We are given a large database of customer transactions. Each transaction consists of items purchased by a customer in a visit. We present an efficient algorithm that generates all significant association rules between items in the database. The algorithm incorporates buffer management and novel estimation and pruning techniques. We also present results of applying this algorithm to sales data obtained from a large retailing company, which shows the effectiveness of the algorithm.

Abstract. Admissible heuristics are an important class of heuristics worth discovering: they guarantee shortest path solutions in search algorithms such as A * and they guarantee less expensively produced, but boundedly longer solutions in search algorithms such as dynamic weighting. Unfortunately, effective (accurate and cheap to com-pute) admissible heuristics can take years for people to discover. Several researchers have suggested that certain transformations of a problem can be used to generate admissible heuristics. This article defines a more general class of transformations, called abstractions, that are guaranteed to generate only admissible heuristics. It also describes and evaluates an implemented program (Absolver IO that uses a means-ends analysis search control strategy to discover abstracted problems that result in effective admissible heuristics. Absolver I/discovered several well-known and a few novel admissible heuristics, including the first known effective one for Rubik's Cube, thus concretely demonstrating that effective admissible heuristics can be tractably discovered by a machine.

Programs such as Bacon, Abacus, Coper, Kepler and others are designed to find functional relationships of scientific significance in numerical data without relying on the deep domain knowledge scientists normally bring to bear in analytic work. Whether these systems actually perform as intended is an open question, however. To date, they have been supported only by anecdotal evidence---reports that a desirable answer has been found in one or more handselected and often artificial cases. In this paper, I describe a function-finding algorithm which differs radically from previous candidates in three respects. First, it concentrates rather on reliable identification of a few functional forms than on heuristic search of an infinite space of potential relations. Second, it introduces the use of distinction, significance and lack of fit--- three general concepts of value in evaluating apparent functional relationships. Finally, and crucially, the algorithm has been tested prospectively on an...

This article reports the results of a study of domain-independent function finding based on a collection of several hundred real scientific data sets. Prior studies have introduced the controversial idea of discovering functional relationships of interest to scientists directly from the data they collect. The evidence presented here supports the view that this is sometimes possible, but it also suggests how often purely datadriven discovery is not possible and how much more difficult it may be than has often been assumed. Experience with sampled examples of real scientific data suggests as well that emphasis on search in prior studies may have been misplaced; for the function-finding problems studied here, scientists typically propose only a handful of different relationships. The difficulty is not in searching for relationships, but in evaluating a few principal ones to determine if they are likely to be of scientific interest. Running head: Function Finding in Real Data Keywords: Emp...

The paper presents initial results from an emerging new direction in engineering design research, in particular, creative design. It argues that constructive induction, which was originally proposed in the field of machine learning, can serve as a foundation for developing a computational theory of engineering design and design creativity. Constructive induction is a process of creating new knowledge (e.g., design knowledge) by performing two intertwined searches, one---for he most adcquale knowledge representation space, and second---for the best hypothesis in this space. Basic concepts and methods of constructive induction are reviewed and illustrated by examples of heir application to conceptual stxuctural design. Several crucial design concepts, including those of an emergent concept and of a goal-oriented Uausformation of he design represenlation space are interpreted in terms of a construction induction process. It is also shown how constructive induction applies to the conxol of the design creativity level. Several measures of the design complexity and relative creativity are proposed. The conclusion presents some lved problems and a plan for future research.