Crowdsourcing-Based Recommendations in Travel and Tourism Market : A Review

Abstract

Conventional methods of vacation preparation that rely on online reviews from numerous sources are frequently time-consuming and impersonal. Trip planning has a lot of options and many interesting insights can be generated by the rise of user-generated content. It has been examined already that with the aid of crowdsourcing, many difficult problems can be solved in a very limited time with less cost. This review provides a brief overview of all the mechanisms for crowdsourcing where multiple user-generated contents provide travelers with personalized trip information. Generally, the visitor ratings about the hotels are gathered from TripAdvisor. Similarly, image data from Flickr, and transportation rates between locations from Uber are collected. First, information on tourist sites from Flickr using geographic data mining techniques is extracted, and thereafter, natural language processing (NLP) is used to identify the multifaceted properties of hotels, and graph analysis to suggest travel routes. Second, researchers created a web-based interface that allows users to engage with the system and offers integrated recommendations for hotels, attractions, and restaurants. On the other hand, this chapter surveys various state-of-the-art techniques and recognizes the most encouraging research trend in tourism recommendation. The many selected methods evaluate the pipeline for mining tourism data streams to find techniques and technologies for real-time predictions guided by design principles for accountability, responsibility, and transparency.