TIQA: Technical Interview Question Answering dataset

This is a Question Answering dataset comprising a list of technical interview-related questions and answers derived from two textbooks, one for Machine Learning and one for Deep Learning subjects. Each question-answer (QA) pair is accompanied by the subject (Machine Learning or Deep Learning) of the textbook and the specific context (section content from the textbook) from which the QA pair was curated. It also contain the Type information of each question.The dataset is available in JSON lines format, with each line representing a JSON object. Each JSON object includes the following keys: Document, Question, Question Type, Subject, and Answer. The Document key contains the context paragraph(s) from the textbook in the form of a list of sentences. The Question key contains the specific question, Question Type contains the type of the Question, and the Subject key indicates the subject of the textbook, i.e., Machine Learning or Deep Learning. The Answer key includes the indexes of the context sentences that form the answer to the question. Note that the first sentence of the context has an index of 0.

本数据集为问答（Question Answering）数据集，收录了源自两本学科教材的技术面试类问答对，两本教材分别对应机器学习（Machine Learning）与深度学习（Deep Learning）领域。每一组问答（Question Answering，QA）对均附带其来源教材的学科归属、该问答对所提取的教材特定上下文（即教材章节内容），同时该数据集还包含每道问题的题型信息。
本数据集采用JSON Lines格式存储，每一行对应一个独立的JSON对象。每个JSON对象均包含以下字段：Document、Question、Question Type、Subject与Answer。其中，Document字段以句子列表的形式存储来源教材的上下文段落内容；Question字段为具体的问题表述；Question Type字段用于标注问题的题型；Subject字段用于标识来源教材的学科归属，即机器学习或深度学习；Answer字段包含构成该问题答案的上下文句子的索引。请注意，上下文的第一句索引为0。