Facility detection and popularity assessment from text classification of social media and crowdsourced data

Kevin A. Sparks; Roger G. Li; Gautam S. Thakur; Robert N. Stewart; Marie L. Urban

doi:10.1145/3003464.3003466

Facility detection and popularity assessment from text classification of social media and crowdsourced data

Kevin A. Sparks, Roger G. Li, Gautam S. Thakur, Robert N. Stewart, Marie L. Urban

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Advances in technology have continually progressed our understanding of where people are, how they use the environment around them, and why they are at their current location. Having a better knowledge of when various locations become popular through space and time could have large impacts on research fields like urban dynamics and energy consumption. In this paper, we discuss the ability to identify and locate various facility types (e.g. restaurant, airport, stadiums) using social media, and assess methods in determining when these facilities become popular over time. We use standard natural language processing tools and machine learning classifiers to interpret geotagged Twitter text and determine if a user is seemingly at a location of interest when the tweet was sent. On average our classifiers are approximately 85% accurate varying across multiple facility types, with a peak precision of 98%. By using these standard methods to classify unstructured text, geotagged social media data can be an extremely useful tool to better understanding the composition of places and how and when people use them.

Original language	English
Title of host publication	Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016
Editors	Ross S. Purves, Christopher B. Jones
Publisher	Association for Computing Machinery, Inc
ISBN (Electronic)	9781450345880
DOIs	https://doi.org/10.1145/3003464.3003466
State	Published - Oct 31 2016
Event	10th Workshop on Geographic Information Retrieval, GIR 2016 - San Francisco Bay Area, United States Duration: Oct 31 2016 → …

Publication series

Name	Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016

Conference

Conference	10th Workshop on Geographic Information Retrieval, GIR 2016
Country/Territory	United States
City	San Francisco Bay Area
Period	10/31/16 → …

Funding

This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05- 00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes.

Keywords

Crowd-sourced data
Geographic information systems
Machine learning
Occupancy analysis
Participatory sensing
Social media
Text classification

Access to Document

10.1145/3003464.3003466

Cite this

Sparks, K. A., Li, R. G., Thakur, G. S., Stewart, R. N., & Urban, M. L. (2016). Facility detection and popularity assessment from text classification of social media and crowdsourced data. In R. S. Purves, & C. B. Jones (Eds.), Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016 Article 3003466 (Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016). Association for Computing Machinery, Inc. https://doi.org/10.1145/3003464.3003466

Sparks, Kevin A. ; Li, Roger G. ; Thakur, Gautam S. et al. / Facility detection and popularity assessment from text classification of social media and crowdsourced data. Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016. editor / Ross S. Purves ; Christopher B. Jones. Association for Computing Machinery, Inc, 2016. (Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016).

@inproceedings{1e020ed9d19a42c7a5ed57a1a7aa9d6d,

title = "Facility detection and popularity assessment from text classification of social media and crowdsourced data",

abstract = "Advances in technology have continually progressed our understanding of where people are, how they use the environment around them, and why they are at their current location. Having a better knowledge of when various locations become popular through space and time could have large impacts on research fields like urban dynamics and energy consumption. In this paper, we discuss the ability to identify and locate various facility types (e.g. restaurant, airport, stadiums) using social media, and assess methods in determining when these facilities become popular over time. We use standard natural language processing tools and machine learning classifiers to interpret geotagged Twitter text and determine if a user is seemingly at a location of interest when the tweet was sent. On average our classifiers are approximately 85% accurate varying across multiple facility types, with a peak precision of 98%. By using these standard methods to classify unstructured text, geotagged social media data can be an extremely useful tool to better understanding the composition of places and how and when people use them.",

keywords = "Crowd-sourced data, Geographic information systems, Machine learning, Occupancy analysis, Participatory sensing, Social media, Text classification",

author = "Sparks, {Kevin A.} and Li, {Roger G.} and Thakur, {Gautam S.} and Stewart, {Robert N.} and Urban, {Marie L.}",

note = "Publisher Copyright: {\textcopyright} 2016 ACM.; 10th Workshop on Geographic Information Retrieval, GIR 2016 ; Conference date: 31-10-2016",

