The 16th IEEE Conference on Ubiquitous Intelligence and Computing (UIC 2019) was held at the De Montfort University in Leicester, England from August 19 to 23, 2019. Under the guidance of Professor Guo Bin of School of Computer Science, "CrowdTravel: Leveraging Cross-Modal CrowdSourced Data for Fine-grained and Context-based Travel Route Recommendation" by Zhang Jing, a master degree student of the Key Laboratory of Intelligent Perception and Computing, Ministry of Industry and Information Technology was the best student paper of the conference.
The development of the Internet of Things, smart devices and sensors is paving the way for the smart world, and the field of ubiquitous computing has also seen remarkable progress.IEEE UIC conference is an important international conference in the field of ubiquitous computing.At the conference, more than 200 scholars from all over the world conducted in-depth discussions on the latest technologies and major challenges in the field of ubiquitous computing in the form of seminars, presentations, panel discussions and keynote speeches, so as to promote the solution of related issues and the transformation of intelligent services.
Based on group contribution data in cyberspace, this paper proposes a fine-grained contextual association scenic spot route recommendation method based on group intelligence knowledge fusion.Because of the increasing number and different styles of travel notes on various major tourism platforms, the knowledge of them is characterized by low-quality, richness and fragmentation. Even if users spend a lot of time, it is often difficult to find a travel route to meet their specific travel needs.This paper adopts deep learning to extract the features of text and images in travel notes. Through cross-modal semantic mining, the correlation between scenic spot description and image content is realized. And then a population intelligence ranking model, CrowdRank is proposed to optimize various and representative content, making full use of the group intelligence (group preference, group attention, etc.) contained in the group contribution data.Finally, based on user preferences, travel time, travel purposes and other multi-dimensional scenarios, visual travel routes are automatically generated and intelligently recommended.Taking 8 popular scenic spots in China as examples, this paper carries out experiments on travel notes collected from Hornet's Nest and Baidu Tourism. The results show that the proposed method can truly describe scenic spots from multiple perspectives, and the recommended context-based travel routes can meet the specific needs of different groups of people.The paper was highly evaluated, fully affirmed and encouraged by the judges. Finally, the paper won the "Best Student Paper Award" for its innovation, practicality and contribution in the field of group intelligence computing.
Under the leadership of Professor Yu Zhiwen and Professor Zhou Xingshe, the Key Laboratory of Intelligent Perception and Computing, Ministry of Industry and Information Technology has actively explored new research directions such as Group Intelligence Perception Computing and Human-Machine-Object Fusion Computing. The laboratory has undertaken 973, 863, key researches of Natural Science Foundation of China and other national-level important projects. It has over 100 publications about group-aware computing on some authoritative journals such as IEEE Comm. Surv. Tutor. (IF: 22.973), IEEE TMC, TKDE, THMS and prestigious conferences such as UbiComp, KDD, INFOCOM and IJCAI, 6 of which were selected as hot or highly cited papers.Professor Guo Bin of this team, ranks first in the world in Microsoft Academic's Crowd Sensing "Top Authors" ranking. His Reviews and Outlook in the field of Crowd Sensing published in ACM Computing Surveys in 2015 has more than 400 other citations.(Google Academic)