Translated with DeepL
The article "Data modelling - gaining new insights from complex data" in the SME magazine shows how this method can be used to gain usable knowledge from complex data sets and make decisions on a solid basis.
The open source tool Collaboration Spotting X (CSX), co-developed by CERN, puts this approach into practice. It transforms tabular data into dynamic visual networks and visualises relationships that conventional tools often overlook.
This article in the SME magazine is part of a series that has emerged as a follow-up activity to the publication Orientation AI: Challenges and Opportunities for Swiss SMEs. This SATW publication supports small and medium-sized enterprises in recognising the potential of artificial intelligence and planning concrete next steps.
The content was created in close collaboration with stakeholders from the SAIROP (Swiss AI Research Overview Platform) network. SAIROP promotes the exchange between science, business and society, makes Swiss AI expertise visible and provides orientation in the dynamic AI ecosystem.
The article describes how small and medium-sized enterprises (SMEs) can use the hyperbag graph method to model and visually analyse unstructured and complex data - such as customer feedback or operating logs. This creates new opportunities to transform data into knowledge and improve strategic decisions.
SMEs are often faced with the challenge of analysing large amounts of data with limited resources. Conventional business intelligence tools primarily process structured figures, but are less suitable for text data or qualitative information. Hyperbag graphs offer a flexible solution that can also be used to efficiently analyse non-numerical data.
A hyperbag graph is a mathematical model based on so-called graph theory. In contrast to classic graphs, which connect two nodes at a time, a hyperbag graph can link several nodes simultaneously.
In addition, elements can occur multiple times - so-called multisets. This makes it possible to visualise repetitions in texts or ratings, for example when customers mention the same term several times.
SMEs can use the open source tool Collaboration Spotting X (CSX), which was developed at CERN, to convert their data into visual networks.
For example, an e-commerce company can use CSX to analyse which topics occur particularly frequently in product reviews - such as assembly, quality or delivery time - and use these findings for product development or marketing.
Conventional tools such as dashboards or tables primarily display numerical key figures. Hyperbag graphs, on the other hand, also link semantic and text-based information. This enables SMEs to recognise correlations and trends that remain hidden in traditional tables. The approach makes it possible not only to visualise data, but also to understand its content.
Hyperbag graphs help SMEs to present data in an understandable way, recognise complex relationships and set priorities.
As the system works transparently and comprehensibly, decisions can be made on a comprehensible database - without the need for a team of data scientists.
The hyperbag graph method was developed and further refined at CERN by a PhD student at the University of Geneva.
In the article, Gianfranco Moi (University of Geneva, IDE4) and Jean-Marie Le Goff (Dtangle, CERN) present the concept. Both are partners of the Swiss Artificial Intelligence Research Overview Platform (SAIROP), an initiative of the Swiss Academy of Engineering Sciences SATW to promote knowledge transfer in the field of artificial intelligence.
