Travelers have always waisted hours in planning their trips trying to maximize their travel excitement. When planning a trip,a traveler searches the Internet using keywords representingthe city she is visiting, and her interests such as historical places, shopping areas, markets, fun events, local cuisines, etc. However, the experience of planning a trip can be daunting, due to the overwhelming number of recommended choices, and lack of trust on the posted information about these choices.
Often, the recommended choices to travelers lack valuable details essential for planning, such as working hours, other people’s experiences, ratings, etc. Recommender Systems try to alleviate the hardship of trip planning by filtering out the non-relevant choices (e.g., events, and activities), and only recommend the most relevant choices that match the traveler’s keywords. Some known Recommender Systems, such as Tripadvisor, incorporate the ratings of users into the filtering process. Nevertheless, these ratings are subjective, and thus the recommended choices may not match the interests and profiles of other travelers.
Nowadays, major companies including those in the tourism sector are transforming their business operations into the digital world. That is, tourism companies are relying on social media to dissemina te and retrieve information. It is expected that this trend will grow as social media continues to have socialimpact on people’s daily life. Social media platforms such as Facebook, YouTube, Instagram, and Twitter, will continue to influence the plans of events and activities of travelers. However, the success of the tourism industry in providing tailored services to travelers to personalize their trips rely heavily on theavailability of reliable event-related content. Therfore, these companies encourage travelers to generate relevant content through prizes, discounts on certain products, etc.
With the hype of social media, travelers are looking for more personalized content to help them plan their individual trips. By leveraging the stored user proflles, user’s history of likes and dislikes, travel schedule, and the set of user keywords, tourism platforms can search social media sites, filter out the irrelevant content, and recommend a personalized set of activities that matches the traveler’s interests and trip schedule.
Recommender Systems build their knowledge base by first collecting social media posts from one source, or multiple sources. These posts undergo several phases of data pre-processing, data management and mining techniques. Complex machine learning and/or deep learning techniques are used to classify multimedia data (text, images, etc.), and to cluster relevant posts in order to generate social events used in the trip planning process.Moreover, AI techniques can be used to learn from previous users travel patterns and itineraries, and thus the system can compare and recommend similar patterns to like-minded travelers. For example, a European traveler who likes historical places, warm weather and local middle eastern food might get a recommendation based on previous trips of Europeans traveling to Sharjah. That user will be recommended to visit Sharjah old Souq, followed by a walk along the Sharjah Corniche, and stop by the traditional Café to have a drink of local Qahwa.
This is where “Traveln” comes to bridge the gap by introducing an AI-powered trip recommendersystem that automates the process of building personalized trips based on live events and activitiesextracted from social media. Traveln is a smart platform connecting the world’s travelers, allowing them tocollaborate, share their experiences, indulge in activities prepared by locals, discover events and hiddengems from social networks, and build engaging, fun, dynamic and rewarding trips.