Type Extension Trees (TETs)

Type Extension Trees are a powerful representation language for "count-of-count" features characterizing the combinatorial structure of neighborhoods of entities in relational domains. TETs can be used as a feature discovery instrument in relational domains, and a metric on TET features can be constructed, in order to effectively exploit their expressive power in terms of "counts-of-counts". Experiments on bibliographic data (e.g., for the prediction of the future h-index of an author) show the potentiality of such features.


Source code and data sets

The source code for working with TETs, together with some tutorial examples, can be downloaded here.
A JVM and a MySQL data base are required. Some bash scripts for running tutorial examples are included.
The archive also contains the DBLP data set version which has been used in the experiments of our paper published on the Artificial Intelligence Journal (see below): the SQL data set alone can also be downloaded directly here.


References

• Jaeger, M., Lippi, M., Passerini, A., Frasconi, P.,
  Type Extension Trees for feature construction and learning in relational domains
  Artificial Intelligence, in press, 2013
  PDF