Dataset Information

Evaluation of home detection algorithms on mobile phone data using individual-level ground truth.

ABSTRACT: Inferring mobile phone users' home location, i.e., assigning a location in space to a user based on data generated by the mobile phone network, is a central task in leveraging mobile phone data to study social and urban phenomena. Despite its widespread use, home detection relies on assumptions that are difficult to check without ground truth, i.e., where the individual who owns the device resides. In this paper, we present a dataset that comprises the mobile phone activity of sixty-five participants for whom the geographical coordinates of their residence location are known. The mobile phone activity refers to Call Detail Records (CDRs), eXtended Detail Records (XDRs), and Control Plane Records (CPRs), which vary in their temporal granularity and differ in the data generation mechanism. We provide an unprecedented evaluation of the accuracy of home detection algorithms and quantify the amount of data needed for each stream to carry out successful home detection for each stream. Our work is useful for researchers and practitioners to minimize data requests and maximize the accuracy of the home antenna location.

Supplementary information

The online version contains supplementary material available at 10.1140/epjds/s13688-021-00284-9.

SUBMITTER: Pappalardo L

PROVIDER: S-EPMC8170634 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Evaluation of home detection algorithms on mobile phone data using individual-level ground truth.

Pappalardo Luca L Ferres Leo L Sacasa Manuel M Cattuto Ciro C Bravo Loreto L

EPJ data science 20210602 1

Inferring mobile phone users' home location, i.e., assigning a location in space to a user based on data generated by the mobile phone network, is a central task in leveraging mobile phone data to study social and urban phenomena. Despite its widespread use, home detection relies on assumptions that are difficult to check without ground truth, i.e., where the individual who owns the device resides. In this paper, we present a dataset that comprises the mobile phone activity of sixty-five partici ...[more]

PMID: 34094810

Similar Datasets

Project description:BackgroundApproximately 1 in 5 American adults experience mental illness every year. Thus, mobile phone-based mental health prediction apps that use phone data and artificial intelligence techniques for mental health assessment have become increasingly important and are being rapidly developed. At the same time, multiple artificial intelligence-related technologies (eg, face recognition and search results) have recently been reported to be biased regarding age, gender, and race. This study moves this discussion to a new domain: phone-based mental health assessment algorithms. It is important to ensure that such algorithms do not contribute to gender disparities through biased predictions across gender groups.ObjectiveThis research aimed to analyze the susceptibility of multiple commonly used machine learning approaches for gender bias in mobile mental health assessment and explore the use of an algorithmic disparate impact remover (DIR) approach to reduce bias levels while maintaining high accuracy.MethodsFirst, we performed preprocessing and model training using the data set (N=55) obtained from a previous study. Accuracy levels and differences in accuracy across genders were computed using 5 different machine learning models. We selected the random forest model, which yielded the highest accuracy, for a more detailed audit and computed multiple metrics that are commonly used for fairness in the machine learning literature. Finally, we applied the DIR approach to reduce bias in the mental health assessment algorithm.ResultsThe highest observed accuracy for the mental health assessment was 78.57%. Although this accuracy level raises optimism, the audit based on gender revealed that the performance of the algorithm was statistically significantly different between the male and female groups (eg, difference in accuracy across genders was 15.85%; P<.001). Similar trends were obtained for other fairness metrics. This disparity in performance was found to reduce significantly after the application of the DIR approach by adapting the data used for modeling (eg, the difference in accuracy across genders was 1.66%, and the reduction is statistically significant with P<.001).ConclusionsThis study grounds the need for algorithmic auditing in phone-based mental health assessment algorithms and the use of gender as a protected attribute to study fairness in such settings. Such audits and remedial steps are the building blocks for the widespread adoption of fair and accurate mental health assessment algorithms in the future.

Dataset Information

Evaluation of home detection algorithms on mobile phone data using individual-level ground truth.

Supplementary information

Publications

Evaluation of home detection algorithms on mobile phone data using individual-level ground truth.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets