Research

Linguists often try to explain language change, but all our explanations are necessarily post hoc and thus difficult to evaluate. What happens if we turn to the future instead of the past and try to predict language change?

HUMINFRA  är en ny distribuerad, nationell infrastruktur för forskning inom humaniora, konst och samhällsvetenskap.

This project uses Construction Grammar to develop a linguistic network that (a) accounts for Swedish grammatical constructions and (b) connects them to constructions in other languages.

Accessibility of research data is critical for advances in many research fields, but textual data often cannot be shared due to the presence of personal and sensitive information, e.g names, political opinions. GDPR suggests pseudonymization as a solution, but we need to learn more about it before adopting it for manipulation of research data.

Inom ramarna för projektet förvaltas och vidareutvecklas Svenska Akademiens lexikala databas (Salex). Vidare bedrivs arbete med Svenska Akademiens båda samtidsordböcker Svenska Akademiens ordlista (SAOL) och Svensk ordbok utgiven av Svenska Akademien (SO). Arbetet sker på uppdrag av och i samarbete med Svenska Akademien.

ICALL - Intelligent Computer Assisted Language Learning. The aim of the project is to develop an open-source system architecture for supporting ICALL, i.e. CALL that reuses NLP tools and NL resources, with emphasis on the Nordic languages.

Den svenska akademiska ordlistan har utvecklats av forskare med anknytning till forskningsområdena språkteknologi, lexikologi/lexikografi och svenska som andraspråk.

A joint project between Språkbanken Text, FLoV and CLASP, with the purpose of creating and exploring methods for argumentation technology.

Developing tools for systematic studies of text classification

This program has two main aims, firstly to develop corpus-based methods for detecting semantic change (over time) and variation (across social groups and media). This will create general tools for the study and detection of language change at large-scale and directly benefit historical linguistics and lexicography. Secondly, we will collaborate with researchers from social sciences, gender studies, and literature to answer their research questions. We will develop tools, evaluation data, and research methodology for their specific needs.

Consumption patterns and life-style in Swedish literature – novels 1830-1860

We will apply corpus-driven methods as a way to expand and correct existing hand-crafted linguistic resources, and conversely we will use hand-crafted resources as additional sources of supervision when learning meaning representations automatically.

The goal of this project is to create a database of comparable lexical items in a number of representative languages spoken in the Himalayan region in India and to use this database for investigating the Himalayas as a linguistic area.

Digitization of Grierson’s Linguistic Survey of India (LSI; 1903-1927)

A Multilingual Annotated Corpus of Grammars for the World's Languages

The project aims to create a prototype Optical Character Recognition (OCR) web service for processing old Swedish texts that are printed in a blackletter (fraktur) or roman typeface, using one of two open source OCR engines. Our ultimate goal is to provide a service for libraries, museums and archives to upload any digitized document and retrieve an OCRed text with high quality, independent on the quality of the print.

EnetCollect (the European Network for Combining Language Learning with Crowdsourcing Techniques) was a large network project funded as a COST Action that ran from March 2017 till September 2021. It involved stakeholders from more than 40 different countries and has been the catalyst for numerous collaborative research efforts, achievements and publications.

The purpose of this project is to fine-tune and evaluate a test platform for OCR-production that was developed by Kungliga biblioteket (KB) in cooperation with the Norwegian software company Zissor in 2017.

Tolkning och förståelse av funktionella symtom i primärvården

Under vilka förutsättningar försvagas yttrandefrihet och demokrati av hat och hot mot journalister online?

Språkbanken is working on integration of available language technology for Swedish into the area of language learning, so-called ICALL.

Det ettåriga pilotprojektet inventerar och analyserar maskinellt en unik samling av pressmaterial från 1923 års jubileumsutställning i Göteborg. Projektet är ett samarbete mellan GPS400, Språkbanken Text, Universitetsbiblioteket och Riksarkivet.

EU project Kelly - lists of useful vocabulary for language learners in 9 languages

Improved annotations for the Korp corpus infrastructure.

