Multi-Strategy to Improve Reading Comprehension

Reading comprehension is a task that is sufficiently close to information extraction applications such as question answering. A reading comprehension (RC) system attempts to understand a document and returns an answer sentence when posed with a question. This paper proposes an RC system which utilizes different techniques, including pattern matching, semantic relation inference and the context assistance rule set. This approach gave improved RC performances for the Remedia corpus. We give the deep performance analysis across the experimental results, such as the performance impact by different techniques, performance impact by the technique of co-reference resolution, and we also ran pairwise t-tests that show the significance on performance improvements.