面向就医地点推荐的个人时空可达性分析方法

目前如何根据用户需求以及时空制约条件合理安排就医出行,在海量的医疗机构中实现就医地点的合理推荐是人们的迫切需求。区别于常见的个性化医疗信息推荐方法,本文综合考虑医疗机构的实时资源规模、医疗用户的个性化需求以及前往机构的时空制约,提出一种面向就医地点推荐的个人时空可达性分析方法,并将方法运用于居民的就医出行场景中,实现就医地点的有效推荐。以济南市为研究区域,借助从高德地图API以及医疗健康网站中获取的地理信息数据与医疗机构资源数据,在不同用户、时间和地理位置的情境下,设置2处高校图书馆、2处居民点以及2处大型交通站共6个场景检验方法的有效性。实验表明,本文提出的个人时空可达性方法,能够有效度量医疗用户个体至就医地点间的可达程度,可解决不同时空情境下的就医地点推荐问题,同时也为相关智慧医疗技术的功能建设提供方法支撑,具有良好的应用前景。     

Collaborative personalized multi‐criteria wayfinding for wheelchair users in outdoors

While finding the optimal route for users with physical disabilities and personalizing routes for each user, on the one hand, and collaborative wayfinding, on the other hand, have been addressed in the pedestrian navigation systems literature, there has not been much research on combining the two activities. The problem associated with wayfinding approaches solely based on information about network segments and personal preferences is that the information about segments in the database may not, correctly and/or adequately address user preferences. The problem associated with wayfinding approaches solely based on the ratings given to routes by wheelchair users is the lack of rates (or scores) for all possible routes between all possible origin‐destination pairs in the network. This article discusses an approach to combine these two approaches for wayfinding to augment each other's shortcomings. To evaluate the personalized wayfinding approach, we utilize a route index, called a comparison index. The results show that with a P‐value of 9%, the routes obtained from our approach are more accessible than the routes obtained from another approach developed in another study. To evaluate the collaborative wayfinding approach, a Monte Carlo simulation was conducted which reflects the updates in routes as users' feedbacks become available.     

构建面向个性化服务的特大城市云空间数据中心

特大城市测绘地理信息数据具有数据海量、增长快速、多源异构、频繁使用的特点。随着大数据和Web2.0时代的到来,传统的空间数据存储管理与共享服务模式存在成果缺乏整合和挖掘、空间信息服务被动且无差别性等缺点,难以满足用户不断提高的空间数据内容、服务响应速度和服务质量的要求,也难以适应因人而异的个性化空间信息服务的需求。本文对个性化服务和云空间数据中心构建的关键技术进行了研究,提出了个性化空间数据情境应用模式,以及一种基于Web客户端的个人异构空间数据在线集成方法,基于NoSQL实现了个人空间数据与传统空间数据服务集成,并以北京市为例基于私有云构建了安全可靠、高效运行、可扩展、用户参与度高、提供个性化服务的特大城市云空间数据中心。     

Personalized travel route recommendation using collaborative filtering based on GPS trajectories

Travelling is a critical component of daily life. With new technology, personalized travel route recommendations are possible and have become a new research area. A personalized travel route recommendation refers to plan an optimal travel route between two geographical locations, based on the road networks and users’ travel preferences. In this paper, we define users’ travel behaviours from their historical Global Positioning System (GPS) trajectories and propose two personalized travel route recommendation methods – collaborative travel route recommendation (CTRR) and an extended version of CTRR (CTRR+). Both methods consider users’ personal travel preferences based on their historical GPS trajectories. In this paper, we first estimate users’ travel behaviour frequencies by using collaborative filtering technique. A route with the maximum probability of a user’s travel behaviour is then generated based on the naïve Bayes model. The CTRR+ method improves the performances of CTRR by taking into account cold start users and integrating distance with the user travel behaviour probability. This paper also conducts some case studies based on a real GPS trajectory data set from Beijing, China. The experimental results show that the proposed CTRR and CTRR+ methods achieve better results for travel route recommendations compared with the shortest distance path method.     

基于地理标记照片的个性化景点推荐方法

研究如何根据已有的海量旅游信息及数据,为游客个性化推荐旅游景点具有重要意义。本文利用从Flickr网站获取的2013—2018年香港特别行政区范围内的地理标记照片来识别旅游景点,并根据游客游览顺序重建旅游轨迹。在此基础上,针对现有方法尚未考虑游客偏好在旅行过程中会发生动态变化的问题,提出一种基于隐含狄利克雷分布模型（Latent Dirichlet Allocation, LDA）和用户长短期偏好的个性化景点推荐方法（A Recommendation Method Based on LDA and User's Long and Short-Term Preference, L-ULSP）。该方法利用LDA主题模型获取景点特征信息,挖掘景点间的相关性,再利用注意力机制和长短期记忆网络分别学习用户的长期偏好和短期偏好,最后结合长短期偏好捕捉用户偏好的动态变化。实验结果表明, L-ULSP方法所推荐的景点在命中率和平均倒数排名2个指标上均优于现有其他方法,证明了本文所提方法可以从景点序列中有效学习游客偏好,并为游客推荐下一个景点。此外,本文通过对比实验,进一步验证了同时考虑用户的长短期偏好能够更好地学习用户的偏好变化。