• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Collaboration Technologies and Systems >Argument placement recommendation and relevancy assessment in an intelligent argumentation system
【24h】

Argument placement recommendation and relevancy assessment in an intelligent argumentation system

机译：智能论证系统中的论据放置推荐和相关性评估

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Argumentation is a critical process for many social activities that need collaborative intelligence. Existing intelligent argumentation systems allow multiple stakeholders from distributed geographical locations to share their opinions and contribute to a decision making process. In the current system, a stakeholder needs to read all the existing arguments posted by other stakeholders before contributing his/her own ideas/arguments. However, when information accumulates and an argumentation network becomes considerably large, it will cost tremendous time and effort for the stakeholder to read and comprehend all existing arguments. In this paper, we propose methods to implement a recommendation component built into an intelligent argumentation system. The recommendation component can automatically assist a stakeholder to better understand the current state of an argumentation by summarizing and identifying a subset of existing arguments that are relevant to the stakeholder. Thus, our proposed work will allow a stakeholder to efficiently and effectively express his/her thoughts in an intelligent argumentation system in relevant argumentation thread. We also empirically evaluate the effectiveness of the proposed recommendation component, and the empirical results indicate it is effective according to a real dataset.
机译：争论是许多需要协作情报的社交活动的关键过程。现有的智能论证系统允许来自分散地理位置的多个利益相关者分享他们的观点并有助于决策过程。在当前系统中，利益相关者需要先阅读其他利益相关者发表的所有现有论点，然后再提出自己的想法/观点。但是，当信息积累并且争论网络变得相当大时，利益相关者阅读和理解所有现有的争论将花费大量的时间和精力。在本文中，我们提出了实现内置在智能论证系统中的推荐组件的方法。推荐组件可以通过汇总和标识与利益相关者相关的现有论点的子集，自动帮助利益相关者更好地理解论点的当前状态。因此，我们提出的工作将使利益相关者能够在相关论证线程的智能论证系统中有效地表达自己的想法。我们还根据经验评估了建议的推荐组件的有效性，而经验结果表明，根据实际数据集，该推荐组件是有效的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号