Csite László: Stabil mintákon alapuló értékpapírárolyam-előrejelző rendszer fejlesztése

Kulcsszavak: k-means algoritmus, neurális háló, értékpapír árfolyam előrejelzés, technikai analízis

Az emberiség egyik legősibb vágyai közé tartozik a jövő megismerése, még mielőtt az jelenné válna, különösen ha ezen ismeret által még anyagi előnyhöz is juthatunk. Nem véletlen hát, hogy az értékpapírok árfolyamainak előrejelzése megkülönböztetett figyelemmel bír. Diplomamunkám témája egy árfolyamtrend-elmozdulást prognosztizáló rendszer kifejlesztése a számítástudomány és a számítástechnika haladó eredményeinek felhasználásával.

Az elvi háttér a technikai analízis alapkövein, a stabil mintákon nyugszik, melyek konkrét előfordulását és tulajdonságait vizsgálja a diplomamunka első része a Budapesti Értéktőzsde historikus adatait alapul véve egy klaszterezési eljárás segítségével. A második lépcsőben a megtalált, speciálisan a BÉT re jellemző minták alapján, neurális háló felhasználásával készít a rendszer előrejelzést a trend várható elmozdulására.

A feladat egy, a vázolt elemzés végrehajtására képes programrendszer megalkotása volt, melyhez a Microsoft legújabb technológiáit használtam fel, így a hátteret biztosító adatbázis az MS SQL Server lett, a programok nagy részét Visual Studio .NET-ben írtam, és a programok .NET Framework 1.1-es futtató környezetet igényelnek.

A dolgozatban elsőként ismertetem a diplomamunka alapjául szolgáló elméletet, majd részletesen bemutatom a felhasznált módszereket, végül sor kerül az elkészült programegyüttes leírására is. [...]

A teljes dolgozat letölthető.

A témával kapcsolatos on-line irodalom

• Glenn J. Battaglia: Mean Square Error

  Mean square error (MSE) is an accepted measure of control and of quality but its practical use as a measure of quality has been limited. This note presents the fundamentals of MSE, introduces a normalized, new version of MSE which overcomes the most severe shortcomings of MSE, discusses breaking MSE into diagnostic components which can assist to obtain maximum improvement at minimum cost, shows how to use the components of MSE to obtain a valid assessment of statistical control, illustrates how to evaluate six-sigma quality accurately for processes which are not in statistical control, and suggests how to employ MSE effectively.

• Jingtao Yao, Chew Lim Tan: A case study on using neural networks to perform technical forecasting of forex

  This paper reports empirical evidence that a neural network model is applicable to the prediction of foreign exchange rates. Time series data and technical indicators, such as moving average, are fed to neural networks to capture the underlying `rulesa of the movement in currency exchange rates. The exchange rates between American Dollar and "ve other major currencies, Japanese Yen, Deutsch Mark, British Pound, Swiss Franc and Australian Dollar are forecast by the trained neural networks. The traditional rescaled range analysis is used to test the `e$ciencya of each market before using historical data to train the neural networks. The results presented here show that without the use of extensive market data or knowledge, useful prediction can be made and signi"cant paper pro"ts can be achieved for out-of-sample data with simple technical indicators. A further research on exchange rates between Swiss Franc and American Dollar is also conducted. However, the experiments show that with e$cient market it is not easy to make pro"ts using technical indicators or time series input neural networks. This article also discusses several issues on the frequency of sampling, choice of network architecture, forecasting periods, and measures for evaluating the model's predictive power. After presenting the experimental results, a discussion on future research concludes the paper.

• Xinghuo Yu, M. Onder Efe, and Okyay Kaynak: A General Backpropagation Algorithm for Feedforward Neural Networks Learning

  In this letter, a general backpropagation algorithm is proposed for feedforward neural networks learning with time varying inputs. The Lyapunov function approach is used to rigorously analyze the convergence of weights, with the use of the algorithm, toward minima of the error function. Sufficient conditions to guarantee the convergence of weights for time varying inputs are derived. It is shown that most commonly used backpropagation learning algorithms are special cases of the developed general algorithm

• MD. Tanvir Islam (PY), and Yoichi OKABE: A New Pattern Learning Algorithm for Multilayer Feedforward Neural Networks

  We propose a new learning algorithm for multilayer feedforward networks. The algorithm is based on Moderatism, a model of biological characteristics. This new algorithm shows superiority over the well-known Error Backpropagation in some pattern learning experiments.

