Classification, the process of assigning a label (or class) to an observation given its features,... more Classification, the process of assigning a label (or class) to an observation given its features, is a common task in many applications. Nonetheless in most real-life applications, the labels can not be fully explained by the observed features. Indeed there can be many factors hidden to the modellers. The unexplained variation is then treated as some random noise which is handled differently depending on the method retained by the practitioner. This work focuses on two simple and widely used supervised classification algorithms: discrete choice models and artificial neural networks in the context of binary classification. Through various numerical experiments involving continuous or discrete explanatory features, we present a comparison of the retained methods' performance in presence of missing variables. The impact of the distribution of the two classes in the training data is also investigated. The outcomes of those experiments highlight the fact that artificial neural networks outperforms the discrete choice models, except when the distribution of the classes in the training data is highly unbalanced. Finally, this work provides some guidelines for choosing the right classifier with respect to the training data.
Synthetic populations are tools widely spread in the agent-based community for representing a bas... more Synthetic populations are tools widely spread in the agent-based community for representing a baseline population of interest whose dynamics and evolution will be simulated and studied. The dynamic evolution of the synthetic population has been typically performed using a discrete and fixed time step. A continuous approach based on the Gillespie algorithm is proposed in this research. Preliminary experiments illustrate the potential of the new method before future work are discussed
General rights Copyright and moral rights for the publications made accessible in the public port... more General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Journal of Artificial Societies and Social Simulation, 2018
Agent based modelling is nowadays widely used in transport and the social science. Forecasting po... more Agent based modelling is nowadays widely used in transport and the social science. Forecasting population evolution and analysing the impact of hypothetical policies are often the main goal of these developments. Such models are based on sub-models defining the interactions of agents either with other agents or with their environment. Sometimes, several models represent phenomena arising at the same time in the real life. Hence, the question of the order in which these sub-models need to be applied is very relevant for simulation outcomes. This paper aims to analyse and quantify the impact of the change in the order of sub-models on an evolving population modelled using TransMob. This software simulates the evolution of the population of a metropolitan area in South East of Sydney (Australia). It includes five principal models: ageing, death, birth, marriage and divorce. Each possible order implies slightly different results mainly driven by how agents' ageing is defined with respect to death. Furthermore, we present a calendar-based approach for the ordering that decreases the variability of final populations. Finally, guidelines are provided proposing general advices and recommendations for researchers designing discrete time agent-based models.
Nowadays, a wide range of microsimulations are performed thanks to agent based simulations. This ... more Nowadays, a wide range of microsimulations are performed thanks to agent based simulations. This frame-work allows to model the interactions between agents and environment in a flexible way. Such models should be initialised with a given set of agents representing the initial population, and their interactions rules should be set. Very often the initial population of agents is obtained using a synthetic population generator. TransMob is an agent-based model that aims to simulate transport and land use interdependencies for urban planning. TransMob contains two types of modules responsible for: the dynamics of the social structure within the population (ageing, divorces,...); and the travel behaviour of the individuals (assigning diaries, ...). In the current version of TransMob, the processes happening in the agents everyday life are modelled and performed in a specific order: first ageing, then death, divorce, marriage and finally giving birth. This work focus on the impact of the ...
Classification, the process of assigning a label (or class) to an observation given its features,... more Classification, the process of assigning a label (or class) to an observation given its features, is a common task in many applications. Nonetheless in most real-life applications, the labels can not be fully explained by the observed features. Indeed there can be many factors hidden to the modellers. The unexplained variation is then treated as some random noise which is handled differently depending on the method retained by the practitioner. This work focuses on two simple and widely used supervised classification algorithms: discrete choice models and artificial neural networks in the context of binary classification. Through various numerical experiments involving continuous or discrete explanatory features, we present a comparison of the retained methods' performance in presence of missing variables. The impact of the distribution of the two classes in the training data is also investigated. The outcomes of those experiments highlight the fact that artificial neural networks outperforms the discrete choice models, except when the distribution of the classes in the training data is highly unbalanced. Finally, this work provides some guidelines for choosing the right classifier with respect to the training data.
Synthetic populations are tools widely spread in the agent-based community for representing a bas... more Synthetic populations are tools widely spread in the agent-based community for representing a baseline population of interest whose dynamics and evolution will be simulated and studied. The dynamic evolution of the synthetic population has been typically performed using a discrete and fixed time step. A continuous approach based on the Gillespie algorithm is proposed in this research. Preliminary experiments illustrate the potential of the new method before future work are discussed
General rights Copyright and moral rights for the publications made accessible in the public port... more General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Journal of Artificial Societies and Social Simulation, 2018
Agent based modelling is nowadays widely used in transport and the social science. Forecasting po... more Agent based modelling is nowadays widely used in transport and the social science. Forecasting population evolution and analysing the impact of hypothetical policies are often the main goal of these developments. Such models are based on sub-models defining the interactions of agents either with other agents or with their environment. Sometimes, several models represent phenomena arising at the same time in the real life. Hence, the question of the order in which these sub-models need to be applied is very relevant for simulation outcomes. This paper aims to analyse and quantify the impact of the change in the order of sub-models on an evolving population modelled using TransMob. This software simulates the evolution of the population of a metropolitan area in South East of Sydney (Australia). It includes five principal models: ageing, death, birth, marriage and divorce. Each possible order implies slightly different results mainly driven by how agents' ageing is defined with respect to death. Furthermore, we present a calendar-based approach for the ordering that decreases the variability of final populations. Finally, guidelines are provided proposing general advices and recommendations for researchers designing discrete time agent-based models.
Nowadays, a wide range of microsimulations are performed thanks to agent based simulations. This ... more Nowadays, a wide range of microsimulations are performed thanks to agent based simulations. This frame-work allows to model the interactions between agents and environment in a flexible way. Such models should be initialised with a given set of agents representing the initial population, and their interactions rules should be set. Very often the initial population of agents is obtained using a synthetic population generator. TransMob is an agent-based model that aims to simulate transport and land use interdependencies for urban planning. TransMob contains two types of modules responsible for: the dynamics of the social structure within the population (ageing, divorces,...); and the travel behaviour of the individuals (assigning diaries, ...). In the current version of TransMob, the processes happening in the agents everyday life are modelled and performed in a specific order: first ageing, then death, divorce, marriage and finally giving birth. This work focus on the impact of the ...
Uploads
Papers by Morgane Dumont