Model-based Exception Mining for Object-Relational Data

TL;DR

This paper extends exception mining to object-relational data using probabilistic graphical models, enabling detection of outliers based on deviations in association patterns, validated on synthetic and real-world datasets.

Contribution

It introduces a novel likelihood ratio metric for outlier detection in complex object-relational data using Bayesian networks.

Findings

Achieved superior detection accuracy over baseline methods.

Validated on synthetic, soccer, and movie datasets.

Effective in identifying exceptional objects in heterogeneous networks.

Abstract

This paper is based on a previous publication [29]. Our work extends exception mining and outlier detection to the case of object-relational data. Object-relational data represent a complex heterogeneous network [12], which comprises objects of different types, links among these objects, also of different types, and attributes of these links. This special structure prohibits a direct vectorial data representation. We follow the well-established Exceptional Model Mining framework, which leverages machine learning models for exception mining: A object is exceptional to the extent that a model learned for the object data differs from a model learned for the general population. Exceptional objects can be viewed as outliers. We apply state of-the-art probabilistic modelling techniques for object-relational data that construct a graphical model (Bayesian network), which compactly represents probabilistic associations in the data. A new metric, derived from the learned object-relational model, quantifies the extent to which the individual association pattern of a potential outlier deviates from that of the whole population. The metric is based on the likelihood ratio of two parameter vectors: One that represents the population associations, and another that represents the individual associations. Our method is validated on synthetic datasets and on real-world data sets about soccer matches and movies. Compared to baseline methods, our novel transformed likelihood ratio achieved the best detection accuracy on all datasets.