Related Experiment Video
Updated: Nov 2, 2025
06:49
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
Published on: December 11, 2015
9.1K
Evaluation of home detection algorithms on mobile phone data using individual-level ground truth
Luca Pappalardo1, Leo Ferres2,3,4, Manuel Sacasa3
1Institute of Information Science and Technologies (ISTI), National Research Council (CNR), Pisa, Italy.
EPJ Data Science
|June 7, 2021
Summary
Accurately determining mobile phone users' home locations is crucial for urban studies. This study evaluates home detection algorithms using real user data, quantifying data needs for improved accuracy and minimized requests.
Area of Science:
- Mobile phone data analysis
- Urban studies
- Social science research
Background:
- Home detection from mobile phone data is vital for urban and social studies.
- Current methods lack ground truth validation, making accuracy assessment difficult.
- A new dataset with known participant residence locations is introduced.
Purpose of the Study:
- To present a novel dataset of mobile phone activity (CDR, XDR, CPR) linked to known home locations.
- To conduct an unprecedented evaluation of home detection algorithm accuracy.
- To quantify data requirements for successful home detection across different data streams.
Main Methods:
- Utilized a dataset of 65 participants with known home coordinates.
- Analyzed Call Detail Records (CDRs), eXtended Detail Records (XDRs), and Control Plane Records (CPRs).
- Evaluated the accuracy of various home detection algorithms and determined data volume needed for each record type.
Main Results:
- Provided a comprehensive accuracy assessment of home detection algorithms.
- Quantified the specific data amounts required from CDRs, XDRs, and CPRs for reliable home detection.
- Established a benchmark for evaluating future home detection methodologies.
Conclusions:
- The study offers a validated approach to assessing home detection accuracy.
- Researchers can now optimize data collection to balance accuracy and efficiency.
- This work facilitates more reliable use of mobile phone data in social and urban research.