year = "2016",

month = oct,

day = "31",

doi = "10.1145/3003464.3003466",

language = "English",

series = "Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016",

publisher = "Association for Computing Machinery, Inc",

editor = "Purves, {Ross S.} and Jones, {Christopher B.}",

booktitle = "Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016",

}

Sparks, KA, Li, RG, Thakur, GS, Stewart, RN & Urban, ML 2016, Facility detection and popularity assessment from text classification of social media and crowdsourced data. in RS Purves & CB Jones (eds), Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016., 3003466, Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016, Association for Computing Machinery, Inc, 10th Workshop on Geographic Information Retrieval, GIR 2016, San Francisco Bay Area, United States, 10/31/16. https://doi.org/10.1145/3003464.3003466

Facility detection and popularity assessment from text classification of social media and crowdsourced data. / Sparks, Kevin A.; Li, Roger G.; Thakur, Gautam S. et al.
Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016. ed. / Ross S. Purves; Christopher B. Jones. Association for Computing Machinery, Inc, 2016. 3003466 (Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Facility detection and popularity assessment from text classification of social media and crowdsourced data

AU - Sparks, Kevin A.

AU - Li, Roger G.

AU - Thakur, Gautam S.

AU - Stewart, Robert N.

AU - Urban, Marie L.

PY - 2016/10/31

Y1 - 2016/10/31

N2 - Advances in technology have continually progressed our understanding of where people are, how they use the environment around them, and why they are at their current location. Having a better knowledge of when various locations become popular through space and time could have large impacts on research fields like urban dynamics and energy consumption. In this paper, we discuss the ability to identify and locate various facility types (e.g. restaurant, airport, stadiums) using social media, and assess methods in determining when these facilities become popular over time. We use standard natural language processing tools and machine learning classifiers to interpret geotagged Twitter text and determine if a user is seemingly at a location of interest when the tweet was sent. On average our classifiers are approximately 85% accurate varying across multiple facility types, with a peak precision of 98%. By using these standard methods to classify unstructured text, geotagged social media data can be an extremely useful tool to better understanding the composition of places and how and when people use them.

AB - Advances in technology have continually progressed our understanding of where people are, how they use the environment around them, and why they are at their current location. Having a better knowledge of when various locations become popular through space and time could have large impacts on research fields like urban dynamics and energy consumption. In this paper, we discuss the ability to identify and locate various facility types (e.g. restaurant, airport, stadiums) using social media, and assess methods in determining when these facilities become popular over time. We use standard natural language processing tools and machine learning classifiers to interpret geotagged Twitter text and determine if a user is seemingly at a location of interest when the tweet was sent. On average our classifiers are approximately 85% accurate varying across multiple facility types, with a peak precision of 98%. By using these standard methods to classify unstructured text, geotagged social media data can be an extremely useful tool to better understanding the composition of places and how and when people use them.

KW - Crowd-sourced data

KW - Geographic information systems

KW - Machine learning

KW - Occupancy analysis

KW - Participatory sensing

KW - Social media

KW - Text classification

UR - http://www.scopus.com/inward/record.url?scp=85005848029&partnerID=8YFLogxK

U2 - 10.1145/3003464.3003466

DO - 10.1145/3003464.3003466

M3 - Conference contribution

AN - SCOPUS:85005848029

T3 - Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016

BT - Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016

A2 - Purves, Ross S.

A2 - Jones, Christopher B.

PB - Association for Computing Machinery, Inc

T2 - 10th Workshop on Geographic Information Retrieval, GIR 2016

Y2 - 31 October 2016

ER -

Sparks KA, Li RG, Thakur GS, Stewart RN, Urban ML. Facility detection and popularity assessment from text classification of social media and crowdsourced data. In Purves RS, Jones CB, editors, Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016. Association for Computing Machinery, Inc. 2016. 3003466. (Proceedings of the 10th Workshop on Geographic Information Retrieval, GIR 2016). doi: 10.1145/3003464.3003466

Facility detection and popularity assessment from text classification of social media and crowdsourced data

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this