The traditional way of feeding Answer Set Programming (ASP) is to define data in structured tables and feed these into the ASP solver to generate solutions. Formalizing the problem space is therefore left to the ASP programmers. The new approach to defining the logical structure in the formal specification language SemaLogic allows shifting part of the problem space definition towards the experts involved. In the case of study regulations, we proposed a first draft of the interface for the SemaLogic ASP communication protocol.

The resulting JSON interface is divided into three generic sections of information:

1.    Attributes: as symbols and their properties as attributed values
2.    Groups: to flexibly name the segmentation of all symbols
3.    Terms: to cover the logical relation between symbols as constraints

 

The OpenAPI interface definition is generic in the sense, that no assumptions about the structure of the coded knowledge are reflected in the data format.

In der modernen Hochschullandschaft ist die effiziente Verwaltung von Informationsbeständen ein zentrales Anliegen der Digitalisierung. In diesem Beitrag argumentieren wir, dass der Einsatz symbolischer künstlicher Intelligenz (KI) besonders geeignet ist, um eine konsistente Repräsentation von Studien- und Prüfungsordnungen (StPO) zu erzeugen und effektiv zu nutzen, und demonstrieren dies am Einsatz der formalen Spezifikationssprache SemaLogic. In der Einleitung erläutern wir, auf welche übergreifenden Probleme der Digitalisierung der Hochschullandschaft wir uns konzentrieren, wie unser Ansatz beschaffen ist und welche spezifischen Anwendungsanforderungen bestehen. Im Hauptteil stellen wir unseren Lösungsansatz im Detail vor: SemaLogic ermöglicht eine maschinenlesbare und formalsprachliche Repräsentation von Regelwerken, sodass eine Validierung und dynamische Visualisierung der Strukturen sowie darauf aufbauende Prozesse wie die Studienverlaufsplanung und die Qualitätssicherung während der Curriculumentwicklung möglich werden. Durch seine flexible Anpassung auf der Ebene von Symbolen, Tokens und Syntax rückt SemaLogic die formale Sprache in die Nähe natürlichsprachlicher Eingaben, was den Einsatzbereich erheblich erweitern sollte. Am Ende des Beitrags gehen wir detailliert auf Anwendungsszenarien eines KI-Assistenzsystems für die individuelle Studienverlaufsplanung oder das Qualitätsmanagement und die Unterstützung von Studienreformen ein. Der Beitrag verdeutlicht, dass die Vorteile einer flexiblen, symbolischen Repräsentation von Regelwerken im Lebenszyklus der Hochschulbildung in Datensparsamkeit, Ressourcenschonung und Interoperabilität mit existierenden Campus-Management-Systemen bestehen und sie damit übergreifende Wirkung in der Digitalisierung der Hochschullandschaft hat.

This dataset contains machine-readable versions of the study regulations for the Master's programme "Cognitive Systems: Language, Learning and Reasoning" at the University of Potsdam. The regulations have been transformed into the technical and formal specification language SemaLogic. 

 

Files in this publication:
* Study regulation in technical language 

* Study regulation in formal language 

* SVGs of both versions 

References: 

Study regulation German: http://www.uni-potsdam.de/am-up/2014/ambek-2014-05-200-216.pdf 

Study regulation English translation: https://www.uni-potsdam.de/fileadmin/projects/studium/docs/03_studium_konkret/07_rechtsgrundlagen/studienordnungen/StO_CogSys_EN.pdf

Three application scenarios for AI-based tools are being analyzed to support the recognition process for university and program changes. These are 1. AI-supported structuring as an AI Module Analyzer, 2. AI-based support for the academic sector in assessing criteria for recognition as an AI Module Matcher and 3. an AI Chatbot to support students in consulting and processing of the application. Practical tests with generative AI and a combination of generative and semantic AI were conducted in several variants, which show as a proof of concept that despite all the difficulties that still exist, there is great potential for AI-based tools to make a decisive contribution to transparent and efficient recognition processes in the near future and thus to increasing permeability and mobility in the European higher education system.

