Learning Relational Bayesian Classifiers from RDF Data

Abstract

Abstract. The increasing availability of large RDF datasets offers an exciting opportunity to use such data to build predictive models using machine learning algorithms. However, the massive size and distributed nature of RDF data calls for approaches to learning from RDF data in a setting where the data can be accessed only through a query interface, e.g., the SPARQL endpoint of the RDF store. In applications where the data are subject to frequent updates, there is a need for algorithms that allow the predictive model to be incrementally updated in response to changes in the data. Furthermore, in some applications, the attributes that are relevant for specific prediction tasks are not known a priori and hence need to be discovered by the algorithm. We present an approach to learning Relational Bayesian Classifiers (RBCs) from RDF data that addresses such scenarios. Specifically, we show how to build RBCs from RDF data using statistical queries through the SPARQL endpoint of the RDF store. We compare the communication complexity of our algorithm with one that requires direct centralized access to the data and hence has to retrieve the entire RDF dataset from the remote location for processing. We establish the conditions under which the RBC models can be incrementally updated in response to addition or deletion of RDF data. We show how our approach can be extended to the setting where the attributes that are relevant for prediction are not known a priori, by selectively crawling the RDF data for attributes of interest. We provide open source implementation and evaluate the proposed approach on several large RDF datasets.

Unsupervised generation of data mining features from linked open data

“…However, generating such features leads to an unmanageable explosion of the feature space. Our approach, like other state-of-the-art approaches [29], foresees fully calculating the metric of each attribute while it is being examined. A more scalable approach would have to rely on quicker approximate estimates of a feature's prospective value instead, performing an on-the-fly selection of features and interesting paths to follow.…”

Learning Classifiers from Distributional Data

“…For example, relational Bayesian classifiers [24], [25] model an object (nodes in a network) to be classified using bags of values of features from those objects that are related to it via relational links.…”

Learning Classifiers from Chains of Multiple Interlinked RDF Data Stores

“…If naive Bayes is learned on the reduced data set then the following is sufficient: number of instances with the class label c and d = agg(B i k ) for every combination of c and d. The former can be expressed by an aggregation query and the later is equivalent to S(G, T , C = c, A k , agg, d) in [10].…”

“…, A K ) a tuple of attributes, and A c is a target attribute. Given an RDF data set D = (G, T , A, A c ), its induced multiset attributed data set [10] is defined as An RDF Learner L [10] is an algorithm that given an RDF data set D = (G, T , A, A c ), its induced multiset attributed data set M(D), a hypothesis class H, and a performance criterion P , outputs a classifier h ∈ H that optimizes P .…”

“…via a query interface such as SPARQL). Barring Lin et al [10] who proposed an approach to learning relational Bayesian classifiers [11] from a single remote RDF store using statistical queries against its SPARQL endpoint, to the best of our knowledge, there has been very little work on this problem.…”