• Charu C. Aggarwal, Philip S. Yu: Outlier Detection for High Dimensional Data

  The outlier detection problem has important applications in the field of fraud detection, network robustness analysis, and intrusion detection. Most such applications are high dimensional domains in which the data can contain hundreds of dimensions. Many recent algorithms use concepts of proximity in order to find outliers based on their relationship to the rest of the data. However, in high dimensional space, the data is sparse and the notion of proximity fails to retain its meaningfulness. In fact, the sparsity of high dimensional data implies that every point is an almost equally good outlier from the perspective of proximity-based definitions. Consequently, for high dimensional data, the notion of finding meaningful outliers becomes substantially more complex and non-obvious. In this paper, we discuss new techniques for outlier detection which find the outliers by studying the behavior of projections from the data set.

• Wentian Li: Applications of Akaike and Bayesian Information Criteria in Epidemiology, Linkage analysis, Microarray Data Analysis, and DNA Sequence Analysis

  talk outline

• Greg Hamerly, Charles Elkan: Alternatives to the k-means algorithm that find better clusterings

  We investigate here the behavior of the standard k-means clustering algorithm and several alternatives to it: the k-harmonic means algorithm due to Zhang and colleagues, fuzzy k-means, Gaussian expectation-maximization, and two new variants of k-harmonic means. Our aim is to find which aspects of these algorithms contribute to finding good clusterings, as opposed to converging to a low-quality local optimum. We describe each algorithm in a unified framework that introduces separate cluster membership and data weight functions. We then show that the algorithms do behave very differently from each other on simple low-dimensional synthetic datasets and image segmentation tasks, and that the k-harmonic means method is superior. Having a soft membership function is essential for finding high-quality clusterings, but having a non-constant data weight function is useful also.

• Hirotaka Mizuno, Michitaka Kosaka, Hiroshi Yajima: Application Of Neural Network To Technical Analysis Of Stock Market Prediction

  This paper presents a neural network model for technical analysis of stock market, and its application to a buying and selling timing prediction system for stock index. When the numbers of learning samples are uneven among categories, the neural network with normal learning has the problem that it tries to improve only the prediction accuracy of most dominant category. In this paper, a learning method is proposed for improving prediction accuracy of other categories, controlling the numbers of learning samples by using information about the importance of each category. Experimental simulation using actual price data is carried out to demonstrate the usefulness of the method.

• Jonathan D. Becher, Pavel Berkhin, Edmund Freeman: Automating Exploratory Data Analysis for Efficient Data Mining

  Having access to large data sets for the purpose of predictive data mining does not guarantee good models, even when the size of the training data is virtually unlimited. Instead, careful data preprocessing is required, including data cleansing, handling missing values, attribute representation and encoding, and generating derived attributes. In particular, the selection of the most appropriate subset of attributes to include is a critical step in building an accurate and efficient model. We describe an automated approach to the exploration, preprocessing, and selection of the optimal attribute subset whose goal is to simplify the KDD process and dramatically shorten the time to build a model. Our implementation finds inappropriate and suspicious attributes, performs target dependency analysis, determiningoptimal attribute encoding, generates new derived attributes, and provides a flexible approach to attribute selection. We present results generated by an industrial KDD environment called the Accrue Decision Series on several real world Web data sets.

• Pavel Berkhin: Survey of Clustering Data Mining Techniques

  No abstract is available

• John S. Breese David Heckerman, Carl Kadie: Empirical Analysis of Predictive Algorithms for Collaborate Filtering

  Collaborative filtering or recommender systems use a database about user preference to predict additional topics or products a new user might like. In this paper we describe several algorithms designed for this task, including techniques based on correlation coefficients, vector-based similarity calculations, and statistical Bayesian methods. We compare the predictive accuracy of the various methods in a set of representative problem domains. We use two basic classes of evaluation metrics. The first characterizes accuracy over a set of individual predictions in terms of average absolute deviation. The second estimates the utility of a ranked list of suggested items. This metric uses an estimate of the probability that a user will see a recommendation in an ordered list. Experiments were run for datasets associated with 3 application areas, d experimental protocols, and the 2 evaluation metrics for the various algorithms. Results indicate that for a wide range of conditions, Bayesian networks with decision trees at each node arid correlation methods outperform Bayesian-clustering and vector-similarity methods. Between correlation arid Bayesian networks, the preferred method depends on the nature of the dataset, nature of the application (ranked versus one-by-one presentation), and the availability of votes with which to make predictions. Other considerations include the size of database, speed of predictions, arid learning time

• Markus Breunig, Hans-Peter Kriegel, Raymond T. Ng, Jörg Sander: LOF: Identifying Density-Based Local Outliers

  For many KDD applications, such as detecting criminal activities in E-commerce, finding the rare instances or the outliers, can be more interesting than finding the common patterns. Existing work in outlier detection regards being an outlier as a binary property. In this paper, we contend that for many scenarios, it is more meaningful to assign to each object a degree of being an outlier. This degree is called the local outlier factor (LOF) of an object. It is local in that the degree depends on how isolated the object is with respect to the surrounding neighborhood. We give a detailed formal analysis showing that LOF enjoys many desirable properties. Using realworld datasets, we demonstrate that LOF can be used to find outliers which appear to be meaningful, but can otherwise not be identified with existing approaches. Finally, a careful performance evaluation of our algorithm confirms we show that our approach of finding local outliers can be practical.

