A Method for Recognizing Location Familiarity to Present Adequate Information to Pedestrians

Pedestrians often use map services that provide route guidance and display nearby buildings with the spread of mobile devices. Conventional map-based services adapt the presented information to the individual by using the user's objective value such as location and the number of visits. We thought that if the presented information on a system is modified based on the degree of feeling familiar about the location, the system can present appropriate information considering the user's situation. Therefore, we defined "location familiarity" as the degree of pedestrians' feeling familiar about a location, and designed a system based on a questionnaire to determine the level of location familiarity that should be recognized and the information to be presented according to the level. Since we consider that a pedestrian's behavior changes when he/she feels familiar, the proposed system recognizes the degree of location familiarity using the acceleration and angular velocity sensor values built-in smartphones, shoes, glasses, and belts and changes the presentation information. The results of the evaluation experiment showed that the recognition accuracy of location familiarity was 0.77. When focusing on only one sensor, the highest recognition accuracy of 0.74 was obtained when using data from the right shoe.