Acknowledging new trends in the research environment is important for many stakeholders, such as researchers, institutional funding bodies, academic publishers, and companies. In particular, being able to identify them as soon as possible can bring an important strategical advantage.
A trend is usually defined as the general direction in which something is evolving. It is often used to describe the popularity of items, such as brands, words, and technologies. In order to detect trends, the relevant items should usually be already recognized and often somewhat popular. For this reason, current methods for detecting trends of research topics usually focus on identifying terms associated with a substantial number of documents, which usually took some years to be produced. Conversely, I theorise that it is possible to perform very early detection of research trends by identify embryonic topics, which have not yet been explicitly labelled or identified by a research community, and that is possible to do so by analysing the dynamics of existent topics. My work is grounded in Kuhn’s theory [1] of the scientific revolution according to which a paradigm shift, also called scientific revolution occurs when a paradigm cannot cope with anomalies, leading to a crisis that will persist until a new outcome redirects research through a new paradigm. In this abstract, I will discuss the state of the art, present an initial study which supports my hypothesis and outline the future directions of my work.
Read more