• Donald E. Brown, Christopher L. Huntley: A Practical Application of Simulated Annealing to Clustering

  We formalize clustering as a partitioning problem with a user-defined internal clustering criterion and present SINICC, an unbiased, empirical method for comparing internal clustering criteria. An application to multi-sensor fusion is described, where the data set is composed of inexact sensor "reports" pertaining to "objects" in an environment. Given these reports, the objective is to produce a representation of the environment, where each entity in the representation is the result of "fusing" sensor reports. Before one can perform fusion, however, the reports must be "associated" into homogeneous clusters. Simulated annealing is used to find a near-optimal partitioning with respect to each of several clustering criteria for a variety of simulated data sets. This method can then be used to determine the "best" clustering criterion for the multi-sensor fusion problem with a given fusion operator.

• Saswat Anand and Wei-Ngan Chin and Siau-Cheng Khoo: Charting Patterns on Price History

  It is an established notion among ?nancial analysts that price moves in patterns and these patterns can be used to forecast future price. As the de?nitions of these patterns are often subjective, every analyst has a need to de?ne and search meaningful patterns from historical time series quickly and efficiently. However, such discovery process can be extremely laborious and technically challenging in the absence of a high level pattern de?nition language. In this paper, we propose a chart-pattern language (CPL for short) to facilitate pattern discovery process. Our language enables ?nancial analysts to (1) de?ne patterns with subjective crite- ria, through introduction of fuzzy constraints, and (2) incre- mentally compose complex patterns from simpler patterns. We demonstrate through an array of examples how real life patterns can be expressed in CPL. In short, CPL provides a high-level platform upon which analysts can de?ne and search patterns easily and without any programming exper- tise. CPL is a domain-speci?c language embedded in Haskell. We show how various features of a functional language, such as pattern matching, higher-order functions, lazy evaluation, facilitate pattern de?nitions and implementation. Further- more, Haskell's type system frees the programmers from an- notating the programs with types.

• Peter Cheeseman, John Stutz: Bayesian Classification (AutoClass): Theory and Results

  We describe AutoClass, an approach to unsupervised classification based upon the classical mixture model, supplemented by a Bayesian method for determining the optimal classes. We include a moderately detailed exposition of the mathematics behind the AutoClass system. We emphasize that no current unsupervised classification system can produce maximally useful results when operated alone. It is the interaction between domain experts and the machine searching over the model space, that generates new knowledge. Both bring unique information and abilities to the database analysis task, and each enhances the others' effectiveness. We illustrate this point with several applications of AutoClass to complex real world databases, and describe the resulting successes and failures.

• George Kollios, Dimitrios Gunopulos, Nick Koudas, Stefan Berchtold: Efficient Biased Sampling for Approximate Clustering And Outlier Detection in Large Datasets

  We investigate the use of biased sampling according to the density of the data set to speed up the operation of general data mining tasks, such as clustering and outlier detection in large multidimensional data sets. In density-biased sampling, the probability that a given point will be included in the sample depends on the local density of the data set. We propose a general technique for density-biased sampling that can factor in user requirements to sample for properties of interest and can be tuned for specific data mining tasks. This allows great flexibility and improved accuracy of the results over simple random sampling. We describe our approach in detail, we analytically evaluate it, and show how it can be optimized for approximate clustering and outlier detection.

• Gasser Auda: Cooperative Modular Neural Network Classifiers

  No abstract is available

• A. Bouguettaya, Q. Le-Viet: Data Clustering Analysis in a Multidimensional Space

  No abstract is available

• Iebeling Kaastra, Milton Boyd: Designing a neural network for forecasting financial and economic time series

  No abstract is available

• Zengyou He, Xiaofei Xu, Shengchun Deng: Discovering Cluster Based Local Outliers

  In this paper, we present the new definition for outlier: cluster-based local outlier, which is meaningful and provides importance to the local data behavior. A measure for identifying the physical significance of an outlier is designed, which is called CBLOF (Cluster-Based Local Outlier Factor). We also propose the FindCBLOF algorithm for discovering outliers. The experimental results show that our approach outperformed the existing methods on identifying meaningful and interesting outliers.

• Pedro Domingos: The Role of Occam's Razor in Knowledge Discovery

  Many KDD systems incorporate an implicit or explicit preference for simpler models, but this use of "Occam's razor" has been strongly criticized by several authors (e.g., Schaffer, 1993; Webb, 1996). This controversy arises partly because Occam's razor has been interpreted in two quite different ways. The first interpretation (simplicity is a goal in itself) is essentially correct, but is at heart a preference for more comprehensible models. The second interpretation (simplicity leads to greater accuracy) is much more problematic. A critical review of the theoretical arguments for and against it shows that it is unfounded as a universal principle, and demonstrably false. A review of empirical evidence shows that it also fails as a practical heuristic. This article argues that its continued use in KDD risks causing significant opportunities to be missed, and should therefore be restricted to the comparatively few applications where it is appropriate. The article proposes and reviews the...

