MOOC统一检索平台的自动构建与应用

随着MOOC平台增多以及同一平台下学习资源剧增,如何实现跨平台的高效语义检索成为目前MOOC亟待解决的问题之一.该文通过网络爬虫工具获取多个知名MOOC平台的学习资源数据,进行相关预处理后存储到Mysql数据库,并根据数据库与本体之间的映射关系自动构建MOOC本体,使用Jena将MOOC本体解析成RDF 3元组,并将RDF 3元组存储至HBase数据库,最终构建出一个MOOC统一检索平台,为学习者推荐符合其检索需求的学习资源.实验结果表明:构建的MOOC统一检索平台可有效地提高检索的查准率和查全率.

The Automatic Construction and Application of MOOC Unified Search Platform

With the increase of MOOC platforms and the proliferation of learning resources under the same platform,how to achieve efficient cross-platform semantic retrieval has become one of the issues that MOOC need to solve urgently.Learning resource data of multiple well-known MOOC platforms is obtained through web crawlers,relevant data is preprocessed and stored into Mysql database,MOOC ontology is automatically constructed according to the mapping relationship between database and ontology,Jena is used to parse the MOOC ontology into RDF triples and the RDF triples are stored to the HBase database,and finally a MOOC unified search platform is built to recommend learning resources for the learners to meet their search requirements.The experimental results show that the constructed semantic MOOC platform can effectively improve the search accuracy and recall rate.