Traditionally, Question Answering (QA) has been an important task in the field of Artificial Intelligence (AI) and a substantial amount of research has been done in the past proposing several methods to answer questions in an automated way. The process of QA has been viewed differently by different researchers in the past. Till date, the current / existing systems cannot answer elementary science questions which are obvious and easy to answer even for a primary school student. This paper documents various methodologies proposed in the research area of QA specifically for elementary science domain and includes the discussion of each technique used by various methodologies. We briefly present the overview of three components of generalized QA system and discuss various metrics used for evaluating the QA systems. The paper also presents an exhaustive review of various datasets used for the purpose with their respective description and details. We conclude the paper by identifying the limitations of the methodologies used till date and pointing out some future directions for research in the arena of science QA.

传统上，问答 (QA) 一直是人工智能 (AI) 领域的一项重要任务，过去进行了大量研究，提出了几种自动回答问题的方法。过去，不同的研究人员对 QA 过程有不同的看法。迄今为止，当前/现有系统无法回答即使是小学生也很容易回答的基础科学问题。本文记录了专门针对基础科学领域的质量保证研究领域提出的各种方法，包括对各种方法使用的每种技术的讨论。我们简要介绍了通用 QA 系统的三个组成部分的概述，并讨论了用于评估 QA 系统的各种指标。该论文还详尽地回顾了用于该目的的各种数据集及其各自的描述和细节。我们通过确定迄今为止使用的方法的局限性并指出科学 QA 领域的一些未来研究方向来总结本文。