This paper introduces a hybrid system termed cascade adaptive resonance theory mapping (ARTMAP) that incorporates symbolic knowledge into neural-network learning and recognition. Cascade ARTMAP, a generalization of fuzzy ARTMAP, represents intermediate attributes and rule cascades of rule-based knowledge explicitly and performs multistep inferencing. A rule insertion algorithm translates if-the...

An increasing amount of behavioral and neurophysiological data suggests that the brain performs optimal (or near-optimal) probabilistic inference and learning during perception and other tasks. Although many machine learning algorithms exist that perform inference and learning in an optimal way, the complete description of how one of those algorithms (or a novel algorithm) can be implemented in...

We consider the general problem of learning multi-category classification from labeled examples. We present experimental results for a nearest neighbor algorithm which actively selects samples from different pattern classes according to a querying rule instead of the a priori class probabilities. The amount of improvement of this query-based approach over the passive batch approach depends on t...

Genetic programming (GP) usually has a wide search space and can use tree structure as its chromosome expression. So, GP may search for global optimum solution. But, in general, GP’s learning speed is not so fast. Apriori algorithm is one of algorithms for generation of association rules. It can be applied to large database. But, It is difficult to define its parameters without experience. We p...

In this paper, an algorithm is proposed based on the concept of pre-large itemsets to maintain discovered generalized association rules for record modification. A pre-large itemset is not truly large, but promises to be large in the future. A lower and an upper support threshold are used to realize this concept. The two user-specified support thresholds make the pre-large itemsets act as a gap ...

We present our experience in applying a rule induction technique to an extremely imbalanced pharmaceutical data set. We focus on using a variety of performance measures to evaluate a number of rule quality measures. We also investigate whether simply changing the distribution skew in the training data can improve predictive performance. Finally, we propose a method for adjusting the learning al...

Traditional association rule mining consider support confident measures to find out frequent item sets, it assumes all items are having equal significance. Where as weighted association rule mining assigns weights to items based on different aspects. Because researchers are more concerned with qualitative aspects of attributes (e.g. significance), as compared to considering only quantitative on...

In this paper, we provide the basic concepts about association rule mining and compared existing algorithms for association rule mining techniques. Of course, a single article cannot describe all the algorithms in detailed, yet we tried to cover the major theoretical issues, which can help the researcher in their researches. KeywordsAssociation rules, algorithm, itemsets, database.

Many algorithms to mine the association rules are divided into two stages, the first is to find the frequent set; the second is use the frequent set to generate association rules. This proposal discuss the respective characteristics and .shortcoming of the current algorithms to mine association rules and propose another method to mine faster; unlike the other algorithms, this algorithm emphasis...

Most of the association rule mining algorithm works based on the assumption that the items present in the dataset are of same kind with similar frequencies. Hence, the algorithms use levelwise support thresholds for mining. When the itemsets are of different frequency and of varied importance, the levelwise support thresholds are not suitable to discover frequent associations. Each item in a le...