Tzung-Pei Hong

Induction of fuzzy rules and membership functions from training examples

Abstract Most fuzzy controllers and fuzzy expert systems must predefine membership functions and fuzzy inference rules to map numeric data into linguistic variable terms and to make fuzzy reasoning work. In this paper, we propose a general learning method as a framework for automatically deriving membership functions and fuzzy if-then rules from a set of given training examples to rapidly build a prototype fuzzy expert system. Based on the membership functions and the fuzzy rules derived, a corresponding fuzzy inference procedure to process inputs is also developed.

Mining association rules from quantitative data

Data-mining is the process of extracting desirable knowledge or interesting patterns from existing databases for specific purposes. Most conventional data-mining algorithms identify the relationships among transactions using binary values, however, transactions with quantitative values are commonly seen in real-world applications. This paper thus proposes a new data-mining algorithm for extracting interesting knowledge from transactions stored as quantitative values. The proposed algorithm integrates fuzzy set concepts and the apriori mining algorithm to find interesting fuzzy association rules in given transaction data sets. Experiments with student grades at I-Shou University were also made to verify the performance of the proposed algorithm.

Integrating fuzzy knowledge by genetic algorithms

We propose a genetic algorithm-based fuzzy knowledge integration framework that can simultaneously integrate multiple fuzzy rule sets and their membership function sets. The proposed approach consists of two phases: fuzzy knowledge encoding and fuzzy knowledge integration. In the encoding phase, each fuzzy rule set with its associated membership functions is first transformed into an intermediary representation and then further encoded as a string. The combined strings form an initial knowledge population, which is then ready for integration. In the knowledge-integration phase, a genetic algorithm is used to generate an optimal or nearly optimal set of fuzzy rules and membership functions from the initial knowledge population. Two application domains, the hepatitis diagnosis and the sugarcane breeding prediction, were used to show the performance of the proposed knowledge-integration approach. Results show that the fuzzy knowledge base derived using our approach performs better than every individual knowledge base.

Finding relevant attributes and membership functions

Fuzzy systems that automatically derive fuzzy if-then rules from numeric data have been developed. Most have to predefine membership functions in order to learn. Hong and Lee proposed a general learning method that automatically derives fuzzy if-then rules and membership functions from a set of given training examples using a decision table. All available attributes were included in the decision table and the initial membership functions for each attribute were built according to the predefined smallest unit. Although Hong and Lees method accurately derives the fuzzy if-then rules and final membership functions, the decision table and the initial membership functions are complex if there are many attributes or if the predefined unit is small. We improve Hong and Lees method by first selecting relevant attributes and building appropriate initial membership functions. These attributes and membership functions are then used in a decision table to derive final fuzzy if-then rules and membership functions. Experimental results on Iris data show that the proposed method effectively induces membership functions and fuzzy if-then rules.

Incrementally fast updated frequent pattern trees

The frequent-pattern-tree (FP-tree) is an efficient data structure for association-rule mining without generation of candidate itemsets. It was used to compress a database into a tree structure which stored only large items. It, however, needed to process all transactions in a batch way. In real-world applications, new transactions are usually inserted into databases. In this paper, we thus attempt to modify the FP-tree construction algorithm for efficiently handling new transactions. A fast updated FP-tree (FUFP-tree) structure is proposed, which makes the tree update process become easier. An incremental FUFP-tree maintenance algorithm is also proposed for reducing the execution time in reconstructing the tree when new transactions are inserted. Experimental results also show that the proposed FUFP-tree maintenance algorithm runs faster than the batch FP-tree construction algorithm for handling new transactions and generates nearly the same tree structure as the FP-tree algorithm. The proposed approach can thus achieve a good trade-off between execution time and tree complexity.

