TU Darmstadt develops argument search engine for Internet texts
Darmstadt, September 14, 2018. Structured decision-making support: The research project "ArgumenText" of the Ubiquitous Knowledge Processing Lab has found a way to filter concrete arguments for and against any topic from amongst the noise of the internet.
How can someone searching concrete arguments on a question - for example, to make an investment decision or to research a topic in depth - find their way on the internet? Scientists from the Ubiquitous Knowledge Processing Lab (UKP) at TU Darmstadt are working on this question. A demonstrator of the search engine recently became available. Anyone researching a subject like "Nuclear Energy" will, after a few seconds, find approximately a hundred arguments for and against nuclear power - from a wide variety of Internet sites. For comparison: a Google search on the same topic yields around 268 million hits.
For this purpose, texts available on the Internet are examined by means of deep neural networks, classified as relevant or not relevant to the search topic, and then tapped for arguments. "Not only are individual words searched, but grammatical structures, contexts and semantics are considered to decide whether a statement is an 'argument' or not and whether it is on the pro or con side," explains Dr. Johannes Daxenberger, who works in the team of Professor Iryna Gurevych as one of the two project coordinators at ArgumenText.
The algorithms behind ArgumenText are being developed by the team in the department itself, building on initial experiments started in 2014 with a body of student essays. "The challenge was to build a system that was trained on various types of texts transferable to any chosen textual form," says Dr. Christian Stab, second project coordinator. "In scientific texts, for example, argumentation is entirely different than in social media."
The project is currently in the validation phase. The main target groups are decision-makers from the business world who must assess whether the use of an innovation is worthwhile, and journalists who must quickly make their way to the core of a subject for any given topic, says Daxenberger. For validation purposes, the participating scientists are currently preparing the method for use with German-language texts as well. It will also be possible to search in real-time in the ever-growing number of texts on the internet. Scientists are working to ensure that the arguments are later available to users as a body of information presented according to given topics.
Argument mining, the recognition of linguistic arguments by means of computer science, is becoming ever more important and visible, say Daxenberger and Stab, in the research of Digital Humanities. The TU made its way to this area early. "We are well and visibly established with our working group," says Professor Iryna Gurevych, head of the UKP. For this purpose, the interdisciplinary team works, for example, with the TU Department of Social and Historical Sciences, as well as with other universities from the Rhine-Main Universities Alliance.
About TU Darmstadt
The Technische Universität (TU) Darmstadt is one of Germany’s leading technical universities. TU Darmstadt incorporates diverse science cultures to create its characteristic profile. The focus is set on engineering and natural sciences, which cooperate closely with outstanding humanities and social sciences. We are enjoying a worldwide reputation for excellent research in our highly-relevant, focused profile areas: cybersecurity, internet and digitalisation, nuclear physics, fluid dynamics and heat- and mass transfer, energy systems and new materials for product innovation. We dynamically develop our portfolio of research and teaching, innovation and transfer, in order to continue opening up important opportunities for the future of society. Our 312 professors, 4,450 scientific and administrative employees and close to 26,000 students devote their talents and best efforts to this goal. Together with Goethe University Frankfurt and Johannes Gutenberg University Mainz, TU Darmstadt has formed the strategic Rhine-Main Universities alliance.
MI-No. 48e/2018, sip
Department of Computer Science
Ubiquitous Knowledge Processing Lab
Prof. Dr. Iryna Gurevych,
Dr.-Ing. Johannes Daxenberger,
Dr.-Ing. Christian Stab,