Face-to-face scientific data collection has been facing a crisis all over the world in recent years. The methodology is becoming increasingly difficult and costly to implement. As a result, there is a growing interest in nonface-to-face data collection methods, including research focusing on users’ ‘digital footprints’, as indicators of human behaviour. Those data may derive from a number of sources, including website traffic, social media activity, smartphone app usage and data on movement and other physical conditions from various device sensors. Data is collected through non-intrusive methods, which do not require constant interaction from users, yet at the same time provide a high level of accuracy and detail about people’s digital lifestyles.
The Octopus project aims to combine the collection of sensor and device usage log-based digital behaviour data and questionnaires capable of displaying multimedia content within a complex, smartphone-based software ecosystem. This will result in an internationally innovative research tool for the scientific and even commercial investigation of a wide range of societal issues and problems, for modelling and predicting human behaviour and, on that basis, for supporting policy-making. The development of the Octopus Research Tools started in autumn 2021 in the framework of the Artificial Intelligence National Laboratory, with the participation of social scientists and IT specialists. It was preceded by methodological groundwork and the specification of system functions. This involved an analysis and reconsideration of the features of software available on the market and developing new proprietary features.
A pilot study was conducted in 2021 on Hungary’s largest online research panel. A vignette technique was used to investigate patterns of respondents’ willingness to share app-based data, the role of research incentives, and the types of data people are more willing to or less willing to share or absolutely refuse to share for scientific, commercial and policy purposes. In the study analysing the research, we looked at the way people put a price tag on their own digital footprint compared to the personal information and opinions available through traditional questionnaires. Our results show that the content and context of data collection significantly affect people’s willingness to respond, while they associate different levels of risk and reliability with the different passive sensory data collection techniques. These methodological studies will continue in 2023, initially in the form of a pilot study with 100 respondents.
The general software framework and the management of basic rights will be in place by summer 2022. As of January 2023, available features include the complex project editing interface, the trigger and notification module for editing the questionnaire and controlling the timing of data collection campaigns, and the sampling function. The smartphone app is currently available for Android devices. In 2023, it will also be released for iPhone devices running iOS. While the core functionality of the Octopus Research Tools does not utilize AI, it is closely linked to AI in several aspects. The various active and passive data collection modules can contribute to the development of more efficient AI-based services. They will enable the collection of data from people in a complex and transparent manner and their subsequent use as input for various self-learning algorithms. Throughout the development process, we strive to include AI-based features in any function where this is relevant.
Egedy Tamás, Ságvári Bence. Urban geographical patterns of the relationship between mobile communication, social networks and economic development – the case of Hungary. Hungarian Geographical Bulletin. 70(2), 129–148., 2021
Ságvári Bence, Gulyás Attila, Koltai Júlia. Attitudes Towards Participation in a Passive Data Collection Experiment. Sensors. 21(18):6085., 2021
Nirprint Hungary Kft.
NRC Marketingkutató és Tanácsadó Kft.
SWT Informatika Kft.
TÁRKI Társadalomkutatási Intézet Zrt.