• Peter C McCluskey: Feedforward and Recurrent Neural Networks and Genetic Programs for Stock Market and Time Series Forecasting

  Adding recurrence to neural networks improves their time series forecasts. Well chosen inputs such as a window of time-delayed inputs or intelligently pre-processed inputs are more important than recurrence. Neural networks do well on moderately noisy and chaotic time series, such as sunspot data. A single neural network or genetic program generalizes poorly on weekly stock market indices due to the low signal to noise ratio. When the responses of a number of networks are averaged, the resulting forecast shows substantial profits on historical data

• A.P. Papliński: Feedforward Multilayer Neural Networks

  órai tananyag

• Mario A. T. Figueiredo, Anil K. Jain: Unsupervised Learning of Finite Mixture Models

  This paper proposes an unsupervised algorithm for learning a finite mixture model from multivariate data. The adjective ?unsupervised? is justified by two properties of the algorithm: 1) it is capable of selecting the number of components and 2) unlike the standard expectation-maximization (EM) algorithm, it does not require careful initialization. The proposed method also avoids another drawback of EM for mixture fitting: the possibility of convergence toward a singular estimate at the boundary of the parameter space. The novelty of our approach is that we do not use a model selection criterion to choose one among a set of preestimated candidate models; instead, we seamlessly integrate estimation and model selection in a single algorithm. Our technique can be applied to any type of parametric mixture model for which it is possible to write an EM algorithm; in this paper, we illustrate it with experiments involving Gaussian mixtures. These experiments testify for the good performance of our approach.

• N. G. Pavlidis, D. K. Tasoulis, M. N. Vrahatis: Financial Forecasting Through Unsupervised Clustering and Evolutionary Trained Neural Networks

  This paper presents a time series forecasting methodology and applies it to generate one?step? ahead predictions for two daily foreign exchange spot rate time series. The methodology draws from the disciplines of chaotic time series analysis, clustering, artifi- cial neural networks and evolutionary computation. In brief, clustering is applied to identify neighborhoods in the reconstructed state space of the system; and subsequently neural networks are trained to model the dynamics of each neighborhood separately. The results obtained through this approach are promising.

• Wei Cheng, Lorry Wagner, and Chien-Hua Lin: Forecasting the 30-year U.S. Treasury Bond with a System of Neural Networks

  A forecasting model based on a system of artificial neural networks (ANN) is used to predict the direction of the 30-Year U.S. Treasury Bond on a weekly basis. At the close of Friday's market, the 23-variable database is updated with the latest information and the data pre-processed for input to the prediction system. Thirty-two feed-forward neural networks are trained on the new data and then individually recalled to predict the following Friday's market direction. These results are then used as input to a decision model that ultimately determines the final prediction. Forecasting began in 1989, with live trades commencing in 1990. The average accuracy of the buy prediction was 67% over a five-year period, with an average annualized return on investment (ROI) of 17.3%. This compares with an ROI of 13.9% for the Lehman Brothers T-Bond Index. This paper describes the methods used for data selection, training and testing, the basic system architecture, and how the decision model improved the total system accuracy as compared to individual networks.

• C. Fraley, A.E. Raftery: How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis

  We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian components with different parameterizations and cross-cluster constraints. Noise and outliers can be modeled by adding a Poisson process component. Partitions are determined by the EM (expectation-maximization) algorithm for maximum likelihood, with initial values from agglomerative hierarchical clustering. Models are compared using an approximation to the Bayes factor based on the Bayesian Information Criterion (BIC); unlike significance tests, this allows comparison of more than two models at the same time, and removes the restriction that the models compared be nested. The problems of determining the number of clusters and the clustering method are...

• Cedric Gegout, Bernard Girau, Fabrice Rossi: Generic Back-Propagation in Arbitrary FeedForward Neural Networks

  In this paper, we describe a general mathematical model for feedforward neural networks. The final form of the network is a vectorial function f of two variables, x (the input of the network) and w (the weight vector). We show that the differential of f can be computed with an extended back-propagation algorithm or with a direct method. By evaluating the time needed to compute the differential with the help of both meth- ods, we show how to chose the best one. We introduce also input sharing and output function which allow us to implement efficiently a multilayer perceptron with our model.

