Crowdtravel: Leveraging heterogeneous crowdsourced data for scenic spot profiling and recommendation

With the prosperity of mobile social networks, more and more people are willing to share their travel experiences and feelings on the Web, which provides abundant knowledge for people who are going to make travel plans. Travel reviews and travelogues are two major ways of social travel sharing. They are complementary in terms of structure, content, and interaction, forming a sort of fragmented travel knowledge. Moreover, the ever-increasing reviews and travelogues may impose the burden on gaining and reorganizing knowledge while making travel plans. Over these issues, this paper proposes CrowdTravel, a multi-source social media data fusion approach for multi-aspect tourism information perception and intelligent recommendation, which can provide travelling assistance for users by crowd intelligence mining. First, we propose a cross-media multi-aspect correlation method to connect fragmented travel information. Second, we mine popular and personalized travel routes from travelogues and make intelligent recommendation based on sequential pattern mining. Finally, we achieve cross-media relevance information based on the similarity between the reviews and image contexts. We conduct experiments over a dataset of eight domestic popular scenic spots, which is collected from two popular social websites about travel, namely Dazhongdianping and Mafengwo. The results indicate that our approach attains fine-grained characterization for the scenic spots and the extracted travel routes can meet different users’ needs.