Matchmaking in Academics & Research: Trends & Practices •

Since last year, the esteem breakthroughs of university-industry demonstrated niche collaboration and what can be achieved when there’s a common agenda and understanding between technology transfer teams and industrial R&D.
Almost everyone has followed high-profile Covid-19 partnerships such as the AstraZeneca-Oxford vaccine, and this focus has been inspiring collaborations around the world. However, despite this successful tie-up, particularly for the life sciences, in some areas of collaboration, there have been setbacks to research being still commercialized.
The teams in academics working in technology transfer offices face many challenges in their role and profile in order for academic research to have as large impact as possible, and in maintaining and improving their reputation within the university and the R&D community.
In order to make this happen, tech transfer teams need to find the most promising research from academics at their institute to engage the most sorted and similar industry to be followed. It is worthy to mention that this process of finding best match also requires identifying not only relevant companies, but also the right people within those companies to be in network.
The definite challenge for academic & research is to fix right and holistic partner in industry. Given the size, complexity and global distribution of R&D-driven companies, it’s unsurprising that this is the most common issue faced when approaching commercialization. The professionals in industry as well as academic faculty also highlighted that they face limited time and resources, resulting in challenges effectively managing all of the research produced within their institute and customizing marketing strategies for individual technologies.
Technically, for such academic & research partner matching of a prospective profile and requirements into university/school programs is solicited & classified into two main approaches: Statistical Machine Learning (SML) and Multi-criteria Decision Analysis.
First, a system inspired by statistical machine learning operates on a series of observed data samples by learning to perform a given task from the data samples. The SML methods are applied to resolve qualifies for a specific theme of research in the university/school (forming a group of selected tie-ups) or not (a group of rejected tie-ups).

The predictive value of this method pivoted in a way to contribute to improving the accuracy of an applicant selection process. However, the effectiveness of statistical machine learning methods has been argued due to the level of sophistication in the decision process, the assumptions made, and the level of matching precession achieved.

On the other hand, Multi-criteria Decision Analyzes are a class of multi-criteria optimization methods that incorporate the use of decision matrices to provide a systematic way for evaluating, or ranking, a set of alternatives (in this case of matching the targeted tie-ups into a university/school), relative to a set of decision criteria (the university’s selection criteria). These decision criteria are usually associated with weights – as to reflect their relative importance.

Must’s Matchmaking (patent-pending) is an emerging group of algorithm (s) that could also foster the research and industry partners’ tie-up and thus more comprehensive developments in automated matchmaking will be visible soon.

follow us in LinkedIn