Semantics in the field of information
In my professional life, I have primarily been concerned with algorithms and semantics, i.e. with various aspects of how we deal with information.
Information plays a part in technology and science, where it can be examined in a technical/scientific manner. However, information also plays a role in completely different fields such as art and philosophy. Are there any conjunctions here? And what do they look like?
On this website, I would like to discuss fundamental and connective aspects of these fields – always from the perspective of information. The background to this is my interest in an integral view of the world. My proposition is that all these fields – technology and science on the one hand and art and philosophy on the other – invariably have something to do with information. We cannot examine everything in all these fields at the same time, but we can examine one thing that they have in common: the way in which they deal with information.
In this context, certain key terms play an important part for me:
Information reduction
A real system – i.e. a human being or a real logical machine (computer) – is incapable of storing the entirety of the accessible information in its surroundings, and therefore the overall volume of information must of necessity be reduced. Information reduction is not only inevitable for real systems but is a characteristic typical of all developed systems, no matter whether they are organic (biology), technological (AI) or cultural (art and society).
A small teaser: with the issue of “information reduction”, we are in the footsteps of two philosophers, namely William of Ockham (Ockham’s razor) and Socrates (I know that I know nothing).
Logodynamics
For many centuries, occidental logic has been concerned with furnishing proof, ranging from Aristotle’s syllogisms to the proof of the existence of God of medieval scholasticism and the more recent first-order logic of mathematics. Despite their mature degree of development, however, these systems do not suffice to map the dynamic processes of thinking. This already starts with the simple IF-THEN, which exists in a static and in a dynamic form. In my view, a dynamic logic is indispensable for a precise depiction of thought processes.
Interpretive system
The theory of the interpretive system stipulates that in the real world, information is always processed within a structured system. This system interprets the information from the environment (input) according to the rules which it finds in an interior space. Naturally, the system has to reduce the volume it absorbs from the environment again and again and profitably concentrate on the information which is important for its objectives.
This information reduction is an active process within the interpretive system.
The Interpretive System is also the title of a book (see below).
Artificial intelligence
Some people look forward to the crucial technological leap, others fear that an infallible artificial intelligence could subjugate humankind. I have worked in the field of knowledge-based systems for a quarter of a century and I mainly marvel about one thing: the fact that we actually attribute genuine intelligence to these – admittedly highly efficient – systems.
Where is the intelligence in so-called artificial intelligence really located? – I am writing a series of blog posts about this, which presents the two fundamental approaches to “artificial intelligence”, namely the rule-based approach and the corpus-based approach, and sheds light on their differences and capabilities.
Book: The Interpretive System
In the 1990s I developed the method of the concept molecules and subsequently described it in a book. Concept molecules are a method for the formal mapping of semantics and for enabling computer programs to interpret free texts. However, we human beings also interpret our environment. The concept molecules are an attempt to transfer as much of this natural interpretation process as possible into the formal world of computers.
The book explains the fundamentals on which a modern semantics can be established. On the basis of the semiotic triangle and the not trivial relationships between its three corners, the phenomenon of information reduction and the position of the interpretive system within the interpreted reality are explained. The annexes contain a brief overview of the concept molecules and a systematisation of the four basic concept architectures.
The book appeared in German in 2001 and was translated in English in the year 2020. It is now available in both languages.
Both editions were published by the ZIM-Verlag of Wolfram Fischer.
→ More about the book in English
→ Buy the book directly from the publisher
→ More about the book in German
→ Buy the German edition directly from the publisher
Semfinder AG
Over a lunch in 1997, Hugo Mosimann and I decided to set up a company for semantic expert systems. The objective was to apply the new method of concept molecules in an NLP program and to use it as a semantic interpretation program for encoding medical texts. Maurus Duelli provided the competent entrepreneurial reinforcement which made it possible for us to navigate the firm even through difficult financial times and finally to install our program in many hundreds of hospitals in Germany, Switzerland, Austria and Spain. In September 2016, we sold the successful company to 3M, which has retained the location, the team and the products.
Acknowledgements
Wolfram Fischer provided me with advice, as well as competent and active support, in the design of this website! The English translations were done by Rachel Waddington, Tony Häfliger and Vivien Blandford. Many thanks!
Translation: Tony Häfliger