Travel Recommendation Based On Data From Social Media

Abstract

— Big data increasingly benefit every analysis and industrial area like health care, finance service and
industrial recommendation. This paper presents a customized travel sequence recommendation from every travelogues
and community-contributed photos and additionally the heterogeneous data (e.g., tags, geo-location, and date taken)
related to these photos. in contrast to most existing travel recommendation approaches, our approach isn't only
customised to user’s travel interest but to boot prepared to recommend a travel sequence rather than individual Points of
Interest (POIs). Topical package house together with representative tags, the distributions of value, visiting time and
visiting season of every topic, is well-mined to bridge the vocabulary gap between user travel preference and travel
routes. we tend to profit of the complementary of a pair of kinds of social media: attraction and community-contributed
photos. we tend to map every user’s and routes’ matter descriptions to the topical package house to induce user topical
package model and route topical package model (i.e., topical interest, cost, time and season). To suggest customised dish
sequence, first, notable routes are stratified consistent with the similarity between user package and route package. Then
prime stratified routes are any optimized by social similar users’ travel records. Representative pictures with viewpoint
and seasonal diversity of POIs are shown to provide a extra comprehensive impression. we tend to value our
recommendation system on a set of seven million Flickr pictures uploaded by 7,387 users and twenty four,008
travelogues covering 864 travel POIs in nine notable cities, and show its effectiveness. we tend to additionally contribute
a brand new dataset with quite 200K photos with heterogeneous information in nine notable cities.