A Machine Learning Framework for Automatically Annotating Web Pages with Simple HTML Ontology Extension (SHOE)

摘要: With enormous amounts of information injected into the Internet every second, manual maintenance knowledge base on is a hopeless task. A reasonable remedy for this problem to create “machine understandable” Internet. To achieve this, Heflin et al. proposed an HTML-based representation language called Simple HTML Ontology Extension (SHOE). SHOE can be used in many application domains, but it requires users manually annotate web pages. overcome shortages SHOE, we created machine learning framework AutoSHOE automatically annotating pages with annotations. framework, easily collect SHOE-annotated as training data, experiment different feature selection methods and algorithms find best approach particular ontology, new trained classifiers rule sets. In addition, allows selectors learners plugged system run anywhere through web. We present architecture then discuss experimental results our proof-of-concept design.