Taking Airtasker to Task: A case study of Airtasker and gig work

Project description

Through scraping data from Airtasker, this project aims to compile a database of tasks, taskers and posters to understand how apps like Airtasker are changing the nature of work. The database will collect information about the types of task-requirements, costs, and duration. It will also collect information about taskers and posters-demographics, and other available metrics. The internship will process data scraped from Airtasker, assist in coding and classification, and explore possible visualisations that will support future analysis.

Project outcome

This project consisted of an exploratory analysis carried out on data taken from the 'gig economy' marketplace Airtasker, seeking to better understand the nature of tasks being outsourced using the platform. The core of the work involved the development of scripts utilising natural language analysis and machine learning techniques in order to carry out automated pre-processing and analysis of data, extract information from text fields, link to other sources of data, and classify tasks based on the type of job requested. Geographic analysis of the resulting data was also carried out in order to identify spatial trends. The project generated an output data set suitable for subsequent statistical analysis in more depth. It also demonstrated the viability of an automated data processing and analysis methodology suitable for application to a larger data set collected on an ongoing basis.

2019 Internship project


Dr Brendan Churchill
School of Social and Political Sciences


Voytek Lapinski
School of Social and Political Sciences