296
Asian Journal of Engineering, Sciences and Technology - Volume 8, Issue 1 2018
By Muhammad Suleman, Mohammad Zoaib, Hussain Shabbir, Hatim Ali Asghar, Humdiya Raza, Tahir.Q.Syed
Keywords: QA systems, Deep Learning, Literature Review, Cognitive Systems, Information Retrieval.
Building Question Answering (QA) systems is a very important task in Information Retrieval (IR) / Natural Language Processing (NLP) domain. In IR / NLP domain there are many tasks which are similar which means solution of one task can be used to solve another task. These tasks include building QA systems, paraphrase detection, semantic similarity between sentences/words, semantic entailment, machine comprehension, slot filling and other like tasks. We found that many of these tasks are tackled in research using different techniques and different datasets. We also found that although standardized but most datasets are very small and they cannot solve generalize semantic assignment problem which lies at the heart of all these problems. Recently new dataset are published with larger size in the hope that they will be useful in building Generalize QA system. We do systematic literature review (SLR) of almost all the papers from 2010 to 2012 related to above mentioned problems. We try to find the correct direction for building a Generalize QA system that will be helpful in QA on any open domain corpus/dataset. We extract all the techniques / features / dataset / evaluation metrics / state of the art results published by different papers developing QA systems or doing like tasks. In the light of this statistics we answer some hypothesized question which we think are really helpful in building Generalize QA systems. For SLR we apply procedure as defined in [38].