TripAdvisor’s target users are the people traveling for vacations.
Satisfy travel’s needs on transportations, hotel bookings, sightseeing, and restaurant selections.
Analyze traveler behaviors and tailor trip plans for them.
The current ranking system takes into account the below three factors:
There is much to do by creating a customer-centered recommendation system, so travelers feel more relevant with the recommender system.
In 2016, TripAdvisor revised their Popularity Ranking Algorithm to fit better customers’ needs, but without customization based on customers’ behavior analysis and build a predictive model in giving recommendations, there is still long way to go in touching customers’ hearts. My recommendation based on the shopping experiences with TripAdvisor:
Know who I am. Based on my hotel search and resort browsing to estimate my economic status and personal preferences. Recommend the hotels and resort places on the same scale over time.
Know what I like. Restaurants, cuisines, in-door, out-door activities. Hotel style and favorite vacation spot. Whom I would like to travel with, and duration of the trip. If I would like to stay in one place or multiple places at a time.
Know my vacation pattern, and recommend tailored trip plan. How often do I travel each year, and what are the places I visit more often. Using Cookies to track sessions of each visits, build a profile under my name, predict next vacation places, activities, and spending.