The discovery of people profiles such as workers, students, families with kids, etc, is of interest for several application domains. For decades, such information has been extracted using census data, and more recently, from social networks, where people’s profile is clearly defined. A new type of data that has not been explored for discovering profiles, but which stores the real movement of people, are trajectories of moving objects. In this paper we propose a rule-based method to represent socio-demographic profiles, a moving object history model to summarize the daily movement of individuals, and define similarity functions for matching the profile model and the history model. We evaluate the method for single and multiple profile discovery.
The discovery of people profiles such as workers, students, families with kids, etc, is of interest for several application domains. For decades, such information has been extracted using census data, and more recently, from social networks, where people's profile is clearly defined. A new type of data that has not been explored for discovering profiles, but which stores the real movement of people, are trajectories of moving objects. In this paper we propose a rule-based method to represent socio-demographic profiles, a moving object history model to summarize the daily movement of individuals, and define similarity functions for matching the profile model and the history model. We evaluate the method for single and multiple profile discovery.
A rule-based method for discovering trajectory profiles
RAFFAETA', Alessandra;
2015-01-01
Abstract
The discovery of people profiles such as workers, students, families with kids, etc, is of interest for several application domains. For decades, such information has been extracted using census data, and more recently, from social networks, where people's profile is clearly defined. A new type of data that has not been explored for discovering profiles, but which stores the real movement of people, are trajectories of moving objects. In this paper we propose a rule-based method to represent socio-demographic profiles, a moving object history model to summarize the daily movement of individuals, and define similarity functions for matching the profile model and the history model. We evaluate the method for single and multiple profile discovery.
| File | Dimensione | Formato | |
|---|---|---|---|
|
seke15paper_143.pdf
accesso aperto
Descrizione: Articolo principale
Tipologia:
Versione dell'editore
Licenza:
Accesso gratuito (solo visione)
Dimensione
390.04 kB
Formato
Adobe PDF
|
390.04 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
