Considering a Method for Generating Human Mobility Model by Reinforcement Learning

2020 ◽  
pp. 121-132
Author(s):  
Yuutaro Iwai ◽  
Akihiro Fujihara
Keyword(s):  
Reinforcement Learning ◽  
Human Mobility ◽  
Mobility Model

scholarly journals A novel geo-hierarchical population mobility model for spatial spreading of resurgent epidemics

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Alexandru Topîrceanu ◽  
Radu-Emil Precup
Keyword(s):  
Computational Models ◽  
Human Mobility ◽  
Discrete Event ◽  
Mobility Model ◽  
Hierarchical Organization ◽  
Population Mobility ◽  
Mobility Patterns ◽  
Fine Grained ◽  
Epidemic Size ◽  
Control Of Epidemics

AbstractComputational models for large, resurgent epidemics are recognized as a crucial tool for predicting the spread of infectious diseases. It is widely agreed, that such models can be augmented with realistic multiscale population models and by incorporating human mobility patterns. Nevertheless, a large proportion of recent studies, aimed at better understanding global epidemics, like influenza, measles, H1N1, SARS, and COVID-19, underestimate the role of heterogeneous mixing in populations, characterized by strong social structures and geography. Motivated by the reduced tractability of studies employing homogeneous mixing, which make conclusions hard to deduce, we propose a new, very fine-grained model incorporating the spatial distribution of population into geographical settlements, with a hierarchical organization down to the level of households (inside which we assume homogeneous mixing). In addition, population is organized heterogeneously outside households, and we model the movement of individuals using travel distance and frequency parameters for inter- and intra-settlement movement. Discrete event simulation, employing an adapted SIR model with relapse, reproduces important qualitative characteristics of real epidemics, like high variation in size and temporal heterogeneity (e.g., waves), that are challenging to reproduce and to quantify with existing measures. Our results pinpoint an important aspect, that epidemic size is more sensitive to the increase in distance of travel, rather that the frequency of travel. Finally, we discuss implications for the control of epidemics by integrating human mobility restrictions, as well as progressive vaccination of individuals.

Predicting Human Mobility with Reinforcement-Learning-Based Long-Term Periodicity Modeling

2021 ◽  
Vol 12 (6) ◽  
pp. 1-23
Author(s):  
Shuo Tao ◽  
Jingang Jiang ◽  
Defu Lian ◽  
Kai Zheng ◽  
Enhong Chen
Keyword(s):  
Reinforcement Learning ◽  
Human Mobility ◽  
Recurrent Network ◽  
Mobility Prediction ◽  
Learning Framework ◽  
Temporal Features ◽  
Wide Range ◽  
Spatio Temporal ◽  
Historical Trajectory

Mobility prediction plays an important role in a wide range of location-based applications and services. However, there are three problems in the existing literature: (1) explicit high-order interactions of spatio-temporal features are not systemically modeled; (2) most existing algorithms place attention mechanisms on top of recurrent network, so they can not allow for full parallelism and are inferior to self-attention for capturing long-range dependence; (3) most literature does not make good use of long-term historical information and do not effectively model the long-term periodicity of users. To this end, we propose MoveNet and RLMoveNet. MoveNet is a self-attention-based sequential model, predicting each user’s next destination based on her most recent visits and historical trajectory. MoveNet first introduces a cross-based learning framework for modeling feature interactions. With self-attention on both the most recent visits and historical trajectory, MoveNet can use an attention mechanism to capture the user’s long-term regularity in a more efficient way. Based on MoveNet, to model long-term periodicity more effectively, we add the reinforcement learning layer and named RLMoveNet. RLMoveNet regards the human mobility prediction as a reinforcement learning problem, using the reinforcement learning layer as the regularization part to drive the model to pay attention to the behavior with periodic actions, which can help us make the algorithm more effective. We evaluate both of them with three real-world mobility datasets. MoveNet outperforms the state-of-the-art mobility predictor by around 10% in terms of accuracy, and simultaneously achieves faster convergence and over 4x training speedup. Moreover, RLMoveNet achieves higher prediction accuracy than MoveNet, which proves that modeling periodicity explicitly from the perspective of reinforcement learning is more effective.

