Multitask Learning with Deep Neural Networks for Community Question Answering

摘要

In this paper, we developed a deep neural network (DNN) that learns to solvesimultaneously the three tasks of the cQA challenge proposed by theSemEval-2016 Task 3, i.e., question-comment similarity, question-questionsimilarity and new question-comment similarity. The latter is the main task,which can exploit the previous two for achieving better results. Our DNN istrained jointly on all the three cQA tasks and learns to encode questions andcomments into a single vector representation shared across the multiple tasks.The results on the official challenge test set show that our approach produceshigher accuracy and faster convergence rates than the individual neuralnetworks. Additionally, our method, which does not use any manual featureengineering, approaches the state of the art established with methods that makeheavy use of it.