Predict demographic information using Word2vec on spatial trajectories

Adir Solomon, Ariel Bar, Chen Yanai, Bracha Shapira, Lior Rokach

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

Inferring socio-demographic attributes of users is an important and challenging task that could help with personalization, recommendation,advertising,etc.Sensor data collected from mobile devices can be utilized for inferring such attributes. Previous works have focused on combining different typesofsensors,such as applications, accelerometer, GPS, battery,and many others,to achieve this task. In this study, we were able to infer attributes,such as gender, age, marital status, and whether the user has children, using solely the GPS sensor. We suggest a novel inference technique, which learns an embeddingrepresentation of preprocessedspatial GPS trajectoriesusing an adaption of the Word2vec approach. Based on the embedding representation, we later train multiple classification models to achieve the inference goals.Our empirical results indicate that the suggested embedding approach outperformsaclassification approach which does not takeinto consideration the embedding patterns.Experiments on real datasets collected from Android devices show that the proposed method achieves over 80% accuracy for variousdemographic prediction tasks.

Original languageEnglish
Title of host publicationUMAP 2018 - Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization
PublisherAssociation for Computing Machinery, Inc
Pages331-339
Number of pages9
ISBN (Electronic)9781450355896
DOIs
StatePublished - 3 Jul 2018
Event26th ACM International Conference on User Modeling, Adaptation and Personalization, UMAP 2018 - Singapore, Singapore
Duration: 8 Jul 201811 Jul 2018

Publication series

NameUMAP 2018 - Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization

Conference

Conference26th ACM International Conference on User Modeling, Adaptation and Personalization, UMAP 2018
Country/TerritorySingapore
CitySingapore
Period8/07/1811/07/18

Keywords

  • DeepLearning
  • Embedding
  • Trajectories
  • Word2vec

Fingerprint

Dive into the research topics of 'Predict demographic information using Word2vec on spatial trajectories'. Together they form a unique fingerprint.

Cite this