Improving Orienteering-based Tourist Trip Planning with Social Sensing

Varování

Publikace nespadá pod Ústav výpočetní techniky, ale pod Fakultu informatiky. Oficiální stránka publikace je na webu muni.cz.

Autoři	PERSIA Fabio PILATO Giovanni GE Mouzhi BOLZONI Paolo D'AURIA Daniela SVEN Helmer
Rok publikování	2020
Druh	Článek v odborném periodiku
Časopis / Zdroj	Future Generation Computer Systems
Fakulta / Pracoviště MU	Fakulta informatiky
Citace
www	http://dx.doi.org/10.1016/j.future.2019.10.028
Doi	http://dx.doi.org/10.1016/j.future.2019.10.028
Klíčová slova	Social sensing;Orienteering;Semantic mapping;Semantic similarity
Popis	We enhance a tourist trip planning framework based on orienteering with category constraints by adding social sensing. This allows us to customize a user’s experience without putting the burden of preference elicitation on the user. We identify the interests of a user by analyzing their Tweets and then match these interests to descriptions of points of interests. For this analysis we adapt different schemes for social sensing to the needs of our orienteering context and compare them to find the most suitable approach. We show that our technique is fast enough for use in real-time dynamic settings and also has a higher accuracy compared to previous approaches. Additionally, we integrate a more efficient algorithm for solving the orienteering problem, boosting the overall performance and utility of our framework further, as demonstrated by the positive user satisfaction received by real users.