Recently, relation classification has gained much success by exploiting deep neural networks. However, it is a lack of an effective model exploiting the unique advantages of both Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), which is suitable for Relation Extraction Task.
In this thesis, we propose a new model effectively combining Segment-level Attention-based Convolutional Neural Networks (SACNNs) and Dependencybased Recurrent Neural Networks (DepRNNs). While the SACNNs allow the model to selectively focus on the important information segment from the raw
sequence, the DepRNNs help to handle the long-distance relations from the shortest dependency path of the related entities. Experiments on the SemEval-2010 Task 8 dataset show that our model is comparable to the state-of-the-art without using any external lexical features.