面向机器学习的知识图谱与问答系统设计

Design of knowledge graph and question answering system for machine learning

作为一种结构化的语义知识库,知识图谱近年来被多个领域所应用,然而在机器学习这一专有领域仍存在空缺。文章描述了如何构建一个面向机器学习领域的知识图谱,并基于该图谱设计了一个问答系统。在图谱的构建过程中,主要使用了爬虫技术以及部分NLP方法对数据进行采集和处理,最终得到1个包含2442个实体的知识图谱,并将其存储在Neo4j图数据库中。针对问答系统设计部分,结合基于规则正则匹配以及基于词向量相似度匹配的方法,构建了问答模块。该领域图谱的构建和问答系统的设计,将使研究人员和爱好者更轻松地获取高质量的机器学习领域的知识。

As a structured semantic knowledge base,knowledge graph has been applied in many fields in recent years,but there are still vacancies in the proprietary field of machine learning.The article describes how to build a knowledge graph for the field of machine learning,and designs a question answering system based on the graph.During the construction of the graph,crawler technology and some NLP methods are mainly used to collect and process the data,and finally a knowledge graph containing 2,442 entities is obtained and stored in the Neo4j graph database.For the design part of the question answering system,a question answering module is constructed by combining the methods based on rule regular matching and word vector similarity matching.The construction of the domain map and the design of the question answering system will make it easier for researchers and enthusiasts to obtain high-quality machine learning domain knowledge.