for Web-based instruction systems because no fixed learning pathway will be appropriate for all learners. Therefore, many researchers focused on developing e-learning systems with personalized learning mechanism to assist on-line Web-based learning and adaptively provide learning pathways. However, although most personalized systems consider learner preferences, interests and browsing behavior in providing personalized curriculum sequencing services, these systems usually neglect to consider whether learner ability and the difficulty level of the recommended courseware are matched to each other or not. Generally, inappropriate courseware leads to learner cognitive overload or disorientation during learning, thus reducing learning effect. Besides, the problem of concept continuity of learning pathways also needs to be considered while implementing personalized curriculum sequencing. Smoother learning pathways increase learning effect, avoiding unnecessarily difficult concepts. This paper presents a prototype of personalized Web-based instruction system (PWIS) based on the proposed modified Item Response Theory (IRT) to perform personalized curriculum sequencing through simultaneously considering courseware difficulty level, learner's ability and the concept continuity of learning pathways during learning. In the proposed modified IRT, the information function is revised to consider the concept continuity of learning pathway as well as considering the difficulty level of courseware and individual learner ability. Experiment results indicate that applying the proposed modified IRT for Web-based learning can construct suitable learning pathway to learners for personalized learning, and help them to learn more effectively.

A new method to medical diagnosis: Artificial immune recognition system (AIRS) with fuzzy weighted pre-processing and application to ECG arrhythmia

Abstract: Changes in the normal rhythm of a human heart may result in different cardiac arrhythmias, which may be immediately fatal or cause irreparable damage to the heart sustained over long periods of time. The ability to automatically identify arrhythmias from ECG recordings is important for clinical diagnosis and treatment. Artificial immune systems (AISs) is a new but effective branch of artificial intelligence. Among the systems proposed in this field so far, artificial immune recognition system (AIRS), which was proposed by A. Watkins, has showed an effective and intriguing performance on the problems it was applied. Previously, AIRS was applied a range of problems including machine-learning benchmark problems and medical classification problems like breast cancer, diabets, liver disorders classification problems. The conducted medical classification task was performed for ECG arrhythmia data taken from UCI repository of machine-learning. Firsly, ECG dataset is normalized in the range of [0,1] and is weighted with fuzzy weighted pre-processing. Then, weighted input values obtained from fuzzy weighted pre-processing is classified by using AIRS classifier system. In this study, fuzzy weighted pre-processing, which can be improved by ours, is a new method and firstly, it is applied to ECG dataset. Classifier system consists of three stages: 50–50% of traing-test dataset, 70–30% of traing-test dataset and 80–20% of traing-test dataset, subsequently, the obtained classification accuries: 78.79, 75.00 and 80.77%.

An evolutionary approach to the combination of multiple classifiers to predict a stock price index

Abstract: Multiple classifier combination is a technique that combines the decisions of different classifiers. Combination can reduce the variance of estimation errors and improve the overall classification accuracy. However, direct application of combination schemes developed for pattern recognition to solving business problems has some limitations, because business problems cannot be explained completely by the results provided by machine-learning-driven classifiers alone owing to their intrinsic complexity. To solve such problems, this paper proposes an approach that is capable of incorporating the subjective problem-solving knowledge of humans into the results of quantitative models. Genetic algorithms (GAs) are used to combine classifiers stemming from machine learning, experts, and users. We use our GA-based method to predict the Korea stock price index (KOSPI).

A study of Taiwan's issuer credit rating systems using support vector machines

Abstract: By providing credit risk information, credit rating systems benefit most participants in financial markets, including issuers, investors, market regulators and intermediaries. In this paper, we propose an automatic classification model for issuer credit ratings, a type of fundamental credit rating information, by applying the support vector machine (SVM) method. This is a novel classification algorithm that is famous for dealing with high dimension classifications. We also use three new variables: stock market information, financial support by the government, and financial support by major shareholders to enhance the effectiveness of the classification. Previous research has seldom considered these variables. The data period of the input variables used in this study covers three years, while most previous research has only considered one year. We compare our SVM model with the back propagation neural network (BP), a well-known credit rating classification method. Our experiment results show that the SVM classification model performs better than the BP model. The accuracy rate (84.62%) is also higher than previous research.

Subscription fraud prevention in telecommunications using fuzzy rules and neural networks

Abstract: A system to prevent subscription fraud in fixed telecommunications with high impact on long-distance carriers is proposed. The system consists of a classification module and a prediction module. The classification module classifies subscribers according to their previous historical behavior into four different categories: subscription fraudulent, otherwise fraudulent, insolvent and normal. The prediction module allows us to identify potential fraudulent customers at the time of subscription. The classification module was implemented using fuzzy rules. It was applied to a database containing information of over 10,000 real subscribers of a major telecom company in Chile. In this database a subscription fraud prevalence of 2.2% was found. The prediction module was implemented as a multilayer perceptron neural network. It was able to identify 56.2% of the true fraudsters, screening only 3.5% of all the subscribers in the test set. This study shows the feasibility of significantly preventing subscription fraud in telecommunications by analyzing the application information and the customer antecedents at the time of application.

Comparison of four different time series methods to forecast hepatitis A virus infection

Abstract: Hepatitis A virus (HAV) infection is not a problem of only developing countries, but also of developed countries. In this study, we compared time series prediction capabilities of three artificial neural networks (ANN) algorithms (multi-layer perceptron (MLP), radial basis function (RBF), and time delay neural networks (TDNN)), and auto-regressive integrated moving average (ARIMA) model to HAV forecasting. To assess the effectiveness of these methods, we used in forecasting 13 years of time series (January 1992–June 2004) monthly records for HAV data, in Turkey. Results show that MLP is more accurate and performs better than RBF, TDNN and ARIMA model.

Modelling sovereign credit ratings: Neural networks versus ordered probit

Abstract: Sovereign credit ratings are becoming increasingly important both within a financial regulatory context and as a necessary prerequisite for the development of emerging capital markets. Using a comprehensive dataset of rating agencies and countries over the period 1989-1999, this paper demonstrates that artificial neural networks (ANN) represent a superior technology for calibrating and predicting sovereign ratings relative to ordered probit modelling, which has been considered by the previous literature to be the most successful econometric approach. ANN have been applied to classification problems with great success over a wide range of applications where there is an absence of a precise theoretical model to underpin the relationships in the data. The results for sovereign credit ratings presented here corroborate other researchers' findings that ANN are highly effective classifiers.

Development of a patent document classification and search platform using a back-propagation network

Abstract: In order to process large numbers of explicit knowledge documents such as patents in an organized manner, automatic document categorization and search are required. In this paper, we develop a document classification and search methodology based on neural network technology that helps companies manage patent documents more effectively. The classification process begins by extracting key phrases from the document set by means of automatic text processing and determining the significance of key phrases according to their frequency in text. In order to maintain a manageable number of independent key phrases, correlation analysis is applied to compute the similarities between key phrases. Phrases with higher correlations are synthesized into a smaller set of phrases. Finally, the back-propagation network model is adopted as a classifier. The target output identifies a patent document’s category based on a hierarchical classification scheme, in this case, the international patent classification (IPC) standard. The methodology is tested using patents related to the design of power hand-tools. Related patents are automatically classified using pre-trained neural network models. In the prototype system, two modules are used for patent document management. The automatic classification module helps the user classify patent documents and the search module helps users find relevant and related patent documents. The result shows an improvement in document classification and identification over previously published methods of patent document management.