一种基于Hadoop集群的自然语言处理平台实现方案

Implementation of Natural Language Pro-cessing Platform Based on Hadoop Cluster

摘要: 本文提出一种基于Hadoop集群的自然语言处理平台实现方案，通过网页收集用户需求和数据，自动调用集群进行处理，返回处理结果。且支持运行第三方开发者编写的Mapreduce插件，平台易扩展。通过实验验证，方案能实现及时响应用户需求，准确调用相关程序，并返回处理结果。

Abstract: This paper proposes a Hadoop-based program of natural language processing platform, collecting user needs and data via web interface, automatically calling the cluster into processing, and returning processing results. The platform supports Mapreduce function packages programmed by third-party developers, and is easy to expand. Experiments confirm the platforms ability of timely responding to user needs, accurately calling relevant programs, and returning processing results.