Maria Grineva, Maxim Grinev, and Dmitry Lizorkin
We present a novel method for extraction of key terms from text documents. The important and novel feature of our method is that it produces groups of key terms, while each group contains key terms semantically related to one of the main themes of the document. Our method bases on a combination of the following two techniques: Wikipedia-based semantic relatedness measure of terms and algorithm for detecting community structure of a network. One of the advantages of our method is that it does not require any training, as it works upon the Wikipedia knowledge base. Our experimental evaluation using human judgments shows that our method produces key terms with high precision and recall.