王成舜, 陈毓芬, 郑束蕾. 顾及眼动数据的网络地图点状符号用户兴趣分析方法[J]. 武汉大学学报 ( 信息科学版), 2018, 43(9): 1429-1437. DOI: 10.13203/j.whugis20160372
引用本文: 王成舜, 陈毓芬, 郑束蕾. 顾及眼动数据的网络地图点状符号用户兴趣分析方法[J]. 武汉大学学报 ( 信息科学版), 2018, 43(9): 1429-1437. DOI: 10.13203/j.whugis20160372
WANG Chengshun, CHEN Yufen, ZHENG Shulei. User Interest Analysis Method of Web Map Point Symbol Considering Eye Movement Data[J]. Geomatics and Information Science of Wuhan University, 2018, 43(9): 1429-1437. DOI: 10.13203/j.whugis20160372
Citation: WANG Chengshun, CHEN Yufen, ZHENG Shulei. User Interest Analysis Method of Web Map Point Symbol Considering Eye Movement Data[J]. Geomatics and Information Science of Wuhan University, 2018, 43(9): 1429-1437. DOI: 10.13203/j.whugis20160372

顾及眼动数据的网络地图点状符号用户兴趣分析方法

User Interest Analysis Method of Web Map Point Symbol Considering Eye Movement Data

  • 摘要: 为解决网络地图个性化推荐过程中点状符号用户兴趣分析结果准确性低的问题,提出了一种基于眼动数据的网络地图点状符号用户兴趣分析方法。利用空间认知测试法筛选39名认知能力一致的被试者参与实验,使用眼动仪采集被试者在浏览4类点状符号素材过程中的眼动数据,同时记录被试者的鼠标数据;分别计算时间、次数与尺寸类型眼动数据用户兴趣度,利用熵权法将3类数据进行整合,设计了一种基于多项眼动数据的用户兴趣度计算方法。研究结果表明,用户兴趣度分析结果正确率为85.9%,优于鼠标数据,证明所提方法能够有效分析用户兴趣,点状符号用户兴趣度计算公式稳定可靠,有助于提升个性化推荐结果的准确度。

     

    Abstract: In order to solve the problem of the poor accuracy of Web map point symbol user interest during the process of Web map personalized recommendation, we proposed a method for calculating user interest degree of Web map point symbols based on eye movement data. Using mental cutting test, 39 subjects with similar cognitive ability were selected to participate in the experiment and thus we collected subjects' eye movement and mouse data in four types Web map point symbols. We filtered time, frequency and size eye movement data to calculate user interest degree, and established a new method for calculating user interest degree based on multiple eye movement data. An experiment using eye-tracking and mouse device was designed to verify the effectiveness of the method. The results indicate that the accuracy of user interest degree is 85.9%, which is better than those of mouse data. It has been proved that this method is able to effectively analyze the user interest degree, and that the user interest formula is stable and reliable, which lays the foundation for personalized recommendation and improves the effectiveness of recommendation results.

     

/

返回文章
返回