Higher education institutions (HEIs) face challenges in adopting digital practices. Neither newly founded nor existing institutions can yet be considered digitally transformed. One important reason for this is the lack of experts which are capable of driving digital transformation (Dx). This shortage of skilled labour for Dx is not only visible in HEIs but throughout society. The solution to this shortage may lie in providing qualified teaching of Dx at the university level in a highly digitised ecosystem that applies Dx to itself. Experiencing digital transformation in problem-based learning scenarios can facilitate a deeper understanding of the methods involved. Breaking up academic programmes into smaller learning units may facilitate lifelong learning and professional development, particularly for experienced individuals. This paper presents the organisational, educational, and scientific perspective of the Digital University (DU) as a mature digital ecosystem which implements this vision for society. The highly interrelated principles between the core functions of the DU enable optimal growth as well as financial and environmental sustainability. The paper discusses bootstrapping new institutions as well as transforming existing universities into this digital ecosystem. In consequence, this paper envisions the DU as a realistic future scenario, which might develop a disruptive power enabling Dx for all HEIs in the long run.

Die Studie wurde im Rahmen des HRK-Projektes MODUS – Mobilität und Durchlässigkeit stärken erstellt.

When implementing study regulations in student information systems, teachers’ intuitive ideas often differ from the later IT mapping. In this article, a rationalisation (both economic and psychological) of such processes is being pursued. The ultimate goal is an AI-based assistance system which provides support for the generation, validation, accreditation and use of study regulations. Advanced AI tools could support a wide range of business capabilities within curriculum design as identified in the Higher Education Reference Model (HERM). Further objectives of this approach are to increase the consistency of individual study planning with study regulations and to support study guidance. Our symbolic approach applies common semantics of natural language and abstract logic which acts as a bridge between the legal norm and the course offerings. The article explains the general concept, introduces the developed technological architecture and presents single tools as well as their integration with the existing IT infrastructure on campus. The curriculum design use case involving extraction and interpretation of the coded knowledge, collaborative editing of study regulations, and fine-grained versioning of study regulations within the creation process is demonstrated and discussed. Finally, the resulting benefits, remaining challenges and future directions are discussed.

As digital workflows evolve around the curriculum life cycle in higher education institutions, advanced digital tools are needed to automate the processing of study regulations. In particular, the use of formal logic is beneficial for any type of validation in the accreditation process. Three key challenges are addressed in this paper: how to model study regulations, how to validate rules contained in modelled programmes, and how to package the contained logic of study rules in flexible communication between different AI services. This report on a case study demonstrates a solution that enables a continuous workflow from editing the rules to automated validation scenarios that support the administrative staff. The use of symbolic logic in conjunction with formal specification languages offers various forms of use cases within the curriculum life cycle.

Student exchange programs such as Erasmus require, to be effective, the transfer of credits from a foreign university to the home university. In order to meet all formal requirements, the rules and regulations of both universities must be matched at the legal, scientific and content levels. The latter two can be intentionally different in order to broaden the horizons of the students. The legal and organizational levels, on the other hand, cause a lot of manual work in the process of formal recognition.
This process is difficult because there is not yet any standardized or generalized way of describing the formal structure of curricula. A solution to this problem -- ideally based on automatic, software-based suggestions -- would enable us to identify similar parts in study programs and their functional role in the curriculum, and also offer “the big picture” at one glance.

This paper shows a first attempt to formally standardize international programs by modelling five BSc in Biology programs from different international universities. The modelling uses SemaLogic, a formal logical description language that was developed specifically to support the overall requirements of curricula. We also present a graphical representation that expresses the identical logical content of the rule set while being clear and easy to understand.

The comparison of the five programs gives the following result:
- The general structures follow either temporal (semesters / years) or thematic pathways. Several alternative models of the same program allow for the distinction between topic-based tracks and their embedding in the overall time structure.
- The degree of flexibility is directly visible through the number of options and their relationship within the general structure. The balance between obligatory elements, free choices, and dependent modules which require certain conditions is immediately apparent.
- The number of alternatives to earn learning credits can be easily estimated. Based on SemaLogic, they can actually be calculated and compared in histograms.

Additional advantages of the logical modelling language became apparent in places where the official documents contained logical errors or undefined relationships of program components, since all SemaLogic-based rules can be verified by a prototype program.

In summary, the formal design of curricula using a logical language such as SemaLogic would greatly enhance the effectiveness of exchange programs both for individual students, as they would be able to understand the requirements better and faster, and for university administrations, as they would be able to realize the formal recognition process with a fraction of the current efforts.

Student exchange programs such as Erasmus require, to be effective, the transfer of credits from a foreign university to the home university. In order to meet all formal requirements, the rules and regulations of both universities must be matched at the legal, scientific and content levels. The latter two can be intentionally different in order to broaden the horizons of the students. The legal and organizational levels, on the other hand, cause a lot of manual work in the process of formal recognition.
This process is difficult because there is not yet any standardized or generalized way of describing the formal structure of curricula. A solution to this problem -- ideally based on automatic, software-based suggestions -- would enable us to identify similar parts in study programs and their functional role in the curriculum, and also offer “the big picture” at one glance.

