In this paper we propose an aggregation strategy for geolocated Twitter posts based on a hierarchical definition of the regular activity patterns within a specific region. The aggregation yields a series of documents that are used to train a topic model. The resulting model is tested against the ones produced by two other aggregation strategies proposed in the literature: aggregation by user and by hashtag. For comparison, we use quality metrics widely used on the literature. The results show that the Geographical Aggregation performs similarly to hashtag aggregation in terms of Jensen-Shannon Divergence and outperforms other aggregation schemes in its ability to reproduce the original cluster labels. One potential application behind this is the discovery of unusual events or as a basis for geolocating messages from text.
KhanA., The role of social media and modern technology in arab spring, Far East Journal of Psychology and Business7(1) (2012), 56–63. Paper 4.
2.
OhO., AgrawalM. and RaoH.R., Information control and terrorism: Tracking the Mumbai terrorist attack through twitter, Information Systems Frontiers13 (2010), 33–43.
3.
CrooksA., CroitoruA., StefanidisA. and RadzikowskiJ., #Earthquake: Twitter as a distributed sensor system, Transactions in GIS17 (2013), 124–147.
4.
BleiD.M., NgA.Y. and JordanM.I., Latent dirichlet allocation, Journal of Machine Learning Research3 (2003), 993–1022.
5.
HongL. and DavisonB.D., Empirical study of topic modeling in Twitter, Proceedings of the First Workshop on Social Media Analytics - SOMA ’10, 2010, pp. 80–88.
6.
WengJ., LimE.-P., JiangJ. and HeQ., TwitterRank, Proceedings of the Third ACM International Conference on Web Search and Data Mining - WSDM ’10, 2010, p. 261.
7.
MehrotraR., SannerS., BuntineW. and XieL., Improving LDA topic models for microblogs via tweet pooling and automatic labeling, Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR ’13, 2013, p. 889.
8.
WangC., WangJ., XieX. and MaW.-y., Mining Geographic Knowledge Using Location Aware Topic Model, Proceedings of the 4th ACM workshop on Geographical information retrieval (GIR), 2007, pp. 65–70.
9.
HuB. and EsterM., Spatial topic modeling in online social media for location recommendation, Proceedings of the 7th ACM conference on Recommender systems - RecSys ’13, 2013, pp. 25–32.
10.
MitchellL., FrankM.R., HarrisK.D., DoddsP.S. and DanforthC.M., The geography of happiness: Connecting twitter sentiment and expression, demographics, and objective characteristics of place, PLoS ONE8 (2013).
11.
GhoshD. and GuhaR., What are we ‘tweeting’ about obesity? Mapping tweets with topic modeling and Geographic Information System, Cartography and Geographic Information Science40 (2013), 90–102.
12.
Frias-MartinezV. and Frias-MartinezE., Spectral clustering for sensing urban land use using Twitter activity, Engineering Applications of Artificial Intelligence35 (2014), 237–245.
13.
LeeR., WakamiyaS. and SumiyaK., Urban area characterization based on crowd behavioral lifelogs over Twitter, Personal and Ubiquitous Computing17 (2013), 605–620.
14.
LeeR., WakamiyaS. and SumiyaK., Discovery of unusual regional social activities using geo-tagged microblogs, World Wide Web14 (2011), 321–349.
15.
KimT., Huerta-CanepaG., ParkJ., HyunS.J. and LeeD., What’s happening: Finding spontaneous user clusters nearby using twitter, Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011, 2011, pp. 806–809.
16.
ChengT. and WicksT., Event detection using twitter: A spatio-temporal approach, PLOS One9 (2014), e97807.
17.
AtefehF. and KhreichW., A survey of techniques for event detection in twitter, Computational Intelligence31 (2015), 132–164.
18.
BoettcherA. and LeeD., EventRadar: A Real-Time Local Event Detection Scheme Using Twitter Stream, IEEE, 2012, pp. 358–367.
19.
SteinskogA.O., TherkelsenJ.F. and GambäckB., Twitter Topic Modeling by Tweet Aggregation, Proceedings of the 21st Nordic Conference of Computational Linguistics, 2017, pp. 77–86.
20.
NewmanD., LauJ., GrieserK. and BaldwinT., Automatic evaluation of topic coherence, in: Proceedings of Human Language Technologies: The 11th Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL HLT 2010), Los Angeles, pp. 100–108.
21.
EisensteinJ., O’connorB., SmithN.A. and XingE.P., A Latent Variable Model for Geographic Lexical Variation, 2010, pp. 1277–1287.
22.
Lopez-RamirezP., Molina-VillegasA., Sanchez-SiordiaO., Chirinos-ColungaM. and Hernandez-ChanG., Multi-scale extraction of regular activity patterns in spatio-temporal events databases: A study using geolocated tweets from central mexico, Research in Computing Science148 (2018).
23.
LeeR. and SumiyaK., Measuring Geographical Regularities of Crowd Behaviors for Twitter-based Geo-social Event Detection, 2010, pp. 1–10.
24.
EsterM., KriegelH.-P., SJ. and XuX., A density-based algorithm for discovering clusters in large spatial databases with noise, AAAI Press, 1996, pp. 226–231.
25.
EdelsbrunnerE. and HerbertM., Three-dimensional alpha shapes, ACM Trans Graph (1994), 43–72.
26.
MimnoD., WallachH.M., TalleyE., LeendersM. and McCallumA., Optimizing semantic coherence in topic models, in: Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, 2, Association for Computational Linguistics, 2011, pp. 262–272.
27.
ArunR., SureshV., Veni MadhavanC.E. and
Narasimha MurthyM.N.
, On Finding the Natural Number of Topics with Latent Dirichlet Allocation: Some Observations, Springer, Berlin, Heidelberg, 2010, pp. 391–402.
