Improving on Q & A Recurrent Neural Networks Using Noun-Tagging
This work addresses incremental improvements in question answering models for researchers, focusing on dataset preprocessing and tuning rather than novel methods.
The researchers tackled the problem of improving a simple Recurrent Neural Network for question answering by fine-tuning hyperparameters and using noun-tagging preprocessing, achieving an entity detection accuracy of 0.984 with a 1-layer LSTM, though relation prediction accuracy remained lower at 0.80.
Often, more time is spent on finding a model that works well, rather than tuning the model and working directly with the dataset. Our research began as an attempt to improve upon a simple Recurrent Neural Network for answering "simple" first-order questions (QA-RNN), developed by Ferhan Ture and Oliver Jojic, from Comcast Labs, using the SimpleQuestions dataset. Their baseline model, a bidirectional, 2-layer LSTM RNN and a GRU RNN, have accuracies of 0.94 and 0.90, for entity detection and relation prediction, respectively. We fine tuned these models by doing substantial hyper-parameter tuning, getting resulting accuracies of 0.70 and 0.80, for entity detection and relation prediction, respectively. An accuracy of 0.984 was obtained on entity detection using a 1-layer LSTM, where preprocessing was done by removing all words not part of a noun chunk from the question. 100% of the dataset was available for relation prediction, but only 20% of the dataset, was available for entity detection, which we believe to be much of the reason for our initial difficulties in replicating their result, despite the fact we were able to improve on their entity detection results.