CIS at TAC Cold Start 2015: Neural Networks and Coreference Resolution for Slot Filling
This is an incremental improvement for slot filling in information extraction, targeting the TAC Cold Start evaluation.
The paper tackles the TAC Cold Start slot filling task by extending a previous system with improved coreference resolution and neural network classification, achieving rank 3 among participating systems.
This paper describes the CIS slot filling system for the TAC Cold Start evaluations 2015. It extends and improves the system we have built for the evaluation last year. This paper mainly describes the changes to our last year's system. Especially, it focuses on the coreference and classification component. For coreference, we have performed several analysis and prepared a resource to simplify our end-to-end system and improve its runtime. For classification, we propose to use neural networks. We have trained convolutional and recurrent neural networks and combined them with traditional evaluation methods, namely patterns and support vector machines. Our runs for the 2015 evaluation have been designed to directly assess the effect of each network on the end-to-end performance of the system. The CIS system achieved rank 3 of all slot filling systems participating in the task.