Computational Social Science in the Age of Big Data
With the increasing accessibility of the internet and the growing abundance of behavioural choices available on social media, there is a need for a non-traditional viewpoint in the scientific analysis of behavioural trends.
Over the past decade, the field of computational social science (CSS) has gained much importance, with thousands of published articles leveraging diverse research findings and experimental models. Our knowledge of significant trends, from socio-economic disparity to the proliferation of infectious diseases, has been significantly enhanced by the associated research findings. Social science can allow us to understand society’s structural inequalities. At the same time, the subject of Computational Social Science tries to open the black-box of data-driven algorithms that make many significant choices and can assist in understanding the possibility of biases that may arise in such systems.
With the increasing accessibility of the internet and the growing abundance of behavioural choices available on social media, there is a need for a non-traditional viewpoint in the scientific analysis of behavioural trends. Social media users are often encouraged to connect and engage with each other through numerous activities, like direct messaging, sending friend-requests, retweeting tweets from others, posting on each other’s walls, etc. In the scientific study of behavioural trends in consumer analytics, social media’s networking and immersive offering calls for a contextual understanding of the underlying user-patterns, often behavioural.
The emergence of Big Data
Amid a transformation in how we study social science phenomena, big data’s emergence has increased researchers’ potential to achieve high impact. With the advent of innovative data collection techniques, coupled with sophistication in data mining and analytics, the research questions we can pose and the research approaches we can implement may witness radical upgradations. This transformation may allow for impactful inferences of real-world phenomena by scientists, business analysts, and researchers. There are numerous sources of big data, as primitive as images and videos, records of online transactions, GPS data, heatmaps, search history, etc. A recent McKinsey report referred to big data as “data sets whose size is beyond the ability of typical database software tools to capture, store, manage, and analyze”.
Nowadays, many businesses have more data than they can accommodate. At the same time, the managers keep on calculating the associated profit opportunity, but the big data revolution may still be in a state of limbo.
Towards Computational Social Science
Computational social science encompasses interdisciplinary subjects that embrace data collection and analysis capabilities. Computational modelling methods can now forecast the behaviour of socio-technical processes that have not historically been analyzed with data samples like that of human experiences and interactions. Apart from the methodological versatility, interdisciplinary flexibility may be beneficial to derive useful insights from a broad dataset. For example, those with competence in sociology and psychology may be ignorant of the perspective explored in the same piece of information using data mining and machine learning methodologies. It will also be beneficial to have different interpretations of statistics and economics.
The efficiency of observations can easily be facilitated by the diverse understanding of a context, especially from the perspective of social science experts who are usually unaccustomed to the machine learning techniques used by computer engineers. The underlying results and capabilities of computing may also be observed by looking at it from the point-of-view of social science. An interdisciplinary approach to evaluate and control the potential outcomes may be a way forward to look at computing from the sustainability angle.