Query Algorithm of Road Section Popularity for Heterogeneous Data

With the development of Internet technology and spatio-temporal data collection technology, urban spatial data is becoming more and more abundant, including not only static spatial object data, but also people's spatial activity trajectory and location-related text comment information. Rich and diverse spatio-temporal data drives the continuous development of new spatial query technologies, such as spatial keyword query, optimal route query, etc. This paper proposes a new spatial route query problem, the query of road section popularity. Given a transportation network graph where references with different keywords are distributed around road segments, the query returns a route that maximizes the popularity of the given keywords satisfying a length constraint. Compared with the existing topic-related spatial route query problems, the query of road section popularity has no end-point constraint. Therefore, this paper designs and implements a 2-hop priority random search strategy and a two-stage random search strategy based on clustering. By establishing a query framework based on time constraints, a route with high popularity value of query keywords is returned to the user in interactive time. In addition, in order to measure the effectiveness of the random search strategy, this paper uses the POT model in extreme value theory to establish the upper bound estimation model of route popularity.