• Jingtao Yao, Chew Lim Tan: Guidelines for Financial Forecasting with Neural Networks

  Neural networks are good at classification, forecasting and recognition. They are also good candidates of financial forecasting tools. Forecasting is often used in the decision making process. Neural network training is an art. Trading based on neural network outputs, or trading strategy is also an art. We will discuss a seven-step neural network forecasting model building approach in this article. Pre and post data processing/analysis skills, data sampling, training criteria and model recommendation will also be covered in this article

• Bin Zhang, Meichun Hsu, Umeshwar Dayal: K-Harmonic Means - A Data Clustering Algorithm

  Data clustering is one of the common techniques used in data mining. A popular performance function for measuring goodness of data clustering is the total within-cluster variance, or the total mean-square quantization error (MSE). The KMeans (KM) algorithm is a popular algorithm which attempts to find a K-clustering which minimizes MSE. The K-Means algorithm is a center-based clustering algorithm. The dependency of the K-Means performance on the initialization of the centers is a major problem; a similar issue exists for an alternative algorithm, Expectation Maximization (EM), although to a lesser extent. In this paper, we propose a new clustering method called the K-Harmonic Means algorithm (KHM). KHM is a center-based clustering algorithm which uses the Harmonic Averages of the distances from each data point to the centers as components to its performance function. It is demonstrated that K-Harmonic Means is essentially insensitive to the initialization of the centers. In certain cases, K-Harmonic Means significantly improves the quality of clustering results comparing with both K-Means and EM, which are the two most popular clustering algorithms used in data exploration and data compression. A unified view of the three performance functions, K-Means?, K-Harmonic Means? and EM?s, are given for comparison. Experimental results of KHM comparing with KM on high dimensional data and visualization of the animation of the convergence of all three algorithms using 2- dimensional data are given

• Anil K. Jain, Robert P.W. Duin, and Jianchang Mao: Statistical Pattern Recognition: A Review

  The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques and methods imported from statistical learning theory have been receiving increasing attention. The design of a recognition system requires careful attention to the following issues: definition of pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design and learning, selection of training and test samples, and performance evaluation. In spite of almost 50 years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition, and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.

• Anil K. Jain, M.N. MURTY, P.J. FLYNN: Data Clustering: A Review

  Clustering is the unsupervised classification of patterns (observations, data items, or feature vectors) into groups (clusters). The clustering problem has been addressed in many contexts and by researchers in many disciplines; this reflects its broad appeal and usefulness as one of the steps in exploratory data analysis. However, clustering is a difficult problem combinatorially, and differences in assumptions and contexts in different communities has made the transfer of useful generic concepts and methodologies slow to occur. This paper presents an overview of pattern clustering methods from a statistical pattern recognition perspective, with a goal of providing useful advice and references to fundamental concepts accessible to the broad community of clustering practitioners. We present a taxonomy of clustering techniques, and identify cross-cutting themes and recent advances. We also describe some important applications of clustering algorithms such as image segmentation, object recognition, and information retrieval.

• Andrew W. Moore: K-means and Hierarchical Clustering

  slide show

• Lynn Kuo, Bani Mallick: Variable Selection for Regression Models (

  A simple method for subset selection of independent variables in regression models is proposed. We expand the usual regression equation to an equation that incorporates all possible subsets of predictors by adding indicator variables as parameters. The vector of indicator variables dictates which predictors to include. Several choices of priors can be employed for the unknown regression coefficients and the unknown indicator parameters. The posterior distribution of the indicator vector is approximated by means of the Markov chain Monte Carlo algorithm. We select subsets with high posterior probabilities. In addition to linear models, we consider generalized linear models.

• Dr. Hamid R. Berenji, David Vengerov: Learning, Cooperation, and Coordination in Multi-Agent Systems

  No abstract is available

• Aristidis Likas, Nikos Vlassis, Jakob J. Verbeek: The Global K-Means Clustering Algorithm

  We present the global k-means algorithm which is an incremental approach to clustering that dynamically adds one cluster center at a time through a deterministic global search procedure consisting of N (with N being the size of the data set) executions of the k-means algorithm from suitable initial positions. We also propose modifications of the method to reduce the computational load without significantly affecting solution quality. The proposed clustering methods are tested on well-known data sets and they compare favorably to the k-means algorithm with random restarts.

• A. Likas, N. Vlassis, J.J. Verbeek: The Global K-Means Clustering Algorithm

  We present the global k-means clustering algorithm which is an incremental approach to clustering that dynamically ne cluster center at a time through a deterministic global search procedure consisting of N (with N being the size of the data set) executions of the k-means algorithm from suitable initial positions. We also propose modifications of the method to reduce the computational load without significantly affecting solution quality. The proposed clustering methods are tested on well-known data sets and they compare favorably to the k-means algorihtm with random restarts.

