Left on read: Human behavior characterization based on messaging service applications - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Left on read: Human behavior characterization based on messaging service applications

Résumé

The objective of this study is to analyze mobile data from messaging applications. We aim to understand messaging app usage and gather information that can enhance infrastructure and quality of life within a smart city context. We will conduct our analysis using the NetMob23 dataset [1], which covers the period from March 16, 2019, to May 31, 2019, and provides uplink and downlink data for messaging applications such as WhatsApp, Telegram, Facebook Messenger, and Apple iMessage. Our observations pertain to the Lyon Metropolis in France. The traffic dataset provided is mapped through GeoJSON files using the WGS84 coordinate system. Each feature represents one square cell (tile) covering an area of (100 × 100) m 2. The Lyon metropolitan area has a total of 54013 tiles, and we used the data provided by the Grand Lyon Portal [2] to label some of these tiles, grouping them in the following classes: C1) Education Centers (208 tiles); C2) Events (74 tiles); C3) Commerce (67 tiles); C4) Hotels (58 tiles); C5) Sports (57 tiles); C6) Restaurants (51 tiles); C7) Religious Centers (41 tiles); C8) Hospitals (31 tiles); C9) Train Station (5 tiles). Figs. 1((a)-(d)) show the tile distribution of 4 different classes across the Lyon metropolitan area. We considered these classes to explore the following hypothesis: "It is possible to characterize, based on information theory, the tiles where users are more likely to engage in online conversations". To verify this hypothesis, we conducted a detailed analysis of the tiles for each class, and looked for similar usage behavior categorized with the Complexity-Entropy Causality Plane (CECP) [3]. Only WhatsApp network traffic was considered in the study because it is the application which generates more traffic.
Fichier principal
Vignette du fichier
_NetMob_2023__Data_Challenge_V2 (1).pdf (5.94 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04269561 , version 1 (03-11-2023)

Identifiants

  • HAL Id : hal-04269561 , version 1

Citer

Geymerson S. Ramos, Gean Santos, Douglas L. L. Moura, Danilo Fernandes, Fabiane Queiroz, et al.. Left on read: Human behavior characterization based on messaging service applications. NetMob 2023: Book of Abstracts, Oct 2023, Madrid, Spain. ⟨hal-04269561⟩
85 Consultations
13 Téléchargements

Partager

Gmail Facebook X LinkedIn More