Abstract: Open access journal is a kind of deep online resources and disperses on the Internet, and it is difficult for the traditional search engines to index these online resources, so the user can not access directly the open access journal via search engines, resulting in a waste of these open resources. This paper proposes a novel focused Web crawler with distributed architecture to collect the open access journal resources scattering throughout the Internet. This architecture adopts the distributed master-slave design, which consists of a master control center and multiple distributed crawler nodes, and proposes an academic information extraction method based on user predefined rules from the open access journals. These distributed crawling nodes can be adjusted dynamically and use Chrome browser based plug-in mechanism to achieve scalability and deployment flexibility.

Key words: distributed Web crawler, open access journal, plug-in mechanism

摘要： 开放存取（open access，OA）期刊属于网络深层资源且分散在互联网中，传统的搜索引擎不能对其建立索引，不能满足用户获取OA期刊资源的需求，从而造成了开放资源的浪费。针对如何集中采集万维网上分散的开放存取期刊资源的问题，提出了一个面向OA期刊的分布式主题爬虫架构。该架构采用主从分布式设计，提出了基于用户预定义规则的OA期刊页面学术信息提取方法，由一个主控中心节点控制多个可动态增减的爬行节点，采用基于Chrome浏览器的插件机制来实现分布式爬行节点的可扩展性和部署的灵活性。

关键词: 分布式爬虫, 开放存取期刊, 插件机制