• Multilayer Neural Networks

  Előadás fólia

• MERLISE CLYDE, GIOVANNI PARMIGIANI and BRANI VIDAKOVIC: Multiple Shrinkage and Subset Selection in Wavelets

  This paper discusses Bayesian methods for multiple shrinkage estimation in wavelets. Wavelets are used in applications for data denoising, via shrinkage of the coefficients towards zero, and for data compression, by shrinkage and setting small coefficients to zero. We approach wavelet shrinkage by using Bayesian hierarchical models, assigning a positive prior probability to the wavelet coefficients being zero. The resulting estimator for the wavelet coefficients is a multiple shrinkage estimator that exhibits a wide variety of nonlinear shrinkage patterns. We discuss fast computational implementations, with a focus on easy-to-compute analytic approximations as well as importance sampling and Markov chain Monte Carlo methods. Multiple shrinkage estimators prove to have excellent mean squared error performance in reconstructing standard test functions. We demonstrate this in simulated test examples, comparing various implementations of multiple shrinkage to commonly used shrinkage rules. Finally, we illustrate our approach with an application to the so-called "glint" data.

• Ah-Hwee Tan, Loo-Nin Teow: Neural Logic Networks For Pattern Recognition Time Series Prediction, and Knowledge Integration

  No abstract is available

• Jingtao Yao, Chew Lim Tan, Hean-Lee Poh: Neural Networks For Technical Analysis: A study on KLCI

  This paper presents a study of artificial neural nets for use in stock index forecasting. The data from major emerging market, Kuala Lumpur Stick Exchange, are epplied as a case tudy. Based on the rescaled range analysis, a backpropagation neural network is used to capture the relationship between the technical indicators and the levels of the index in the market under study over time. Using different trading strategies, a significant paper profit can be achieved by purchasing the indexed stocks in the respective propportions. The resutls show that the neural network model can get better returns compared with the conventional ARIMA models.

• Georg Dorffner: Neural Networks For Time Series Processing

  This paper provides an overview over the most common neural network types for time series processing, i.e. pattern recognition and forecasting in spatio-temporal patterns. Emphasis is put on the relationship between neural networks models and more classical approaches to time series processing, inparticular forecasting. The paper begins with an introduction of the basics of time series processing, and discusses feedforward as well as recurrent neural networks, with respect to their ability to model non-linear dependecies in spatio-temporal patterns.

• Michael I Jordan and Christopher M. Bishop: Neural Networks

  We present an overview of current research on artificial neural networks, emphasizing a statistical perpestcive. We view neural networks as parameterized graphs that make probabilistic assumtions about data, and view learning algorithms methods for finding parameter values that look probable in the light of th edata.

• A. Hotho, S. Staab, A. Maedche: Ontology-based Text Clustering

  Text clustering typically involves clustering in a high dimensional space, which appears difficult with regard to virtually all practical settings. In addition, given a particular clustering result it is typically very hard to come up with a good explanation of why the text clusters have been constructed the way they are. In this paper, we propose a new approach for applying background knowledge during preprocessing in order to improve clustering results and allow for selection between results. We built various views basing our selection of text features on a heterarchy of concepts. Based on these aggregations, we compute multiple clustering results using K-Means. The results may be distinguished and explained by the corresponding selection of concepts in the ontology. Our results compare favourably with a sophisticated baseline preprocessing strategy.

• Tak-chung Fu, Fu-lai Chung?, Vincent Ng and Robert Luk: Pattern Discovery from Stock Time Series Using Self-Organizing Maps

  Pattern discovery from time series is of fundamental importance. Particularly when the domain expert derived patterns do not exist or are not complete, an algorithm to discover specific patterns or shapes automatically from the time series data is necessary. Such an algorithm is noteworthy in that it does not assume prior knowledge of the number of interesting structures, nor does it require an exhaustive explanation of the patterns being described. In this paper, a clustering approach is proposed for pattern discovery from time series. In view of its popularity and superior clustering performance, the self-organizing map (SOM) was adopted for pattern discovery in temporal data sequences. It is a special type of clustering algorithm that imposes a topological structure on the data. To prepare for the SOM algorithm, data sequences are segmented from the numerical time series using a continuous sliding window. Similar temporal patterns are then grouped together using SOM into clusters, which may subsequently be used to represent different structures of the data or temporal patterns. Attempts have been made to tackle the problem of representing patterns in a multi-resolution manner. With the increase in the number of data points in the patterns (the length of patterns), the time needed for the discovery process increases exponentially. To address this problem, we propose to compress the input patterns by a perceptually important point (PIP) identification algorithm. The idea is to replace the original data segment by its PIP?s so that the dimensionality of the input pattern can be reduced. Encouraging results are observed and reported for the application of the proposed methods to the time series collected from the Hong Kong stock market.

