Anna Atramentov

Learn More
We describe an efficient implementation (MRDTL-2) of the Multirelational decision tree learning (MRDTL) algorithm [19] which in turn was based on a proposal by Knobbe et al. [15] We describe some simple techniques for speeding up the calculation of sufficient statistics for decision trees and related hypothesis classes from multi-relational data. Because(More)
Recent advances in high throughput data acquisition, digital storage, and communications technologies have made it possible to gather very large amounts of data in many scientific and commercial domains. Much of this data resides in relational databases. Even when the data repository is not a relational database, it is often convenient to view heterogeneous(More)
We describe experiments with an implementation of Multi-relational decision tree learning (MRDTL) algorithm for induction of decision trees from relational databases using an approach proposed by Knobbe et al. [1999a]. Our results show that the performance of MRDTL is competitive with that of other algorithms for learning classifiers from multiple relations(More)
We present a general approach to speeding up a family of multi-relational data mining algorithms that construct and use selection graphs to obtain the information needed for building predictive models (e.g., decision tree classifiers) from relational database. Preliminary results of our experiments suggest that the proposed method can yield 1-2 orders of(More)
Gathering huge amounts of complex information (data and knowledge) is very common nowadays. This calls for the necessity to represent, store and manipulate complex information (e.g. detect correlations and patterns, discover explanations, construct predictive models etc.). Furthermore, being autonomously maintained, data can change in time or even change(More)
  • 1