TY - CONF
T1 - Predictive Graph Mining
AU - Karwath, Andreas
AU - De Raedt, Luc
PY - 2004
Y1 - 2004
N2 - Graph mining approaches are extremely popular and effective in molecular databases. The vast majority of these approaches first derive interesting, i.e. frequent, patterns and then use these as features to build predictive models. Rather than building these models in a two step indirect way, the SMIREP system introduced in this paper, derives predictive rule models from molecular data directly. SMIREP combines the SMILES and SMARTS representation languages that are popular in computational chemistry with the IREP rule-learning algorithm by Fürnkranz. Even though SMIREP is focused on SMILES, its principles are also applicable to graph mining problems in other domains. SMIREP is experimentally evaluated on two benchmark databases.
AB - Graph mining approaches are extremely popular and effective in molecular databases. The vast majority of these approaches first derive interesting, i.e. frequent, patterns and then use these as features to build predictive models. Rather than building these models in a two step indirect way, the SMIREP system introduced in this paper, derives predictive rule models from molecular data directly. SMIREP combines the SMILES and SMARTS representation languages that are popular in computational chemistry with the IREP rule-learning algorithm by Fürnkranz. Even though SMIREP is focused on SMILES, its principles are also applicable to graph mining problems in other domains. SMIREP is experimentally evaluated on two benchmark databases.
KW - cheminformatics, graph mining, machine learning, QSAR
UR - https://www.scopus.com/pages/publications/35048822976
U2 - 10.1007/978-3-540-30214-8_1
DO - 10.1007/978-3-540-30214-8_1
M3 - Paper
SP - 1
EP - 15
ER -