Mining and Using Sets of Patterns through Compression

Int. J. Artif. Intell. Tools

Nijssen

et al. 2014

Pattern mining provides useful tools for exploratory data analysis. Numerous efficient algorithms exist that are able to discover various types of patterns in large datasets. Unfortunately, the problem of identifying patterns that are genuinely interesting to a particular user remains challenging. Current approaches generally require considerable data mining expertise or effort from the data analyst, and hence cannot be used by typical domain experts.To address this, we introduce a generic framework for interactive learning of userspecific pattern ranking functions. The user is only asked to rank small sets of patterns, while a ranking function is inferred from this feedback by preference learning techniques. Moreover, we propose a number of active learning heuristics to minimize the effort required from the user, while ensuring that accurate rankings are obtained. We show how the learned ranking functions can be used to mine new, more interesting patterns.We demonstrate two concrete instances of our framework for two different pattern mining tasks, frequent itemset mining and subgroup discovery. We empirically evaluate the capacity of the algorithm to learn pattern rankings by emulating users. Experiments demonstrate that the system is able to learn accurate rankings, and that the active learning heuristics help reduce the required user effort. Furthermore, using the learned ranking functions as search heuristics allows discovering patterns of higher quality than those in the initial set. This shows that machine learning techniques in general, and active preference learning in particular, are promising building blocks for interactive data mining systems.

Section: Introductionmentioning

confidence: 99%

Interactive Learning of Pattern Rankings

Dzyuba

Int. J. Artif. Intell. Tools

Nijssen

et al. 2014

“…We formalize this problem using the minimum description length (MDL) principle [4], which, informally, states that the best model is the one that compresses the data best. The MDL principle perfectly fits our purposes because (1) it allows to select the simplest model that adequately explains the data, and (2) it has been previously shown to be very effective for the selection of pattern-based models (e.g., [7,11]).…”

Section: Introductionmentioning

confidence: 77%

“…Let us now briefly introduce the basic notions of the minimum description length (MDL) principle [4] as it is commonly used in compression-based pattern mining [7]. Given a set of models M and a dataset D, the best model M ∈ M is the one that minimizes L(D, M) = L(M ) + L(D|M ), with L(M ) the length, in bits, of the encoding of M , and L(D|M ) the length, in bits, of the encoding of the data given M .…”

Section: Minimum Description Length (Mdl)mentioning

confidence: 99%

“…As typically done in compression-based pattern mining [7], we use a two-part MDL code that leads to decomposing the total encoded length L(α, S) into two parts: L(S) and L(α|S), with the relation L(α, S) = L(S) + L(α|S). In the upcoming subsection we define L(S), i.e., the encoded length of a signature, after which Subsect.…”

Section: An Encoding For Signaturesmentioning

confidence: 99%

See 1 more Smart Citation

Widening for MDL-Based Retail Signature Discovery

Gautrais

Cellier

Lecture Notes in Computer Science

et al. 2020

Self Cite

Signature patterns have been introduced to model repetitive behavior, e.g., of customers repeatedly buying the same set of products in consecutive time periods. A disadvantage of existing approaches to signature discovery, however, is that the required number of occurrences of a signature needs to be manually chosen. To address this limitation, we formalize the problem of selecting the best signature using the minimum description length (MDL) principle. To this end, we propose an encoding for signature models and for any data stream given such a signature model. As finding the MDL-optimal solution is unfeasible, we propose a novel algorithm that is an instance of widening, i.e., a diversified beam search that heuristically explores promising parts of the search space. Finally, we demonstrate the effectiveness of the problem formalization and the algorithm on a real-world retail dataset, and show that our approach yields relevant signatures.

“…In contrast to existing pattern-based modeling approaches (e.g., [42,41]), we deal with a supervised setting in which the goal is to learn a mapping from instances to class labels. This implies that we are not looking for structure within instance data X, but for structure in X that helps to predict Y .…”

Section: For Multiclass Classificationmentioning

confidence: 99%

Interpretable multiclass classification by MDL-based rule lists

Proença

2020

Information Sciences

Self Cite

Interpretable classifiers have recently witnessed an increase in attention from the data mining community because they are inherently easier to understand and explain than their more complex counterparts. Examples of interpretable classification models include decision trees, rule sets, and rule lists. Learning such models often involves optimizing hyperparameters, which typically requires substantial amounts of data and may result in relatively large models.In this paper, we consider the problem of learning compact yet accurate probabilistic rule lists for multiclass classification. Specifically, we propose a novel formalization based on probabilistic rule lists and the minimum description length (MDL) principle. This results in virtually parameter-free model selection that naturally allows to trade-off model complexity with goodness of fit, by which overfitting and the need for hyperparameter tuning are effectively avoided. Finally, we introduce the Classy algorithm, which greedily finds rule lists according to the proposed criterion.We empirically demonstrate that Classy selects small probabilistic rule lists that outperform state-of-the-art classifiers when it comes to the combination of predictive performance and interpretability. We show that Classy is insensitive to its only parameter, i.e., the candidate set, and that compression on the training set correlates with classification performance, validating our MDL-based selection criterion.