This paper shows a first attempt to formally standardize international programs by modelling five BSc in Biology programs from different international universities. The modelling uses SemaLogic, a formal logical description language that was developed specifically to support the overall requirements of curricula. We also present a graphical representation that expresses the identical logical content of the rule set while being clear and easy to understand.

The comparison of the five programs gives the following result:
- The general structures follow either temporal (semesters / years) or thematic pathways. Several alternative models of the same program allow for the distinction between topic-based tracks and their embedding in the overall time structure.
- The degree of flexibility is directly visible through the number of options and their relationship within the general structure. The balance between obligatory elements, free choices, and dependent modules which require certain conditions is immediately apparent.
- The number of alternatives to earn learning credits can be easily estimated. Based on SemaLogic, they can actually be calculated and compared in histograms.

Additional advantages of the logical modelling language became apparent in places where the official documents contained logical errors or undefined relationships of program components, since all SemaLogic-based rules can be verified by a prototype program.

In summary, the formal design of curricula using a logical language such as SemaLogic would greatly enhance the effectiveness of exchange programs both for individual students, as they would be able to understand the requirements better and faster, and for university administrations, as they would be able to realize the formal recognition process with a fraction of the current efforts.

Eine wesentliche Aufgabe im Zuge der Digitalisierung der Verwaltungsprozesse einer Universität ist die Abbildung der Prüfungs-und Studienordnungen (PStO) im Campus-Management-System (CMS). Denn nur wenn die in diesen Ordnungen enthaltenen Regeln logisch auswertbar sind, können sie Abläufe im Student-Life-Cycle-Prozess steuern. Dieser Beitrag schlägt eine drastische Vereinfachung als technische Lösung vor. Auf Basis einer semantischen Repräsentation, die direkt aus den Ordnungen erstellt werden kann, wird die menschlich lesbare Form der Ordnung zum ,,Programmcode“ und kann direkt Einfluss auf die Umsetzung im CMS nehmen. Für die Erzeugung einer solchen sprachlichen Übersetzung, welche die notwendige Flexibilität, aber eindeutige Repräsentation erzeugen kann, wurden allgemeine Anforderungen aus der Literatur sowie 20 Studienordnungen analysiert. Die getroffene Auswahl deckt dabei alle Bundesländer, den Großteil der Hochschulformen (Volluniversitäten, Technische Universität, Hochschulen für angewandte Wissenschaften, Musik-und Kunsthochschulen), ein breites Fächerspektrum und vier verbreitete Campus-Management-Systeme ab. Im Weiteren wird aufgezeigt, warum die auf dieser Grundlage entwickelte Sprache SemaLogic die dargelegten Anforderungen erfüllt und dass der Entwicklungsstand der darauf basierenden prototypischen Umsetzung bereits in der Lage ist, die semantische Repräsentation zu erzeugen und logische Prüfungen durchzuführen. Ein Ausblick auf die potentiellen Anwendungsfälle schließt den Beitrag ab.

A method for generating control commands includes providing a document that includes at least one declaration that formulates a rule set, parsing the at least one declaration to generate a plurality of syntactical blocks, constructing terms and symbols from the syntactical blocks to generate a semantic representation, validating the semantic representation, and generating at least one control command based on the validated semantic representation, wherein the at least one control command corresponds to the rule set formulated by the at least one declaration. The at least one control command can advantageously be converted into a plurality of platform-specific instructions for driving a target platform. There is further defined a device for generating control commands and a corresponding system.

The traditional way of feeding Answer Set Programming (ASP) is to define data in structured tables and feed these into the ASP solver to generate solutions. Formalizing the problem space is therefore left to the ASP programmers. The new approach to defining the logical structure in the formal specification language SemaLogic allows shifting part of the problem space definition towards the experts involved. In the case of study regulations, we proposed a first draft of the interface for the SemaLogic ASP communication protocol.

The resulting JSON interface is divided into three generic sections of information:

1.    Attributes: as symbols and their properties as attributed values
2.    Groups: to flexibly name the segmentation of all symbols
3.    Terms: to cover the logical relation between symbols as constraints

 

The OpenAPI interface definition is generic in the sense, that no assumptions about the structure of the coded knowledge are reflected in the data format.