In this paper, we propose a new efficient data reduction algorithm through combining lattice with rough set. On the basis of lattice learning, the algorithm applies the concept of attribute reduction in the theory of ...
详细信息
ISBN:
(纸本)1424400600
In this paper, we propose a new efficient data reduction algorithm through combining lattice with rough set. On the basis of lattice learning, the algorithm applies the concept of attribute reduction in the theory of Rough Sets and calculates the importance degree of attributes automatically by a density based approach. Under acceptable classification precision and complexity, it reduces row and column together and generates concise classification rules. the algorithm represents a solution to the problem of attribute generalization on the basis of lattice learning and automatic estimation of attribute weights independently of domain experts. Attributes in the classification rules are ordered by the importance degree of attribute. So in the classification and by the sequence of importance degree of attribute, from one attribute to another, we can exclude the objects which dissatisfy the constraint from the attribute. And then it can, to a large extent, reduces the size of data set of object classified by scanning attribute of the rules, and thereby the efficiency of classification is improved greatly.
Induce learning is one of the most important research areas on application of artificial intelligence. the classification algorithm is the core of the machinelearning. the article advanced a kind of decision classifi...
详细信息
ISBN:
(纸本)1424400600
Induce learning is one of the most important research areas on application of artificial intelligence. the classification algorithm is the core of the machinelearning. the article advanced a kind of decision classification algorithm that is in View of attribute value statistical regularity, on the characteristic and insufficient of ID3, C4.5 Algorithm. (In short of BS-CA) this algorithm takes the statistical regularity as foundation, and takes the multiplication as comprehensive tactics, simplifying the trimming and optimization processes of decision tree. the theory analysis and experiment shows that BS-CA possesses high accuracy, high classification speed, simple and high activity.
In patternrecognition, feature selection aims to choose the smallest subset of features that is necessary and sufficient to describe the target concept. In this paper, a mutual information-based constructive criterio...
详细信息
ISBN:
(纸本)9781424404759
In patternrecognition, feature selection aims to choose the smallest subset of features that is necessary and sufficient to describe the target concept. In this paper, a mutual information-based constructive criterion under arbitrary information distributions of input features is presented for feature selection. this criterion can capture boththe relevance to the output classes and the redundancy with respect to the already,-selected features without any parameters like 8 in MIFS or MIFS-U methods to be preset. Furthermore, a modified greedy feature selection algorithm called MICC is proposed, and experimental results demonstrate the good performance of MICC on both synthetic and benchmark data sets.
the proceedings contain 68 papers. the topics discussed include: a comparative analysis of data distribution methods in an agent-based neural system for classification tasks;stochastic differential portfolio games wit...
详细信息
ISBN:
(纸本)0769526624
the proceedings contain 68 papers. the topics discussed include: a comparative analysis of data distribution methods in an agent-based neural system for classification tasks;stochastic differential portfolio games with regime switching model;extracting symbolic rules from clustering of gene expression data;a novel microarray gene selection method based on consistency;combining greedy method and genetic algorithm to identify transcription factor binding sites;investigation of a new artificial immune system model applied to patternrecognition;RLM: a new method of encoding weights in DNA strands;shape representation and distance measure based on retational graph;fast modeling of curved object from two images;research on an improved gray gradient orientation algorithm in anisotropic high-pass filtering;and image color reduction based on self-organizing maps and growing self-organizing neural networks.
In some cases, an ambiguous pattern may belong to more than one class, however it is forcibly classified to one of these classes in conventional support vector machine. Handling those ambiguous patterns in this way ma...
详细信息
ISBN:
(纸本)1424400600
In some cases, an ambiguous pattern may belong to more than one class, however it is forcibly classified to one of these classes in conventional support vector machine. Handling those ambiguous patterns in this way may loss the uncertainly information of the patterns. therefore, we prefer to keep the uncertainty information in the ambiguous patterns. In this work, instead of two-class classification, we propose to classify samples into four classes: namely positive, negative ambiguous and outlier classes.
the effective representation and amalgamating extraction method of frequent itemsets in distributed databases is important to improve the result of distributed association rules mining. the common methods are ineffici...
