Limits...
Understanding Human Mobility from Twitter.

Jurdak R, Zhao K, Liu J, AbouJaoude M, Cameron M, Newth D - PLoS ONE (2015)

Bottom Line: Our analysis shows that geotagged tweets can capture rich features of human mobility, such as the diversity of movement orbits among individuals and of movements within and between cities.We also find that short- and long-distance movers both spend most of their time in large metropolitan areas, in contrast with intermediate-distance movers' movements, reflecting the impact of different modes of travel.Our study provides solid evidence that Twitter can indeed be a useful proxy for tracking and predicting human movement.

View Article: PubMed Central - PubMed

Affiliation: CSIRO, Brisbane, Australia.

ABSTRACT
Understanding human mobility is crucial for a broad range of applications from disease prediction to communication networks. Most efforts on studying human mobility have so far used private and low resolution data, such as call data records. Here, we propose Twitter as a proxy for human mobility, as it relies on publicly available data and provides high resolution positioning when users opt to geotag their tweets with their current location. We analyse a Twitter dataset with more than six million geotagged tweets posted in Australia, and we demonstrate that Twitter can be a reliable source for studying human mobility patterns. Our analysis shows that geotagged tweets can capture rich features of human mobility, such as the diversity of movement orbits among individuals and of movements within and between cities. We also find that short- and long-distance movers both spend most of their time in large metropolitan areas, in contrast with intermediate-distance movers' movements, reflecting the impact of different modes of travel. Our study provides solid evidence that Twitter can indeed be a useful proxy for tracking and predicting human movement.

No MeSH data available.


Related in: MedlinePlus

(a)The isotropy ratio σ steadily decreases with rg before increasing again from 200km to around 1000km, indicating that popular intercity trips contribute to increased isotropy. (b) The probability of return to the most popular location and more generally the preference for previously visited locations both drop significantly with increasing rg.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4496063&req=5

pone.0131469.g005: (a)The isotropy ratio σ steadily decreases with rg before increasing again from 200km to around 1000km, indicating that popular intercity trips contribute to increased isotropy. (b) The probability of return to the most popular location and more generally the preference for previously visited locations both drop significantly with increasing rg.

Mentions: Next, we analyse the capacity of geotagged Tweets to capture the spatial orbit of movement for groups of the population defined according to their radius of gyration rg. We explore the probability density function P(x, y), i.e. the probability that a user is observed at location (x, y) in its intrinsic reference frame (see [7] for details). We measure this function for user groups of different radius of gyration rg, as shown in Fig 4. We use the isotropy ratio [7]σ = δy/δx, where δy is the standard deviation of P(x, y) along the y-axis and δx is the standard deviation of P(x, y) along the x-axis, to characterise the orbit of each rg group. At very short rg up to 4km, the isotropy ratio slightly increases (see Fig 5(a)). As rg increases furthers, we observe an increase in anisotropy in P(x, y) as in [7]; however, this correlation between increased anisotropy and rg is only valid for shorter distances between 4km to 200km, which maps well to typical distances for the use of cars as a transport mode. Movement patterns become more diffusive (isotropic) once again for rg between 200km and 1000km. In fact, we observe an unexpected steady rise in σ for distances between 200km and 1000km, where people typically consider modes of transport other than cars, such as trains or planes. The peak in σ is most likely a product of the population distribution in Australia. The top 3 cities account for more than half the population, and the distances between the largest city and commercial capital (Sydney) and the next two cities (Melbourne and Brisbane) are around 963km and 1010km respectively. This result suggests that frequent travellers among these cities are less directed and more diffusive in their movement within the cities, thus the higher isotropic ratio.


Understanding Human Mobility from Twitter.

Jurdak R, Zhao K, Liu J, AbouJaoude M, Cameron M, Newth D - PLoS ONE (2015)

(a)The isotropy ratio σ steadily decreases with rg before increasing again from 200km to around 1000km, indicating that popular intercity trips contribute to increased isotropy. (b) The probability of return to the most popular location and more generally the preference for previously visited locations both drop significantly with increasing rg.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4496063&req=5

pone.0131469.g005: (a)The isotropy ratio σ steadily decreases with rg before increasing again from 200km to around 1000km, indicating that popular intercity trips contribute to increased isotropy. (b) The probability of return to the most popular location and more generally the preference for previously visited locations both drop significantly with increasing rg.
Mentions: Next, we analyse the capacity of geotagged Tweets to capture the spatial orbit of movement for groups of the population defined according to their radius of gyration rg. We explore the probability density function P(x, y), i.e. the probability that a user is observed at location (x, y) in its intrinsic reference frame (see [7] for details). We measure this function for user groups of different radius of gyration rg, as shown in Fig 4. We use the isotropy ratio [7]σ = δy/δx, where δy is the standard deviation of P(x, y) along the y-axis and δx is the standard deviation of P(x, y) along the x-axis, to characterise the orbit of each rg group. At very short rg up to 4km, the isotropy ratio slightly increases (see Fig 5(a)). As rg increases furthers, we observe an increase in anisotropy in P(x, y) as in [7]; however, this correlation between increased anisotropy and rg is only valid for shorter distances between 4km to 200km, which maps well to typical distances for the use of cars as a transport mode. Movement patterns become more diffusive (isotropic) once again for rg between 200km and 1000km. In fact, we observe an unexpected steady rise in σ for distances between 200km and 1000km, where people typically consider modes of transport other than cars, such as trains or planes. The peak in σ is most likely a product of the population distribution in Australia. The top 3 cities account for more than half the population, and the distances between the largest city and commercial capital (Sydney) and the next two cities (Melbourne and Brisbane) are around 963km and 1010km respectively. This result suggests that frequent travellers among these cities are less directed and more diffusive in their movement within the cities, thus the higher isotropic ratio.

Bottom Line: Our analysis shows that geotagged tweets can capture rich features of human mobility, such as the diversity of movement orbits among individuals and of movements within and between cities.We also find that short- and long-distance movers both spend most of their time in large metropolitan areas, in contrast with intermediate-distance movers' movements, reflecting the impact of different modes of travel.Our study provides solid evidence that Twitter can indeed be a useful proxy for tracking and predicting human movement.

View Article: PubMed Central - PubMed

Affiliation: CSIRO, Brisbane, Australia.

ABSTRACT
Understanding human mobility is crucial for a broad range of applications from disease prediction to communication networks. Most efforts on studying human mobility have so far used private and low resolution data, such as call data records. Here, we propose Twitter as a proxy for human mobility, as it relies on publicly available data and provides high resolution positioning when users opt to geotag their tweets with their current location. We analyse a Twitter dataset with more than six million geotagged tweets posted in Australia, and we demonstrate that Twitter can be a reliable source for studying human mobility patterns. Our analysis shows that geotagged tweets can capture rich features of human mobility, such as the diversity of movement orbits among individuals and of movements within and between cities. We also find that short- and long-distance movers both spend most of their time in large metropolitan areas, in contrast with intermediate-distance movers' movements, reflecting the impact of different modes of travel. Our study provides solid evidence that Twitter can indeed be a useful proxy for tracking and predicting human movement.

No MeSH data available.


Related in: MedlinePlus