Gender Analysis and Identification using Big Data

Dalberg Data Insights partnered with GSMA Connected Women to develop the Gender Identification and Analysis Toolkit (GAIT) with one primary purpose: to allow operators to predict the gender of their subscribers on an individual, MSISDN level. The information gap the toolkit addresses is an important one; understanding the nature and scale of the mobile gender gap is a prerequisite for closing it. GAIT is a machine learning algorithm that analyses mobile usage patterns to estimate the gender of subscribers. By training the algorithm on a small accurately gender-tagged sample of the customer base to analyse usage patterns by gender, unknown genders for the rest of the subscriber base can be identified with little need for expensive primary research. In Bangladesh, a pilot implementation achieved 84.5% accuracy. The toolkit can then be used to predict the gender of new subscribers as they sign up to and begin using the service. Link: https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2018/09/GSMA-Gender-Analysis-and-Identification-Toolkit-GAIT-August-2018.pdf

Companies

  • G

    GSMA

    Skills

    Inspired by this project? Showcase projects you’ve worked on and inspire other people.

    working to build a more inclusive and sustainable world.
    Recruitment Manager
    Associate Partner & Co-founder, Dalberg Data Insights