详细信息
ISBN:
(纸本)1424400600
the effective representation and amalgamating extraction method of frequent itemsets in distributed databases is important to improve the result of distributed association rules mining. the common methods are inefficient due either to higher number of database scan or to larger amount of candidate itemsets for communication. Based on discussing the relation between the concept of pruned concept lattice (PCL) and the representation of frequent itemsets, the Closed Frequent Itemsets of PCL is defined. UMPCL_I, an approximate amalgamation and extraction method of frequent itemsets in horizontally partitioned databases based on multiple PCL, is proposed. the main ideas of this method are using a frequent concept to represent some few of frequent itemsets, and using a local support slightly lower than global support to prune sub-lattices before been amalgamated to decrease the size of exchanged messages. the theoretic analysis and experiment show that such method is efficient.
Condition monitoring and classification of machinery health state is of great practical significance in manufacturing industry, because it provides updated information regarding machine status on-line, thus avoiding t...
详细信息
ISBN:
(纸本)9783540490265
Condition monitoring and classification of machinery health state is of great practical significance in manufacturing industry, because it provides updated information regarding machine status on-line, thus avoiding the production loss and minimizing the chances of catastrophic machine failures. this is a patternrecognition problem and a condition monitoring system based on a hybrid of neural network and hidden Markov model (HMM) is proposed in this paper. Neural network realizes dimensionality reduction for Lipschitz exponent functions obtained from vibration data as input features and hidden Markov model is used for condition classification. the machinery condition can be identified by selecting the corresponding HMM which maximizes the probability of a given observation sequence. In the end, the proposed method is validated using gearbox vibration data.
In this paper, we investigate the problem of learningthe decision functions for sequential data describing complex objects that are composed of subobjects. the decision function maps sequence of attribute values into...
详细信息
ISBN:
(纸本)3540476938
In this paper, we investigate the problem of learningthe decision functions for sequential data describing complex objects that are composed of subobjects. the decision function maps sequence of attribute values into a relational structure, representing properties of the object described by the sequence. this relational structure is constructed in a way that allows us to answer questions from a given language. the decision function is constructed by means of rule system. the rules are learned incrementally in a dialog with an expert. We also present an algorithm that implements the rule system and we apply it to real life problems.
Clustering is one branch of unsupervised machinelearningtheory, which has a wide variety of applications in patternrecognition, image processing, economics, document categorization, web mining, etc. Today, we const...
详细信息
ISBN:
(纸本)3540465359
Clustering is one branch of unsupervised machinelearningtheory, which has a wide variety of applications in patternrecognition, image processing, economics, document categorization, web mining, etc. Today, we constantly face how to handle a large number of similar data items, which drives many researchers to contribute themselves to this field. Support vector machine provides a new pathway for clustering, however, it behaves bad in handling massive data. As an emergent theory, artificial immune system can effectively recognize antigens and produce the memory antibodies. this mechanism is constantly used to achieve representative or feature data from raw data. A combinational clustering method is proposed in this paper based on artificial immune system and support vector machine, Experimentation in functionality and performance is done in detail. Finally a more challenging application in elevator industry is conducted. the results strongly indicate that this combinational clustering in this paper is of feasibility and of practice.
A conceptual clustering program CLUSTER3 is described that, given a set of objects represented by attribute-value tuples, groups them into clusters described by generalized conjunctive descriptions in attributional ca...
详细信息
ISBN:
(纸本)1845641787
A conceptual clustering program CLUSTER3 is described that, given a set of objects represented by attribute-value tuples, groups them into clusters described by generalized conjunctive descriptions in attributional calculus. the descriptions are optimized according to a user-designed multi-criterion clustering quality measure. the clustering process in CLUSTER3 depends on a viewpoint underlying the clustering goal, and employs the view-relevant attribute subsetting method (VAS) that selects for clustering only attributes relevant to this viewpoint. the program is illustrated by a simple designed problem and by its application to clustering of US Congressional voting records. the ongoing research concerns application of CLUSTER3 to large and complex datasets such as collections of web pages.
暂无评论