个性化知识的表示方法

Representation method of personalized knowledge

在当前信息暴涨的时代,网络信息正在面临着各取所需的要求,信息检索、话题检测、信息推荐等应用技术都逐渐开始面向个性化的发展趋势。然而目前的个性化技术大都依赖于对用户行为的了解,根据用户的历史行为,判断和预测用户的目的,没有同用户的当前所具有的知识结合起来。提出一种用户个性化知识的粗略表示方法——词形关系图,作为个性化应用技术的基础。具体内容包括:词形关系图表示知识的方式,结合遗忘规律从用户语料库中获取个性化词形关联的方法,以及结合实验结果对该表示方法应用可行性的初步分析。

Now, along with the sudden increase of the amount of network information, the personalization is becoming the critical component of information retrieval, topic detection, and information recommendation. However, most of the current personalization technologies rely on user behavior, which is based on the historical actions of the user to determine and predict the user＇s destination, not combining with a user＇s knowledge. This paper presents a method of rough knowledge representation for individual users-word relationship graph as the basis for personalized applications. Topics include the method of knowledge representation using word relationship graph, the acquirement of word relationship graph combined the forget law from the user corpus, and the experimental results analysis for the preliminary feasibility of this representation.