• Amir Assadi: A Basic Introduction To Pattern Recognition

  Órai Jegyzet

• Dan Pelleg, Andrew Moore: X-means: Extending K-means with Efficient Estimation of the Number of Clusters

  Despite its popularity for general clustering, K-means suffers three major shortcomings; it scales poorly computationally, the number of clusters K has to be supplied by the user, and the search is prone to local minima. We propose solutions for the first two problems, and a partial remedy for the third. Building on prior work for algorithmic acceleration that is not based on approximation, we introduce a new algorithm that efficiently, searches the space of cluster locations and number of clusters to optimize the Bayesian Information Criterion (BIC) or the Akaike Information Criterion (AIC) measure. The innovations include two new ways of exploiting cached sufficient statistics and a new very efficient test that in one K-means sweep selects the most promising subset of classes for refinement. This gives rise to a fast, statistically founded algorithm that outputs both the number of classes and their parameters. Experiments show this technique reveals the true num...

• Rakesh Agrawal, Giuseppe Psaila Edward L Wimmers, Mohamed Zait: Querying Shapes of Histories

  We present a shape defintion language, called SDL for retrieving objects based on on shapes contained in the histories associated with these objects..

• Shi Zhong, Joydeep Ghosh: Scalable, Balanced Model-based Clustering

  This paper presents a general framework for adapting any generative (model-based) clustering algorithm to provide balanced solutions, i.e., clusters of comparable sizes. Partitional, model-based clustering algorithms are viewed as an iterative two-step optimization process---iterative model re-estimation and sample re-assignment. Instead of a maximum-likelihood (ML) assignment, a balanceconstrained approach is used for the sample assignment step. An e#cient iterative bipartitioning heuristic is developed to reduce the computational complexity of this step and make the balanced sample assignment algorithm scalable to large datasets. We demonstrate the superiority of this approach to regular ML clustering on complex data such as arbitraryshape 2-D spatial data, high-dimensional text documents, and EEG time series.

• Rakesh Agrawal, Christos Faloutsos, Arun Swami: Efficient Similarity Search In Sequence Databases

  talk outline

• Ian Davidson and Ashwin Satyanarayana: Speeding up k-means Clustering by Bootstrap Averaging

  No abstract is available

• David Vengerov: Stock Market Pattern Recognition with Neural Networks

  No abstract is available

• Bruno A. Olshausen: Supervised learning in multilayer feedforward networks - "backpropagation"

  This handout describes the backpropagation learning rule for multilayer feedforward networks composed of McCulloch-Pitts type neurons. These meth- ods were originally worked out by Rumelhart, Hinton, and Williams in the mid- 1980's. The derivation of a learning rule for multilayer networks represented a critical breakthrough in the field, because it allows for complex input-output relationships to be learned that could not be achieved by a single-stage network due to the limitation of linear separability.

• Serhiy Shtovba, Yakiv Mashnitskiy: The Backpropagation Multilayer Feedforward Neural Network Based Competition Task Solution

  The competition task solution, which was based on the use of the neural network and analysis of results are presented in the paper.

• JEFF FLEMING, CHRIS KIRBY, BARBARA OSTDIEK: The Economic Value of Volatility Timing

  Numerous studies report that standard volatility models have low explanatory power, leading some researchers to question whether these models have economic value. We examine this question by using conditional mean-variance analysis to assess the value of volatility timing to short-horizon investors. We find that the volatility timing strategies outperform the unconditionally efficient static portfolios that have the same target expected return and volatility. This finding is robust to estimation risk and transaction costs.

• James Theiler, Galen Gisler: A Contiguity-Enhanced K-Means Clustering Algorithm for Unsupervised Multispectral Image Segmentation

  The recent and continuing construction of multi- and hyper-spectral imagers will provide detailed data cubes with information in both the spatial and spectral domain. This data shows great promise for remote sensing applications ranging from environmental and agricultural to national security interests. The reduction of this voluminous data to useful intermediate forms is necessary both for downlinking all those bits and for interpreting them. Smart on-board hardware is required, as well as sophisticated earth-bound processing. A segmented image (in which the multispectral data in each pixel is classified into one of a small number of categories) is one kind of intermediate form which provides some measure of data compression. Traditional image segmentation algorithms treat pixels independently and cluster the pixels according only to their spectral information. This neglects the implicit spatial information that is available in the image. We will suggest a simple approach --- a varian...

• Leejay Wu May: Thesis Proposal: Designing Distance Functions

  Distance function design is a fundamental problem underlying much of data mining. Numerous methods designed to solve a variety of problems ranging from clustering to nearest-neighbor retrieval require some concept of distance with which to compare objects. Approaches can be classified as user-driven methods, in which complex distance functions are induced from user input or other sources of labels and feedback; and data-driven methods, which perform distance function design as a preprocessing step without such additional input. With regards to the first, the FALCON system provides example-driven query development via relevance-feedback; the second approach includes a scaling and selection system. Both are works in progress. I will focus on the latter, with an eye towards adding reconstruction to the scaling and selection components as well as developing and testing a fair evaluation scheme to quantify how good a set of scales is. The former, FALCON, will also be extended but to a much lesser degree.

