This method extracts some sentences that have higher weight and builds text relationship map according to similarity computation of sentences,and then uses the algorithm of community partition in complex networks to solve the problem of sub-topic partition.
抽取文本中权重较高的句子,通过句子的相似度计算建立文本关系图,利用社区划分算法解决子主题划分的问题。
参考来源 - 多主题文本摘要抽取的研究与实现·2,447,543篇论文数据，部分数据来源于NoteExpress
系统内用户的状态可基于当前主题/子主题、主题表现等定义。
The state of a user can be defined based on the current topic/sub-topic, performance in topics, and so on.
我们尽可能提供相关详细信息的适当参考资料，让您能够进一步研究相关的子主题。
Wherever possible, references to appropriate resources are provided so that you can further explore related subtopics.
如何正确有效地确定文档的子主题边界对于自动文摘、问答系统等自然语言处理应用是非常重要的。
How to subdivide text into multi-paragraph units is an important issue in many NLP applications such as automatic summarization and QA system.
应用推荐