Automatic clustering method 



An automatic pattern recognition method has a short processing time, can be applied to nonlinear separation problems, and can perform similarity calculations. The method: Divides a plurality of sample data of known categories into a plurality of classes; When the sample data in a divided class is not all in the same category, repeats dividing the sample data into subclasses until sample data in a subclass has only one category; Expresses the relationship between classes and subclasses in a tree-structure representation and determines the standard pattern for each class and subclass from the sample data contained there; and Checks which of the tree-structured classes input data of unknown category is nearest, by calculating the distance to the standard pattern of each class, and then, when the class has subclasses, performs a similar check until the lowest-level subclass is reached to determine the subclass the input data is closest to. The category of the lowest-level subclass is taken as the category of the input data. 

Method and system for data clustering for very large databases 



Multi-dimensional data contained in very large databases is efficiently and accurately clustered to determine patterns therein and extract useful information from such patterns. Conventional computer processors may be used which have limited memory capacity and conventional operating speed, allowing massive data sets to be processed in a reasonable time and with reasonable computer resources. The clustering process is organized using a clustering feature tree structure wherein each clustering feature comprises the number of data points in the cluster, the linear sum of the data points in the cluster, and the square sum of the data points in the cluster. A dense region of data points is treated collectively as a single cluster, and points in sparsely occupied regions can be treated as outliers and removed from the clustering feature tree. The clustering can be carried out continuously with new data points being received and processed, and with the clustering feature tree being restructured as necessary to accommodate the information from the newly received data points. 

Method for mining causality rules with applications to electronic commerce 



For mining causality rules in an event database, the rules are obtained by iteratively generating candidate rules and counting their occurrences in the event database. Newly identified causality rules are used to generate the next set of candidate rules to be evaluated, by increasing the size of the set of consequential events triggered by triggering events and/or the number of triggering events. The preferred embodiment uses an iterative approach to deriving the causality rules in order of the consequential set sizes and triggering set sizes. The detection of an occurrence of a causality rule in an event sequence is handled as a sub-sequence matching problem using a novel hierarchical matching method to improve efficiency. 

Object-oriented data mining and decision making system 



The Object-Oriented Data Mining and Decision Making System is invented and disclosed. This system is an integration of two subsystems, the object-oriented data mining subsystem which is an object-oriented machine learning system, and the object-oriented decision making subsystem which is also called the object-oriented expert system. In this invention, object-oriented technology, a new technology in software design and development is applied. The difference between the methodology used in this invention and that used in all other learning systems and expert systems is that in this invention, all objects have the same attributes and actions and are in the same class in the whole process. At the same time, only objects themselves, nothing else are processed, without adding any additional structures, such as trees, hierarchies, and relations to them. Therefore a single class can cover all working objects in the whole process. Besides, statistical methodology is introduced to each object in the class. 

System and method for data mining from relational data by sieving through iterated relational reinforcement 



A system and method are provided for performing the process known as "data mining" on a database of raw data records having common data elements, to obtain categorical cluster rules as to what elements of the data tend to occur in common in multiple records. Initial values are assigned to the elements. In an iterative process, the associated value for each given one of the elements is recalculated based on the values of other elements which occur in records together with the given element. Thus, the associated values will tend to grow for elements occurring together in multiple records. Those common occurrences of elements in multiple records represent categorical cluster rules the owner of the data is likely to want to know about. Thus, these rules may be identified based on the growth of the associated values for the records.