• Thomas Klarik and Gottfried Rudorfer: Time Series Forecasting Using Neural Networks

  Artificial neural networks are suitable for many tasks in pattern recognition and machine learning. In this paper we present an APL system for forecasting univariate time series with neural networks. Unlike conventional tchniques for time series analysis, an artificial neural network needs littl einformation about the time series data and can be applied to a broad range if problems. ...

• Jonathan Shlens: Time Series Prediction with Artificial Neural Networks

  This research project investigates the ablity of artificial neural networks to predict time series. Specifically, this project examined two applications: predicting electrical power demand and earthquake tremors in Los Angeles. ?

• Junmyung Lee, Sungzoon Cho and Jinwoo Baek: Trend Detection Using Auto-Associative Neural Networks : Intraday KOSPI 200 Futures

  This paper reports the results of a new neural network based trend detector. An auto-associative neural network was trained with the ?trend? data obtained from the intra-day KOSPI 200 future price. It was then used to predict a trend. Simple investment strategies based on the detector achieved a one-year return of 31.2 points with no leverage.

• Eamonn Keogh : A Tutorial on Indexing and Mining Time Series Data

  presentation

• Ramon Lawrence: Using Neural Networks to Forecast Stock Market Prices

  This paper is a survey on the application of neural networks in forecasting stock market prices. With their ability to discover patterns in nonlinear and chaotic systems, neural networks offer the ability to predict market directions more accurately than current techniques. Common market analysis techniques such as technical analysis, fundamental analysis, and regression are discussed and compared with neural network performance. Also, the Efficient Market Hypothesis (EMH) is presented and contrasted with chaos theory and neural networks. This paper refutes the EMH based on previous neural network work. Finally, future directions for applying neural networks to the financial markets are discussed.

• Amitabh Chaudhary, Alexander S. Szalay, and Andrew W. Moore Dep: Very Fast Outlier Detection in Large Multidimensional Data Sets

  Outliers are objects that do not comply with the general behavior of the data. Applications such as exploration in science databases need fast interactive tools for outlier detection in data sets that have unknown distributions, are large in size, and are in high dimensional space. Existing algorithms for outlier detection are too slow for such applications. We present an algorithm based on an innovative use of k-d trees that doesn't assume any probability model and is linear in the number of objects and in the number of dimensions. We also provide experimental results that show that this is indeed a practical solution to the above problem.

• Eamonn Keogh : Exact Indexing of Dynamic Time Warping

  The problem of indexing time series has attracted much research interest in the database community. Most algorithms used to index time series utilize the Euclidean distance or some variation thereof. However is has been forcefully shown that the Euclidean distance is a very brittle distance measure. Dynamic Time Warping (DTW) is a much more robust distance measure for time series, allowing similar shapes to match even if they are out of phase in the time axis.

• Vitorino Ramos, Pedro Pina, Fernando Muge: From Feature Extraction to Classification: A Multidisciplinary Approach applied to Portuguese Granites

  The purpose of this paper is to present a complete methodology based on a multidisciplinary approach, that goes from the extraction of features till the classification of a set of different portuguese granites. The set of tools to extract the features that characterise polished surfaces of the granites is mainly based on mathematical morphology. The classification methodology is based on a genetic algorithm capable of search the input feature space used by the nearest neighbour rule classifier. Results show that is adequate to perform feature reduction and simultaneous improve the recognition rate. Moreover, the present methodology represents a robust strategy to understand the proper nature of the images treated, and their discriminant features.

• Kenji Yamanishi, Jun-ichi Takeuchi, Knowledge Discovery and Data Mining: On-line Unsupervised Outlier Detection Using Finite Mixtures with Discounting Learning Algorithms

  Outlier detection is a fundamental issue in data mining, speci cally in fraud detection, network intrusion detection, network monitoring, etc. SmartSifter, which we abbreviate as SS, is an outlier detection engine adrressing this problem from the viewpoint of statistical learning theory. This paper provides a theoretical basis for SS and empirically demonstrates its effectiveness. SS detects outliers in an online process through the on-line unsupervised learning of a probabilistic model (using a finite mixture model) of the information source. Each time a datum is input SS employs an on-line discounting learning algorithm to learn the probabilistic model. A score is given to the datum based on the learned model, with a high score indicating a high possibility of being a statistical outlier. The novel features of SS are: 1) it is adaptive to non-stationary sources of data; 2) a score has a clear statistical/information-theoretic meaning; 3) it is computationally inexpensive; and 4) it can handle both categorical and continuous variables. An experimental application to network intrusion detection shows that SS was able to identify data with high scores that corresponded to attacks, with low computational costs. Further experimental application has identified a number of meaningful rare cases in actual health insurance pathology data from Australia's Health Insurance Commission. 1.