Disaster Relief Management Using Reinforcement Learning-Based Routing

2021 ◽  
Vol 17 (1) ◽  
pp. 24-37
Author(s):  
Gajanan Madhavrao Walunjkar ◽  
Anne Koteswara Rao ◽  
V. Srinivasa Rao
Keyword(s):  
Reinforcement Learning ◽  
Routing Protocol ◽  
Ad Hoc ◽  
Average Energy ◽  
High Mobility ◽  
Adaptive Routing ◽  
Mobility Model ◽  
Delivery Ratio ◽  
Disaster Area ◽  
Disaster Scenario

Effective disaster management is required for the peoples who are trapped in the disaster scenario but unfortunately when disaster situation occurs the infrastructure support is no longer available to the rescue team. Ad hoc networks which are infrastructure-less networks can easily deploy in such situation. In disaster area mobility model, disaster area is divided into different zones such as incident zone, casualty treatment zones, transport areas, hospital zones, etc. Also, in order to tackle high mobility of nodes and frequent failure of links in a network, there is a need of adaptive routing protocol. Reinforcement learning is used to design such adaptive routing protocol which shows good improvement in packet delivery ratio, delay and average energy consumed.

scholarly journals Characterizing Pairwise Social Relationships Quantitatively: Interest-Oriented Mobility Modeling for Human Contacts in Delay Tolerant Networks

2013 ◽  
Vol 2013 ◽  
pp. 1-15 ◽  
Author(s):  
Jiaxu Chen ◽  
Yazhe Tang ◽  
Chengchen Hu ◽  
Guijuan Wang
Keyword(s):  
Social Relationships ◽  
Mobile Networks ◽  
Human Mobility ◽  
Synthetic Data ◽  
Delay Tolerant Networks ◽  
Mobility Model ◽  
Interaction Matrix ◽  
Mobility Modeling ◽  
The Social ◽  
Delay Tolerant

Human mobility modeling has increasingly drawn the attention of researchers working on wireless mobile networks such as delay tolerant networks (DTNs) in the last few years. So far, a number of human mobility models have been proposed to reproduce people’s social relationships, which strongly affect people’s daily life movement behaviors. However, most of them are based on the granularity of community. This paper presents interest-oriented human contacts (IHC) mobility model, which can reproduce social relationships on a pairwise granularity. As well, IHC provides two methods to generate input parameters (interest vectors) based on the social interaction matrix of target scenarios. By comparing synthetic data generated by IHC with three different real traces, we validate our model as a good approximation for human mobility. Exhaustive experiments are also conducted to show that IHC can predict well the performance of routing protocols.

scholarly journals Community-based Mobility Model and Probabilistic ORBIT Mobility Model Implementations in OMNeT++

10.29007/4tv9 ◽  
2018 ◽  
Author(s):  
Vishnupriya Kuppusamy ◽  
Leonardo Sarmiento ◽  
Asanga Udugama ◽  
Anna Förster
Keyword(s):  
Real World ◽  
Human Mobility ◽  
Mobility Model ◽  
Mobility Models ◽  
Community Based ◽  
Opportunistic Networking

Simulations of Opportunistic Networking (OppNet) protocols require the use of suitable synthetic mobility models or real world traces. Many synthetic mobility models have been proposed based on the study of human mobility individually and in groups. Opportunistic Protocol Simulator (OPS) is a budding simulator which is based on OMNeT++ to simulate OppNets. However, compared to other OppNet simulators in the literature, only very few synthetic mobility models exist in OMNeT++ currently, restricting the simulation of OppNets to using the existing mobility models or traces. In this paper, we develop two more synthetic mobility models in OMNeT++ namely community-based mobility model and probabilistic ORBIT based mobility, which can enhance the simulating environment available for OppNets in OMNeT++.

scholarly journals SRMM: A Social Relationship-Aware Human Mobility Model

Electronics ◽  
2020 ◽  
Vol 9 (2) ◽  
pp. 221 ◽  
Author(s):  
Dat Van Anh Duong ◽  
Seokhoon Yoon
Keyword(s):  
Social Context ◽  
Social Relationships ◽  
Social Relationship ◽  
Human Mobility ◽  
Real Life ◽  
Human Movement ◽  
Mobility Model ◽  
Power Law Distribution ◽  
Road Map ◽  
Human Movements

Since human movement patterns are important for validating the performance of wireless networks, several traces of human movements in real life have been collected. However, collecting data about human movements is costly and time-consuming. Moreover, multiple traces are demanded to test various network scenarios. As a result, a lot of synthetic models of human movement have been proposed. Nevertheless, most of the proposed models were often based on random generation, and cannot produce realistic human movements. Although there have been a few models that tried to capture the characteristics of human movement in real life (e.g., flights, inter-contact times, and pause times following the truncated power-law distribution), those models still cannot reflect realistic human movements due to a lack of consideration for social context among people. To address those limitations, in this paper, we propose a novel human mobility model called the social relationship–aware human mobility model (SRMM), which considers social context as well as the characteristics of human movement. SRMM partitions people into social groups by exploiting information from a social graph. Then, the movements of people are determined by considering the distances to places and social relationships. The proposed model is first evaluated by using a synthetic map, and then a real road map is considered. The results of SRMM are compared with a real trace and other synthetic mobility models. The obtained results indicate that SRMM is consistently better at reflecting both human movement characteristics and social relationships.

Sign in / Sign up

Export Citation Format

Share Document