A Method to Identify Feature Constraints Based on Feature Selections Mining.

Creating Nested Mappings with Clio.

Deadline and QoS Aware Data Warehouse.

Network%2DBased Problem Detection for Distributed Systems.

Algorithms for Finding Attribute Value Group for Binary Segmentation of Categorical Databases.



Mining Structured Association Patterns from Databases.

Mining Optimized Association Rules for Numeric Attributes.
Given a huge database, we address the problem of finding association rules for numeric attributes, such as (Balance c 1) + (CardLoan = yes), which implies that bank customers whose balances fall in a range 1 are likely to use card loan with a probability greater than p. The above rule is interesting only if the range 1 has some special feature with respect to the interrelation between Balance and CardLoan. It is required that the number of customers whose balances are contained in 1 (called the support of 1) is sufficient, and also that the probability p (called the confidence raho) should be much higher than the average probability of the condition being met. Our goal is to realize a system that finds such appropriate ranges automatically. We mainly focus on computing two
Parallel Processing of Multiple Aggregate Queries on Shared%2DNothing Multiprocessors.

Algorithms for Mining Association Rules for Binary Segmentations of Huge Categorical Databases.
We consider the problem of finding association rules that make nearly optimal binary segmen-tations of huge categorical databases. The op-timality of segmentation is defined by an ob-jective function suitable for the user’s objec-tive. An objective function is usually defined in terms of the distribution of a given target attribute. Our goal is to find association rules that split databases into two subsets, optimiz-ing the value of an objective function. The problem is intractable for general ob-jective functions, because letting N be the number of records of a given database, there are 2N possible binary segmentations, and we may have to exhaustively examine all of them. However, when the objective function is convex, there are feasible algorithms for finding nearly optimal binary segmentations, and we prove that typical criteria, such as “entropy (mutual information), ” “x2 (correla-tion), ” and “gini index (mean squared error),” are actually convex. We propose practical algorithms that use computational geometry techniques to handle
Computing Optimized Rectilinear Regions for Association Rules.

Interval Finding and Its Application to Data Mining.

Mining Optimized Association Rules for Numeric Attributes.
Given a huge database, we address the problem of finding association rules for numeric attributes, such as (Balance c 1) + (CardLoan = yes), which implies that bank customers whose balances fall in a range 1 are likely to use card loan with a probability greater than p. The above rule is interesting only if the range 1 has some special feature with respect to the interrelation between Balance and CardLoan. It is required that the number of customers whose balances are contained in 1 (called the support of 1) is sufficient, and also that the probability p (called the confidence raho) should be much higher than the average probability of the condition being met. Our goal is to realize a system that finds such appropriate ranges automatically. We mainly focus on computing two
Data Mining Using Two%2DDimensional Optimized Accociation Rules Scheme

SONAR System for Optimized Numeric AssociationRules.
We propose an extension of an entropy-based heuristic of Quinlan [Q93] for constructing a decision tree from a large database with many numeric attributes. Quinlan pointed out that his original method (as well as other existing methods) may be inefficient if any numeric attributes are strongly correlated. Our approach offers one solution to this problem. For each pair of numeric attributes with strong correlation, we compute a two-dimensional association rule with respect to these attributes and the objective attribute of the decision tree. In particular, we consider a family R of grid-regions in the plane associated with the pair of attributes. For R 2 R, the data can be split into two classes: data inside R and data outside R. We compute the region R opt 2 R that minimizes the entropy of the splitting, and add the splitting associated with R opt (for each pair of strongly correlated attributes) to the set of candidate tests in Quinlan&#039;s entropy-based heuristic.  We give efficient alg...
Constructing Efficient Decision Trees by Using Optimized Numeric Association Rules.
We propose an extension of an entropy-based heuristic of Quinlan [Q93] for constructing a decision tree from a large database with many numeric attributes. Quinlan pointed out that his original method (as well as other existing methods) may be inefficient if any numeric attributes are strongly correlated. Our approach offers one solution to this problem. For each pair of numeric attributes with strong correlation, we compute a two-dimensional association rule with respect to these attributes and the objective attribute of the decision tree. In particular, we consider a family R of grid-regions in the plane associated with the pair of attributes. For R 2 R, the data can be split into two classes: data inside R and data outside R. We compute the region R opt 2 R that minimizes the entropy of the splitting, and add the splitting associated with R opt (for each pair of strongly correlated attributes) to the set of candidate tests in Quinlan&#039;s entropy-based heuristic.  We give efficient alg...
