Anytime discovery of a diverse set of patterns with Monte Carlo tree search

Abstract

The discovery of patterns that accurately discriminate one class label from another remains a challenging data mining task. Subgroup discovery (SD) is one of the frameworks that enables to elicit such interesting patterns from labeled data. A question remains fairly open: How to select an accurate heuristic search technique when exhaustive enumeration of the pattern space is infeasible? Existing approaches make use of beam-search, sampling, and genetic algorithms for discovering a pattern set that is nonredundant and of high quality w.r.t. a pattern quality measure. We argue that such approaches produce pattern sets that lack of diversity: Only few patterns of high quality, and different enough, are discovered. Our main contribution is then to formally define pattern mining as a game and to solve it with Monte Carlo tree search (MCTS). It can be seen as an exhaustive search guided by random simulations which can be stopped early (limited budget) by virtue of its best-first search property. We show through a comprehensive set of experiments how MCTS enables the anytime discovery of a diverse pattern set of high quality. It outperforms other approaches when dealing with a large pattern search space and for different quality measures. Thanks to its genericity, our MCTS approach can be used for SD but also for many other pattern mining tasks.

Mining Frequent Patterns in 2D+t Grid Graphs for Cellular Automata Analysis

“…This may come from the fact that deep patterns are not diverse enough to be selected by the post-processing step which in turn suggests that, when the timeout is reached, the diversity of the mined pattern is not high enough. To further increase this diversity, stochastic search methods such as Monte-Carlo Tree Search [2] could be integrated in our algorithm.…”

Actionable Subgroup Discovery and Urban Farm Optimization

“…For instance, in Fig. 1, the domain of m 1 is {1, 2, 3, 4} and the intent [2,4], [1,3] (see the definition of interval patterns later) denotes a subgroup whose extent is {g 3 , g 4 , g 5 , g 6 }.…”

“…Heuristic [2,15] and exhaustive [1,5] solutions have been proposed for subgroup discovery. Usually, these approaches consider a set of nominal attributes with a binary label.…”

Optimal Subgroup Discovery in Purely Numerical Data

“…A quality measure has to capture discrepancies in the target label distribution between the selected subset of objects and the overall dataset. A large panel of exhaustive [1,10] and heuristic [5,16] subgroup discovery algorithms have been proposed so far. Most of these approaches consider a set of nominal attributes with a binary label.…”