Mark-A. Krogel

KDD Cup 2001 report

This paper presents results and lessons from KDD Cup 2001. KDD Cup 2001 focused on mining biological databases. It involved three cutting-edge tasks related to drug design and genomics.

Comparative Evaluation of Approaches to Propositionalization

Propositionalization has already been shown to be a promising approach for robustly and effectively handling relational data sets for knowledge discovery. In this paper, we compare up-to-date methods for propositionalization from two main groups: logic-oriented and database-oriented techniques. Experiments using several learning tasks – both ILP benchmarks and tasks from recent international data mining competitions – show that both groups have their specific advantages. While logic-oriented methods can handle complex background knowledge and provide expressive first-order models, database-oriented methods can be more efficient especially on larger data sets. Obtained accuracies vary such that a combination of the features produced by both groups seems a further valuable venture.

Multi-Relational Learning, Text Mining, and Semi-Supervised Learning for Functional Genomics

We focus on the problem of predicting functional properties of the proteins corresponding to genes in the yeast genome. Our goal is to study the effectiveness of approaches that utilize all data sources that are available in this problem setting, including relational data, abstracts of research papers, and unlabeled data. We investigate a propositionalization approach which uses relational gene interaction data. We study the benefit of text classification and information extraction for utilizing a collection of scientific abstracts. We study transduction and co-training for using unlabeled data. We report on both, positive and negative results on the investigated approaches. The studied tasks are KDD Cup tasks of 2001 and 2002. The solutions which we describe achieved the highest score for task 2 in 2001, the fourth rank for task 3 in 2001, the highest score for one of the two subtasks and the third place for the overall task 2 in 2002.

Transformation-Based Learning Using Multirelational Aggregation

Given the very widespread use of multirelational databases, ILP systems are increasingly being used on data originating from such warehouses. Unfortunately, even though not complex in structure, such business data often contain highly non-determinate components, making them difficult for ILP learners geared towards structurally complex tasks. In this paper, we build on popular transformation-based approaches to ILP and describe how they can naturally be extended with relational aggregation. We experimentall y show that this results in a multirelational learner that outperforms a structurally-oriented ILP system both in speed and accuracy on this class of problems.

Combining data and text mining techniques for yeast gene regulation prediction: a case study

In order to solve task 2 of the KDD Cup 2002, we exploited various available information sources. In particular, use of relational information describing the interactions among genes and information automatically extracted from scientific abstracts improves the accuracy of our predictions.

Feature Selection for Propositionalization

Following the success of inductive logic programming on structurally complex but small problems, recently there has been strong interest in relational methods that scale to real-world databases. Propositionalization has already been shown to be a particularly promising approach for robustly and effectively handling larger relational data sets. However, the number of propositional features generated here tends to quickly increase, e.g. with the number of relations, with negative effects especially for the efficiency of learning. In this paper, we show that feature selection techniques can significantly increase the efficiency of transformation-based learning without sacrificing accuracy.

Effectiveness of information extraction, multi-relational, and semi-supervised learning for predicting functional properties of genes

We focus on the problem of predicting functional properties of the proteins corresponding to genes in the yeast genome. Our goal is to study the effectiveness of approaches that utilize all data sources that are available in this problem setting, including unlabeled and relational data, and abstracts of research papers. We study transduction and co-training for using unlabeled data. We investigate a propositionalization approach which uses relational gene interaction data. We study the benefit of information extraction for utilizing a collection of scientific abstracts. The studied tasks are KDD Cup tasks of 2001 and 2002. The solutions which we describe achieved the highest score for task 2 in 2001, the fourth rank for task 3 in 2001, the highest score for one of the two subtasks and the third place for the overall task 2 in 2002.