1) Topic type
话题类型
2) topic-flow pattern
话题流类型
3) semantic types of topic
话题语义类型
4) topic classification
话题分类
1.
According to the features of topics in TDT(Topic Detection and Tracking) system which data source comes from news Web site,the concept of topic classification was proposed.
根据数据源为新闻网站的TDT(topic detection and tracking)系统的话题特点,首先提出了话题分类思想,然后将KNN(k-nearest neighbor)算法与网站频道信息相融合对话题进行分类。
5) Topic Model
话题模型
1.
In this paper,we introduce a weakly-supervised topic model to resolve class ambiguity of named entities by leveraging weak supervision from human.
本文通过引入一个弱指导话题模型,利用少量的人工指导信息,很好地解决了实体的类别模糊性,提高了挖掘的有效性。
6) Topic Kind Center(TKC)
话题类中心
1.
This method uses Topic Kind Center(TKC) to substitute topic model and uses named entities VSM to express TKC,updates TKC in topic tracking until TKC is stable.
针对传统自适应话题追踪中话题模型的不足,提出基于K-Modes聚类的自适应话题追踪方法(K-MATT方法),用话题类中心代替话题模型,把命名实体向量空间模型作为话题类中心,在追踪过程中不断迭代更新话题类中心,直到话题类中心稳定。
补充资料：话题
谈话的中心：～转了ㄧ换个～接着说。
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条