idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
The University of Stuttgart will be the first European institution to join the AI Horizons Network of IBM in order to advance AI research on the interaction of language and knowledge as part of a multi-year cooperation. The AI Horizons Network is a global network of researchers and doctoral students founded by IBM to further develop the use of artificial intelligence, machine learning, natural language processing, and related technologies in a series of research projects and experiments. Up to now, more than 80 scientific papers from the network have been published worldwide.
We interact with chatbots almost every day, which by now serve as a user-friendly interface to connect with information sources and various services. Regardless of whether you have questions about insurance benefits, the processing status of a loan, or parcel tracking - most of these interactions are question-and-answer scenarios that can be implemented within a given context using natural language processing (NLP). One of the long-term goals of the AI research community is to further develop NLP in such a way that, in the future, artificial intelligence will be able to interpret human communication and interactions so well that systems can formulate responses to queries independently and in a context-dependent manner. This stage of development is called “natural language understanding” (NLU).
Thanks to NLU, AI systems could in future provide answers based on the analysis of context information. For example, a voice assistant on a mobile device could use GPS information to infer not only the relevance of certain place names, but also that the environment inspires the user to change the subject (“Do we have partner companies in this area that work in logistics?”). To achieve this, scientists still have a number of challenges to overcome. This includes the development of an interface between unstructured data such as oral statements, but also tweets and comments in social media, for one thing, and structured data as they are available in tables or databases for another.
The development of such an interface between language and structured knowledge is the aim of the three-year cooperation project, entitled “Language-Knowledge Interaction”, between the Institute for Natural Language Processing at the University of Stuttgart and IBM Research Europe.
“Perhaps the most striking feature of human language is that its expressions are like chameleons: They can adjust their meaning depending on the context. As a result, people can be amazingly brief when communicating complex issues - even in rapidly changing contexts. Natural language understanding systems have to find evidence in language and text that points to the right context of interpretation,” says Jonas Kuhn, Professor of Computational Linguistics at the University of Stuttgart and project manager for the university. “Today, deep-learning processes are able to recognize relevant patterns in training data. But it’s still a big challenge to go beyond learning the right solutions to a very specific task. Many application scenarios require systems that can generalize and that provide reasons for the decisions that are made.”
“As part of the AI Horizons Network, IBM researchers are working with internationally renowned faculties and brilliant students on a number of ambitious research projects and experiments designed to accelerate the use of AI in order to bring added value to society,” says Dr. Anika Schumann, Manager for Artificial Intelligence at IBM Research and responsible for the project on the part of the company. “The results of this collaboration have the potential to influence and change the use of AI in sectors that are quite different from each other, such as health care, materials science, and finance.”
The network’s projects are aimed at using different applications of AI, for example in the fields of health, the environment, logistics, and education. The network deals with the entire so-called AI stack, from the analysis of the unstructured and structured data that are required for training the systems, to the build-up of new computer infrastructures that are used to optimize the new data-intensive workloads in a digital world.
The Knowledge-Language Interaction project will specifically focus on developing an automated interface between unstructured and structured data using machine learning. Though manually created translation rules are precise, they are limited to a fixed pattern of differentiation. Thanks to recent advances in so-called deep-learning methods, it is now possible to automatically induce complex, multidimensional text and language representations from the natural usage patterns in data.
The collaboration will make the University of Stuttgart the first institution in Europe to join the IBM AI Horizons Network. Leading universities from all over the world, such as the Massachusetts Institute of Technology (MIT), IIT Bombay, Université de Montréal, and the University of Massachusetts at Amherst, collaborate with IBM within the network.
Prof. Jonas Kuhn, University of Stuttgart, Institute of Natural Language Processing, phone +49 711 685-81365, mailto:jonas.kuhn@ims.uni-stuttgart.de
https://www.ibm.com/ibm
https://www.research.ibm.com/artificial-intelligence/horizons-network/
Merkmale dieser Pressemitteilung:
Journalisten, Wissenschaftler
Informationstechnik, Medien- und Kommunikationswissenschaften, Sprache / Literatur
überregional
Forschungs- / Wissenstransfer, Kooperationen
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
