[1]孙晓燕,朱利霞,陈杨.基于可能性条件偏好网络的交互式遗传算法及其应用[J].郑州大学学报(工学版),2017,38(06):1-5.
点击复制

基于可能性条件偏好网络的交互式遗传算法及其应用()
分享到:

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
38
期数:
2017年06期
页码:
1-5
栏目:
出版日期:
2017-11-20

文章信息/Info

Title:
Probabilistic Conditional Preference Network Assisted Interactive Genetic Algorithm and Its Application
作者:
孙晓燕朱利霞陈杨
文献标志码:
A
摘要:
根据用户实施的人机交互行为而隐式地获取用户偏好的交互式进化优化算法,可有效减轻用户疲劳,提高个性化搜索或推荐的效率. 但是,已有研究没有考虑用户交互行为和偏好的不确定性,影响了对用户偏好的拟合精度以及基于该偏好表达的进化搜索. 针对该问题,本文提出基于可能性条件偏好网络的交互式遗传算法,以刻画用户交互行为和偏好的不确定性,并提高算法的搜索性能. 首先,采用交互时间表示交互行为,考虑交互行为的不确定性,给出交互时间可信度的定义,并基于该定义,给出了用户不确定偏好的表达函数;其次,利用可信交互时间和偏好函数,定义了用户对评价对象的偏好权重,并利用该权重,设计(更新)可定量表示用户不确定偏好的可能性条件偏好网络,以更好地拟合用户偏好;然后,结合评价不确定性和可能性条件偏好网络,提出了改进的个体适应值估计策略,以更好地引导搜索. 最后,将所提算法应用于图书个性化搜索中,结果表明了算法搜索的可靠性和高效性.
Abstract:
Interactive evolutionary algorithms with user preference implicitly extracted from interactions of user are more powerful in alleviating user fatigue and improving the exploration in personalized search or recommendation. However, the uncertainties existing in user interactions and preferences have not been considered in the previous research, which will greatly impact the reliability of the extracted preference model, as well as the effective exploration of the evolution with that model. Therefore, an interactive genetic algorithm with probabilistic conditional preference networks (PCP-nets)is proposed , in which, the uncertainties are further figured out according to the interactions, and a PCP-net is designed to depict user preference model with higher accuracy by involving those uncertainties. First, the interaction time is adopted to mathematically describe the relationship between the interactions and user preference, and the reliability of the interaction time is further defined to reflect the interactive uncertainty.The preference function with evaluation uncertainty is established with the reliability of interaction time. Second, the preference weights on each interacted object are assigned on the basis of preference function and reliability. With these weights, the PCP-nets are designed and updated by involving the uncertainties into the preference model to improve the approximation. Third, a more accurate fitness function is delivered to assign fitness for the individuals. Last, the proposed algorithm is applied to a personalized book search and its superiority in exploration and feasibility is experimentally demonstrated.
更新日期/Last Update: