MobIE: A German Dataset for Named Entity Recognition, Entity Linking and Relation Extraction in the Mobility Domain

MobIE: A German Dataset for Named Entity Recognition, Entity Linking and Relation Extraction in the Mobility Domain Leonhard Hennig Phuc Tran Truong Aleksandra Gabryszak German Research Center for Artificial Intelligence (DFKI) Speech and Language Technology Lab {leonhard.hennig,phuc tran.truong,aleksandra.gabryszak}@dfki.de Abstract We present MobIE, a German-language dataset, which is human-annotated with 20 coarse- and fine-grained entity types and entity linking information for geographically link- able entities. The dataset consists of 3,232 so- cial media texts and traffic reports with 91K tokens, and contains 20.5K annotated enti- ties, 13.1K of which are linked to a knowl- edge base. A subset of the dataset is human- annotated with seven mobility-related, n-ary relation types, while the remaining documents are annotated using a weakly-supervised label- ing approach implemented with the Snorkel framework. To the best of our knowledge, this is the first German-language dataset that com- bines annotations for NER, EL and RE, and thus can be used for joint and multi-task learn- ing of these fundamental information extrac- tion tasks. We make MobIE public at https: //github.com/dfki-nlp/mobie. 1 Introduction Named entity recognition (NER), entity linking (EL) and relation extraction (RE) are fundamental tasks in information extraction, and a key compo- nent in numerous downstream applications, such as question answering (Yu et al., 2017) and knowledge base population (Ji and Grishman, 2011). Recent neural approaches based on pre-trained language models (e.g., BERT (Devlin et al., 2019)) have shown impressive results for these tasks when fine- tuned on supervised datasets (Akbik et al., 2018; De Cao et al., 2021; Alt et al., 2019). However, annotated datasets for fine-tuning information ex- traction models are still scarce, even in a com- paratively well-resourced language such as Ger- man (Benikova et al., 2014), and generally only contain annotations for a single task (e.g., for NER CoNLL’03 German (Tjong Kim Sang and De Meul- der, 2003), GermEval 2014 (Benikova et al., 2014); entity linking GerNED (Ploch et al., 2012)). In addition, research in multi-task (Ruder, 2017) and joint learning (Sui et al., 2020) has shown that mod- els can benefit from exploiting training signals of related tasks. To the best of our knowledge, the work of Schiersch et al. (2018) is the only dataset for German that includes two of the three tasks, namely NER and RE, in a single dataset. In this work, we present MobIE, a German- language information extraction dataset which has been fully annotated for NER, EL, and n-ary RE. The dataset is based upon a subset of documents provided by Schiersch et al. (2018), but focuses on the domain of mobility-related events, such as traffic obstructions and public transport issues. Fig- ure 1 displays an example traffic report with a Can- celed Route event. All relations in our dataset are n- ary, i.e. consist of two or more arguments, some of which are optional. Our work expands the dataset of Schiersch et al. (2018) with the following contri- butions: • We significantly extend the dataset with 1,686 annotated documents, more than doubling its size from 1,546 …

Content truncated to 500 words