Fuzzy data mining for interesting generalized association rules

Due to the increasing use of very large databases and data warehouses, mining useful information and helpful knowledge from transactions is evolving into an important research area. Most conventional data-mining algorithms identify the relationships among transactions using binary values and find rules at a single concept level. Transactions with quantitative values and items with hierarchy relation are, however, commonly seen in real-world applications. In this paper, we thus introduce the problem of mining fuzzy generalized association rules from quantitative data. A fuzzy mining algorithm based on Srikant and Agrawals method is proposed for extracting implicit generalized knowledge from transactions stored as quantitative values. It integrates fuzzy-set concepts and generalized data mining technologies to achieve this purpose. Items in rules may be from any level of the given taxonomy. The effect of numbers of fuzzy regions on the performance of the proposed algorithm is also discussed.

Mining association rules with multiple minimum supports using maximum constraints

Abstract Data mining is the process of extracting desirable knowledge or interesting patterns from existing databases for specific purposes. Most of the previous approaches set a single minimum support threshold for all the items or itemsets. But in real applications, different items may have different criteria to judge its importance. The support requirements should then vary with different items. In this paper, we provide another point of view about defining the minimum supports of itemsets when items have different minimum supports. The maximum constraint is used, which is well explained and may be suitable to some mining domains. We then propose a simple algorithm based on the Apriori approach to find the large-itemsets and association rules under this constraint. The proposed algorithm is easy and efficient when compared to Wang et al.’s under the maximum constraint. The numbers of association rules and large itemsets obtained by the proposed mining algorithm using the maximum constraint are also less than those using the minimum constraint. Whether to adopt the proposed approach thus depends on the requirements of mining problems. Besides, the granular computing technique of bit strings is used to speed up the proposed data mining algorithm.

Processing individual fuzzy attributes for fuzzy rule induction

Fuzzy systems that can automatically derive fuzzy if–then rules and membership functions from numeric data have recently been developed. In this paper, we propose two new fuzzy learning methods for automatically deriving membership functions and fuzzy if–then rules from a set of given training examples. The proposed methods first select relevant attributes and build appropriate initial membership functions. They then simplify the intervals and the membership functions of each attribute before forming a decision table. These attributes and membership functions are then used in a decision table to derive the final fuzzy if–then rules and membership functions. Experimental results for the Iris data show that our methods can achieve a high degree of accuracy. The proposed methods are thus useful in constructing membership functions and in managing uncertainty and vagueness. They can also reduce the time and effort needed to develop a fuzzy knowledge base.

An effective tree structure for mining high utility itemsets

Research highlights? In this paper, the high utility pattern tree (HUP tree) is designed. ? The HUP-growth mining algorithm is proposed to derive high utility patterns effectively and efficiently. ? The proposed approach integrates the previous two-phase procedure for utility mining and the FP-tree concept to utilize the downward-closure property and generate a compressed tree structure. In the past, many algorithms were proposed to mine association rules, most of which were based on item frequency values. Considering a customer may buy many copies of an item and each item may have different profits, mining frequent patterns from a traditional database is not suitable for some real-world applications. Utility mining was thus proposed to consider costs, profits and other measures according to user preference. In this paper, the high utility pattern tree (HUP tree) is designed and the HUP-growth mining algorithm is proposed to derive high utility patterns effectively and efficiently. The proposed approach integrates the previous two-phase procedure for utility mining and the FP-tree concept to utilize the downward-closure property and generate a compressed tree structure. Experimental results also show that the proposed approach has a better performance than Liu et al.s two-phase algorithm in execution time. At last, the numbers of tree nodes generated from three different item ordering methods are also compared, with results showing that the frequency ordering produces less tree nodes than the other two.

A GA-based Fuzzy Mining Approach to Achieve a Trade-off Between Number of Rules and Suitability of Membership Functions

Data mining is most commonly used in attempts to induce association rules from transaction data. Transactions in real-world applications, however, usually consist of quantitative values. This paper thus proposes a fuzzy data-mining algorithm for extracting both association rules and membership functions from quantitative transactions. We present a GA-based framework for finding membership functions suitable for mining problems and then use the final best set of membership functions to mine fuzzy association rules. The fitness of each chromosome is evaluated by the number of large 1-itemsets generated from part of the previously proposed fuzzy mining algorithm and by the suitability of the membership functions. Experimental results also show the effectiveness of the framework.