Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Urban Region Embeddings from Service-Specific Mobile Traffic Data

Giulio Loddi, Chiara Pugliese, Francesco Lettich, Fabio Pinelli, Chiara Renso

TL;DR

This work presents a methodology for creating urban region embeddings from service-specific mobile traffic data, employing a temporal convolutional network-based autoencoder, transformers, and learnable weighted sum models to capture key urban features.

Abstract

With the advent of advanced 4G/5G mobile networks, mobile phone data collected by operators now includes detailed, service-specific traffic information with high spatio-temporal resolution. In this paper, we leverage this type of data to explore its potential for generating high-quality representations of urban regions. To achieve this, we present a methodology for creating urban region embeddings from service-specific mobile traffic data, employing a temporal convolutional network-based autoencoder, transformers, and learnable weighted sum models to capture key urban features. In the extensive experimental evaluation conducted using a real-world dataset, we demonstrate that the embeddings generated by our methodology effectively capture urban characteristics. Specifically, our embeddings are compared against those of a state-of-the-art competitor across two downstream tasks. Additionally, through clustering techniques, we investigate how well the embeddings produced by our methodology capture the temporal dynamics and characteristics of the underlying urban regions. Overall, this work highlights the potential of service-specific mobile traffic data for urban research and emphasizes the importance of making such data accessible to support public innovation.

Urban Region Embeddings from Service-Specific Mobile Traffic Data

TL;DR

This work presents a methodology for creating urban region embeddings from service-specific mobile traffic data, employing a temporal convolutional network-based autoencoder, transformers, and learnable weighted sum models to capture key urban features.

Abstract

With the advent of advanced 4G/5G mobile networks, mobile phone data collected by operators now includes detailed, service-specific traffic information with high spatio-temporal resolution. In this paper, we leverage this type of data to explore its potential for generating high-quality representations of urban regions. To achieve this, we present a methodology for creating urban region embeddings from service-specific mobile traffic data, employing a temporal convolutional network-based autoencoder, transformers, and learnable weighted sum models to capture key urban features. In the extensive experimental evaluation conducted using a real-world dataset, we demonstrate that the embeddings generated by our methodology effectively capture urban characteristics. Specifically, our embeddings are compared against those of a state-of-the-art competitor across two downstream tasks. Additionally, through clustering techniques, we investigate how well the embeddings produced by our methodology capture the temporal dynamics and characteristics of the underlying urban regions. Overall, this work highlights the potential of service-specific mobile traffic data for urban research and emphasizes the importance of making such data accessible to support public innovation.

Paper Structure

This paper contains 10 sections, 4 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Preliminaries and problem definition
  4. Methodology
  5. Experimental evaluation
  6. Datasets
  7. Application of our methodology to the NetMob 2023 dataset
  8. Addressing EQ1: evaluating the quality of embeddings
  9. Addressing EQ2: evaluation of temporal dynamics
  10. Conclusions and Future Works

Figures (3)

  • Figure 1: Summary diagram of the methodology: multivariate aggregated MTC time series are inputted to an autoencoder (Step 1), which outputs low-dimensional MTC embeddings. These embeddings are then given as input to a Cell Aggregator (Step 2), which acts as an MTC aggregator to generate the final urban region embeddings. The aggregation is performed under a contrastive learning objective, which captures spatial dependencies between regions.
  • Figure 2: Clustering on full-day time series embeddings ($k=9$).
  • Figure 3: Clustering of the embeddings of the three time slots series ($k=9$).

Theorems & Definitions (4)

  • Definition 1: Mobile Traffic Cell tessellation
  • Definition 2: Cellular Time Series (CTS)
  • Definition 3: Target tessellation
  • Definition 4: Problem Definition