摘要
提出了基于行为语言的智能交互代理,利用关键词匹配技术获取用户需求关键信息,触发cplus2asp系统根据行为语言描述的知识及规则进行推理得出调控策略,实现智能人机交互。该智能交互代理的实现方法解决了传统智能交互引擎交互接口不友好、交互过程复杂等问题的局限性。实验结果验证了该方法的有效性。
An intelligent interaction agent based on action language is proposed.The keyword matching technology is used to obtain the key information of the users＇ requirement,and the spus2 asp system is triggered to drive the regulation strategy based on the knowledge and rules described in the action language to achieve intelligent human-computer interaction.After that,it uses its reasoning module to generate a sequence of actions for home services and convert them into equipment control instructions to eventually achieve intelligent management of the home environment.
作者
顾海军
田雅倩
崔莹
GU Hai-jun,TIAN Ya-qian,CUI Ying(College of Communication Engineering, Jilin University, Changchun 130012,China)
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2018年第5期1578-1585,共8页
Journal of Jilin University:Engineering and Technology Edition
基金
国家自然科学基金面上项目(61671219)
关键词
人工智能
智能交互代理
行为语言
知识框架
关键词匹配
artificial intelligence
intelligent interactive agent
action language
knowledge framekeyword matching
