Using Semantic Technology To Model Persona For Adaptable Agents

Abstract

In state of the art research a growing interest in the application of agent models for the simulation of road traffic can be observed. Software agents are particularly suitable for the representation of travellers and their goal-oriented behaviour. Although numerous applications based on these types of models are already available, the options for modelling and calibration of the agents as goal-oriented individuals are either simplified to aggregated parameters or associated with overly complex and opaque implementation details. This makes it difficult to reuse available simulation models. In this paper, we demonstrate how the combination of persona models together with semantic methods can be applied to achieve a well-structured agent model that allows for improved reusability.

Modelling the Impact of Individual Preferences on Traffic Policies

“…In particular, individual preferences on the travel and activity domains are inferred from survey data. Census properties are given to the agent by assigning persona profiles according to the local population (see [25]). These properties are reflected into the ontology and serve as input to rules to infer preferences.…”

Matching Synthetic Populations with Personas: A Test Application for Urban Mobility

“…Recent studies report more quantitatively informed personas through surveys (Schäfer et al, 2019;Goodman-Deane et al, 2021) or big data (Stevenson and Mattson, 2019). On the other hand, on the synthetic population side, attempts are made to enrich traveller agents by persona model attributes by using semantic technology (Nguyen et al, 2021). This semantic structure bears the potential to reuse the agents in different scenarios, which is otherwise a complicated task (Nguyen et al, 2021).…”

“…On the other hand, on the synthetic population side, attempts are made to enrich traveller agents by persona model attributes by using semantic technology (Nguyen et al, 2021). This semantic structure bears the potential to reuse the agents in different scenarios, which is otherwise a complicated task (Nguyen et al, 2021). While existing approaches provide valuable insights on various processes, such as using a large quantity of data to build a persona set, they neither fully consider the bi-directional communication between both parts nor adapt to future situations.…”

Unlocking Urban Simulation Data with a Semantic City Planning System - Ontologically representing and integrating MATSim output data in a knowledge graph

“…Despite existing domain ontologies, efforts to use transportation ontologies to semantically represent traffic simulation models have been limited. Examples we encountered include modelling agent belief or choice systems (Nguyen et al, 2021), using ontologies to improve data for autonomous vehicle simulations (Schwab et al, 2020), and as a means to explore the results of simulation model for household agent travel schedule (Katsumi and Fox, 2020). However, we found no prior work where MATSim simulation outputs have been semantically represented.…”