30 December 2016:
Gheorghiua, Andreescuc & Curajd (2016) present a review of their experience with horizon scanning, that is to collect and process weak signals, i.e. information on short cycle technologies, peripheral trends and opportunities arising in connection with key enabling technologies.
They state that “a functional weak signal radar faces significant challenges. In broad terms, perhaps the three most important ones are the following:
- Finding good inputs, i.e., a large and varied number of relevant sources of technological news and other information;
- Designing a good signal-filtering system, i.e., keeping the feed of potential weak signals both relevant and numerically manageable, so as to prevent informational overflow; and
- Generating useful categories of weak signals.”
The authors piloted a radar for technological weak signals with the goal to “… grow into a fully automated procedure, among others with the assistance of artificial intelligence (specifically, machine learning AI based on natural language processing).” While no evidence has been given that AI produced useful results, they selected and categorised weak signals based on a community of practice (in their case graduate students) via structured, game-like interactions on the TAGy platform.
After a one-year pilot, the group collected and classified 7,200 weak signals out of 152,000 candidates and a database consisting of 740,000 pieces of information, which in turn are directing future strategies.
They conclude that a “priority-selection agreement” on shared assumption (and information) is a key enabler for a successful entrepreneurial dialogue. This led to their evidence based foresight toolkit for entrepreneurial discovery (figure 1), which is driven by weak signals that have been processed through a qualitative, collaborative method.
Gheorghiua, Radu; Andreescuc, Liviu & Curajd, Adrian 2016: “A Foresight Toolkit for Smart Specialization and Entrepreneurial Discovery”, Futures, vol. 80, p. 33-44.
Link to full text article: