2019 – 2021

Computational Pun-derstanding

Computer-assisted translation of humorous wordplay

The translation of wordplay is one of the most extensively researched problems in translation studies, but until now it has attracted little attention in the fields of artificial intelligence and language technology. In Computational Pun-derstanding, we study how professional translators process wordplay, with particular attention to the tools, knowledge sources, and working processes they employ. We then decompose these processes and look for parts that can be modelled computationally as part of an interactive, computer-assisted translation system. With this “machine-in-the-loop” paradigm, language technology is applied only to those subtasks it can perform best, such as searching a large vocabulary space for translation candidates matching certain phonetic and semantic constraints. Subtasks that depend heavily on real-world background knowledge—such as selecting the candidate that best fits the wider humorous context—are left to the human translator.

2015 – 2020


Evolution and Function of the Environmental Protein Sequence Universe

Protein sequences are generated in large quantities by DNA sequencing and represent one of the most important reservoirs of molecular biological data. Protein sequences point to the molecular functions and biological roles of their gene products through blueprints of the function and structure of their encoded proteins and their connected evolutionary relationships. During the last decade, the sequencing of metagenomes directly from environmental samples without cultivation has significantly expanded the known protein sequence universe. However, the environmental protein universe is still mainly unstructured and awaits specific utilization in computational biology; although, hundreds of metagenomes have been deeply sequenced and thereby account for the majority of protein sequences stored in databases. The central aim of this proposal is investigating the fundamental evolutionary structures behind the environmental protein sequences previously obtained. We will cluster the entire protein sequence universe, including metagenomes, into evolutionary related families. Based on established concepts, such as orthology or protein domains, this project will develop novel clustering methods for large protein networks.

2019 – 2022


On Valid and Reliable Experiments in Music Information Retrieval

Music Information Retrieval (MIR), as the interdisciplinary science of retrieving information from music, conducts experiments with a multitude of methods from machine learning, statistics, signal processing, artificial intelligence, etc. It relies on the proper evaluation of all these methods to measure the success of new algorithms, or, in more general terms, chart the progress of the whole field of MIR. The principal role of computer experiments and their statistical evaluation within MIR is now widely accepted and understood, but the more fundamental notions of validity and reliability in MIR experiments are still rarely discussed within the field. This lack of awareness for valid and reliable MIR experimentation is at the heart of a number of seemingly puzzling phenomena in recent MIR research and will be tackled in this project. The project is currently located at the Johannes Kepler Universität Linz.

2019 – 2022

Dust and Data

The Art of Curating in the Age of Artificial Intelligence

Artificial intelligence and machine learning are both a challenge and a chance for today’s museums and their growing digital collections. „Dust and Data – The Art of Curating in the Age of Artificial Intelligence” (DAD) as a 2-years artistic research project strives to explore the curatorial potential of artificial intelligence within a set of co-operations with several museum institutions. DAD is located at the Academy of Fine Arts, Vienna and the Institute of Computational Perception, Johannes Kepler Universität Linz. During the first nine month, DAD was also located at the Austrian Research Institute for Artificial Intelligence (OFAI). DAD is funded by the FWF PEEK-program.

2016 – 2019


Coordination and Collaborative Creativity in Music Ensembles

Music ensemble performance requires precise temporal coordination between performers. As an art form, it also requires the creative interpretation of existing music, if not the creation of new music altogether. Thus, though ensemble musicians aim to sound unique, they are simultaneously constrained by the need to remain predictable to each other. How ensembles achieve well-coordinated performances in musical contexts requiring creative interpretation or improvisation is the question driving this research. Our aim is to identify the cognitive mechanisms underlying musical creativity in groups. We consider both human-human and human-computer collaboration to test how factors normally present in live human interaction (e.g. opportunity for communication, perception of co-performers as intention and responsive) affect creative collaboration.

2015 – 2019


Semantic Annotation by Learned Structured and Adaptive Signal Representations

The goal of SALSA is to bridge the semantic gap in music information research (MIR) by using adaptive and structured signal representations. The semantic gap is the difference in information content between signal representations or models used in MIR and high-level semantic descriptions used by musicians and audiences. Examples are the mapping from signal representation to concrete content such as instrumentation or to more abstract tags such as the emotional experience of music.

2016 – 2019


Robotic Action-Language Learning through Interaction

Future social robots will need the ability to acquire new tasks and behaviours on the job both through observation and through natural language instruction, for robot designers cannot build in all environmental and task contingencies. In this project, we tackle the critical subproblem of learning new actions and their corresponding words by the artificial system observing how those actions are performed and expressed by humans. Inspired from psychological studies, we develop experimentation-based algorithms for word learning, integrated with natural language understanding and generation.

2015 – 2019


Automatische Aggregation, Kontextualisierung, Zusammenfassung und Reformulierung von Nachrichten und Meldungen

Das Projekt beschäftigte sich mit der Zusammenfassung von News-Artikeln für News-Aggregatoren. Dabei wurden Lösungen für extraktive und abstraktive Summarization basierend auf Heuristiken und unter Anwendung maschinellen Lernens auf ihre Eignung für den konkreten Anwendungsfall untersucht. Entsprechende Funktionalitäten wurden in einem Prototyp implementiert. Im letzten Drittel des Projektes wurden in einer Pilotphase Ergebnisse aus dem Bereich News-Summarization auf die Rechtsdomäne übertragen und adaptiert.

2015 – 2018


Character Mining and Generation

The hero, the villain, the servant, the mentor, and many more ... movie and drama continue to rely on a repertoire of archetypical characters. But what makes a character? The proposed project CHARMinG will develop and apply AI methods from text and sentiment mining, natural language processing and machine learning to identify, from electronic sources of fictional dialogues (movie scripts, transcripts, drama texts), a set of indicators that convey the core of the relational/functional features and personality of characters, thereby leading to the generation of more colourful and engaging virtual characters.

2017 – 2018

De-escalation Bot

AI for a better discourse in online news fora

OFAI together with the Austrian (online) newspaper Der Standard developed the De-Escalation Bot, a series of machine learning based classifiers which scan forum postings for ones which are expected to provide valuable contributions to a forum and therefore should be visible not only for a rather short period of time but should be accessible to all users in a forum. The classifiers help forum moderators to identify such postings from a plethora of posts being produced by the forum users. Pinning respective posts on top of the forum has shown to improve the discussion quality, reports Der Standard (article in German).