Abstract
We define Location Category Inference (LCI) as a task of predicting the category of a visited venue, such as bar, restaurant or university, given user location GPS coordinates and a set of venue candidates. LCI is an essential part of the hyper-personalization systems as its output provides deep insights into user lifestyle (has children, owns a dog) and behavioral patterns (regularly exercises, visits museums). Due to such factors as signal obstruction, especially in urban canyons, the GPS positioning is inaccurate. The noise in the GPS signal makes the problem of LCI challenging and requires researchers to explore models that incorporate additional information such as the time of day, duration of stay or user lifestyle in order to overcome the noise-induced errors. In this paper we propose an embeddable on-device LCI model which fuses spatial and temporal features. We discuss how initial clustering of locations helps limiting the GPS noise. Then, we propose a multi-modal architecture that incorporates socio-cultural information on when and for how long people typically visit venues of different categories. Finally, we compare our model with one nearest neighbor, a simple fully connected neural network and a random forest model and show that the multi-modal neural network achieves f1 score of 73.2% which is 6.6% better than the best of benchmark models. Our model outperforms benchmark models while being almost 180 times smaller in size at around 1.9Mb.
Supported by Sentiance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Shaw, B., Shea, J., Sinha, S., Hogue, A.: Learning to rank for spatiotemporal search. In: WSDM 2013 (2013). https://doi.org/10.1145/2433396.2433485
Alsudais, A., Leroy, G., Corso, A.: We know where you are tweeting from: assigning a type of place to tweets using natural language processing and random forests. In: 2014 IEEE International Congress on Big Data, pp. 594–600 (2014). https://doi.org/10.1109/BigData.Congress.2014.91. ISSN 2379-7703
Sabatelli, M., Osmani, V., Mayora, O., Gruenerbl, A., Lukowicz, P.: Correlation of significant places with self-reported state of bipolar disorder patients. In: 2014 4th International Conference on Wireless Mobile Communication and Healthcare - Transforming Healthcare Through Innovations in Mobile and Wireless Technologies (MOBIHEALTH), pp. 116–119 (2014). https://doi.org/10.1109/MOBIHEALTH.2014.7015923
Bao, J., Zheng, Yu., Wilkie, D., Mokbel, M.: Recommendations in location-based social networks: a survey. GeoInformatica 19(3), 525–565 (2015). https://doi.org/10.1007/s10707-014-0220-8
McKenzie, G., Janowicz, K.: Where is also about time: a location-distortion model to improve reverse geocoding using behavior-driven temporal semantic signatures. Comput. Environ. Urban Syst. 54, 1–13 (2015). https://doi.org/10.1016/j.compenvurbsys.2015.05.003
Yang, D., Zhang, D., Chen, L., Qu, B.: NationTelescope: monitoring and visualizing large-scale collective behavior in LBSNs. J. Netw. Comput. Appl. 55, 170–180 (2015). https://doi.org/10.1016/j.jnca.2015.05.010. ISSN 1084-8045
Yang, D., Zhang, D., Qu, B.: Participatory cultural mapping based on collective behavior data in location-based social networks. ACM Trans. Intell. Syst. Technol. 7(3), 30:1–30:23 (2016). https://doi.org/10.1145/2814575. ISSN 2157-6904
Pang, J., Zhang, Y.: DeepCity: a feature learning framework for mining location check-ins. In: Eleventh International AAAI Conference on Web and Social Media (2017)
He, J., Li, X., Liao, L., Cheung, W.K.: Personalized next Point-of-Interest Recommendation via Latent Behavior Patterns Inference (2018). arXiv:1805.06316
Keßler, C., McKenzie, G.: A geoprivacy manifesto. Trans. GIS 22(1), 3–19 (2018). https://doi.org/10.1111/tgis.12305. ISSN 1467-9671
Duan, Y., Lu, W., Xing, W., Bao, P., Wei, X.: PBEM: a pattern-based embedding model for user location category prediction. In: 2019 Twelfth International Conference on Mobile Computing and Ubiquitous Network (ICMU), pp. 1–6 (2019). https://doi.org/10.23919/ICMU48249.2019.9006662
Kim, Y.M., Song, H.Y.: Analysis of relationship between personal factors and visiting places using random forest technique. In: 2019 Federated Conference on Computer Science and Information Systems (FedCSIS), pp. 725–732 (2019). https://doi.org/10.15439/2019F318. ISSN 2300-5963
Wongvibulsin, S., Martin, S., Saria, S., Zeger, S., Murphy, S.: An individualized, data-driven digital approach for precision behavior change. Am. J. Lifestyle Med. 14, 155982761984348 (2019). https://doi.org/10.1177/1559827619843489
Yi, J., Lei, Q., Gifford, W.M., Liu, J., Yan, J., Zhou, B.: Fast unsupervised location category inference from highly inaccurate mobility data. In: Proceedings of the 2019 SIAM International Conference on Data Mining (SDM), Proceedings, pp. 55–63. Society for Industrial and Applied Mathematics (2019). https://doi.org/10.1137/1.9781611975673.7
Andrade, T., Cancela, B., Gama, J.: Mining human mobility data to discover locations and habits. In: Cellier, P., Driessens, K. (eds.) ECML PKDD 2019. CCIS, vol. 1168, pp. 390–401. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-43887-6_32
Angmo, R., Aggarwal, N., Mangat, V., Lal, A., Kaur, S.: An improved clustering approach for identifying significant locations from spatio-temporal data. Wireless Pers. Commun. 121(1), 985–1009 (2021). https://doi.org/10.1007/s11277-021-08668-w
De Maio, C., Gallo, M., Hao, F., Yang, E.: Who and where: context-aware advertisement recommendation on Twitter. Soft. Comput. 25(1), 379–387 (2020). https://doi.org/10.1007/s00500-020-05147-5
Islam, M.A., Mohammad, M.M., Das, S.S.S., Ali, M.E.: A survey on deep learning based point-of-interest (POI) recommendations. Neurocomputing 472, 306–325 (2022). https://doi.org/10.1016/j.neucom.2021.05.114. ISSN 0925-2312
McKenzie, G., Romm, D., Zhang, H., Brunila, M.: PrivyTo: a privacy-preserving location-sharing platform. Trans. GIS 26(4), 1703–1717 (2022). https://doi.org/10.1111/tgis.12924
Zhang, H., McKenzie, G.: Rehumanize geoprivacy: from disclosure control to human perception. GeoJournal 88(1), 189–208 (2022). https://doi.org/10.1007/s10708-022-10598-4. ISSN 1572-9893
Sentiance journeys app for android. https://play.google.com/store/apps/details?id=com.sentiance.journeys. Accessed 01 Apr 2023
Sentiance journeys app for IoS. https://apps.apple.com/be/app/journeys-by-sentiance/id984087229. Accessed 01 Apr 2023
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Appendix
A Appendix
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Musaev, G., Mets, K., Tamošiūnas, R., Uvarov, V., De Schepper, T., Hellinckx, P. (2023). On-device Deep Learning Location Category Inference Model. In: Calders, T., Vens, C., Lijffijt, J., Goethals, B. (eds) Artificial Intelligence and Machine Learning. BNAIC/Benelearn 2022. Communications in Computer and Information Science, vol 1805. Springer, Cham. https://doi.org/10.1007/978-3-031-39144-6_7
Download citation
DOI: https://doi.org/10.1007/978-3-031-39144-6_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-39143-9
Online ISBN: 978-3-031-39144-6
eBook Packages: Computer ScienceComputer Science (R0)