The L2 P project studies lexical and grammatical competence in second language learner Swedish through two corpora: coursebooks and essays.

At Språkbanken Text, we have an ongoing research theme that relates to computational modeling of word meaning called (computational) lexical semantic change. In this theme, we have both research projects and a larger research program.

The market Language primarily is funded by MAW in which we look at the changing concepts around “the market”. They have transitioned from implying a concrete physical market to increasingly abstract markets like Europe-wide iron markets, as well as marriage and dating markets. They have also increasingly become actors in our lives, “the market reacted badly to the new corona restrictions”. We will complement the conceptual historians in-depth analyses with computational models of change. This project ranges 2022-2025.

Developing automatic annotation tools for Old Swedish texts.

En svensk medicinsk testkollektion

The META-NORD project aims to establish an open linguistic infrastructure in the Baltic and Nordic countries.

In this project we want to utilize a useful collection of 9000 digitized grammatical descriptions covering over a thousand languages in order to significantly expand the ability to make major language comparisons.

MOLTO's goal is to develop a set of tools for translating texts between multiple languages in real time with high quality. Languages are separate modules in the tool and can be varied; prototypes covering a majority of the EU's 23 official languages will be built.

Person-Centred Information and Communication for patients undergoing Colo-Rectal Cancer Surgery

The aim of the project is to investigate the role and importance of rumouring for the vaccination skepticism growing on the internet, and how it can be understood as an expression of civic engagement in the present digital times entailing crucial transformations for everyday civic culture.

The aim of this project is to build a versatile lexical infrastructure for Swedish language technology (LT).

With an increasing aging pyramid the number of people with cognitive dysfunctions, such as various types of dementia, has grown at a high rate. However, years before the clinical onset symptoms of dementia, patients exhibit serious deficits in their oral and written communication and visual short-term memory, signs that can be measured and serve as a complement to medical evidence to discriminate the performance of healthy (elderly) controls or even predict poor cognitive health in late life.

This project aims to make lexical resources for language technology available in the form of linked open data.

The goal of this project is to finalize the evaluation framework SuperLim by contributing training data for the current collection of test sets, a reference implementation (baseline), and a standardized web-based test environment for comparison between models and publication of results (leaderboard).

The purpose of this project is to create high-quality test sets for att enable all actors within Swedish NLP to evaluate and compare language models.

The goal of the SweFN++ project is to build an open-content -- i.e., freely available and modifiable -- integrated lexical resource for Swedish -- so far lacking -- to be used as a basic infrastructural component in Swedish language technology (LT) research and in the development of LT applications for Swedish.

Målet med Svenskt Språkdatalabb är att skapa en nationell kunskapsnod inom språkteknologi, och ta fram svenska referensdatamängder för NLP som sedan tillgängliggörs med öppen access i AI Innovation of Swedens datafabrik.

The aim of this project is to develop a Swedish so-called constructicon, a database of Swedish constructions.

The main focus of this project is on producing data, tools and workflows for research an and evaluation of L2 Swedish.

Computerised text classification is used to help identify documents on the topic of patient-centred care.

The project studies the rise of complex verb constructions in Germanic.

The main aim of this research program is to advance the state of the art in language technology resources and methods for semantic processing of Swedish text, in order to provide researchers and others with more sophisticated tools for working with the information contained in large volumes of digitized text, e.g., by being able to correlate and compare the content of texts and text passages on a large scale.

In this project, we aim to find automatic, corpus-based methods for detecting semantic change and lexical replacement for Swedish and English.

This project investigates which strategies are employed when North Germanic personal names are adapted to medieval German, French and Latin in multilingual contexts. It aims at surveying the variation patterns evident in the adaptations and seeks to develop a theoretical model that explains why different strategies were used.

Språkbanken Text collaborates with the Department of Philosophy, Linguistics and Theory of Science to create an annotated corpus of Xhosa, an underresourced Bantu language of South Africa (also known as isiXhosa and Xosa).