Maxim Konca

Staff Member

Goethe-Universität Frankfurt am Main
Robert-Mayer-Straße 10
Room 401d
D-60325 Frankfurt am Main
D-60054 Frankfurt am Main (use for package delivery)
Postfach / P.O. Box: 154
Phone:
Mail:

Office Hour: TBA

Publications

2024

Maxim Konca, Andy Lücking and Alexander Mehler. May, 2024. German SRL: Corpus Construction and Model Training. Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), 7717–7727.
BibTeX
@inproceedings{Konca:et:al:2024,
  abstract  = {A useful semantic role-annotated resource for training semantic
               role models for the German language is missing. We point out some
               problems of previous resources and provide a new one due to a
               combined translation and alignment process: The gold standard
               CoNLL-2012 semantic role annotations are translated into German.
               Semantic role labels are transferred due to alignment models.
               The resulting dataset is used to train a German semantic role
               model. With F1-scores around 0.7, the major roles achieve competitive
               evaluation scores, but avoid limitations of previous approaches.
               The described procedure can be applied to other languages as well.},
  address   = {Torino, Italy},
  author    = {Konca, Maxim and L{\"u}cking, Andy and Mehler, Alexander},
  booktitle = {Proceedings of the 2024 Joint International Conference on Computational
               Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
  editor    = {Calzolari, Nicoletta and Kan, Min-Yen and Hoste, Veronique and Lenci, Alessandro
               and Sakti, Sakriani and Xue, Nianwen},
  month     = {may},
  pages     = {7717--7727},
  publisher = {ELRA and ICCL},
  title     = {{G}erman {SRL}: Corpus Construction and Model Training},
  url       = {https://aclanthology.org/2024.lrec-main.682},
  poster    = {https://www.texttechnologylab.org/wp-content/uploads/2024/05/LREC_2024_Poster_GERMAN_SRL.pdf},
  year      = {2024}
}

2023

Alexander Mehler, Mevlüt Bagci, Alexander Henlein, Giuseppe Abrami, Christian Spiekermann, Patrick Schrottenbacher, Maxim Konca, Andy Lücking, Juliane Engel, Marc Quintino, Jakob Schreiber, Kevin Saukel and Olga Zlatkin-Troitschanskaia. 2023. A Multimodal Data Model for Simulation-Based Learning with Va.Si.Li-Lab. Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management, 539–565.
BibTeX
@inproceedings{Mehler:et:al:2023:a,
  abstract  = {Simulation-based learning is a method in which learners learn
               to master real-life scenarios and tasks from simulated application
               contexts. It is particularly suitable for the use of VR technologies,
               as these allow immersive experiences of the targeted scenarios.
               VR methods are also relevant for studies on online learning, especially
               in groups, as they provide access to a variety of multimodal learning
               and interaction data. However, VR leads to a trade-off between
               technological conditions of the observability of such data and
               the openness of learner behavior. We present Va.Si.Li-Lab, a VR-L
               ab for Simulation-based Learn ing developed to address this trade-off.
               Va.Si.Li-Lab uses a graph-theoretical model based on hypergraphs
               to represent the data diversity of multimodal learning and interaction.
               We develop this data model in relation to mono- and multimodal,
               intra- and interpersonal data and interleave it with ISO-Space
               to describe distributed multiple documents from the perspective
               of their interactive generation. The paper adds three use cases
               to motivate the broad applicability of Va.Si.Li-Lab and its data
               model.},
  address   = {Cham},
  author    = {Mehler, Alexander and Bagci, Mevl{\"u}t and Henlein, Alexander
               and Abrami, Giuseppe and Spiekermann, Christian and Schrottenbacher, Patrick
               and Konca, Maxim and L{\"u}cking, Andy and Engel, Juliane and Quintino, Marc
               and Schreiber, Jakob and Saukel, Kevin and Zlatkin-Troitschanskaia, Olga},
  booktitle = {Digital Human Modeling and Applications in Health, Safety, Ergonomics
               and Risk Management},
  editor    = {Duffy, Vincent G.},
  isbn      = {978-3-031-35741-1},
  pages     = {539--565},
  publisher = {Springer Nature Switzerland},
  title     = {A Multimodal Data Model for Simulation-Based Learning with Va.Si.Li-Lab},
  year      = {2023},
  doi       = {10.1007/978-3-031-35741-1_39}
}

2022

Alexander Mehler, Maxim Konca, Marie-Theres Nagel, Andy Lücking and Olga Zlatkin-Troitschanskaia. March, 2022. On latent domain-specific textual preferences in solving Internet-based generic tasks among graduates/young professionals from three domains.
BibTeX
@misc{Mehler:et:al:2022,
  author    = {Mehler, Alexander and Konca, Maxim and Nagel, Marie-Theres and L\"{u}cking, Andy
               and Zlatkin-Troitschanskaia, Olga},
  year      = {2022},
  month     = {03},
  howpublished = {Presentation at BEBF 2022},
  title     = {On latent domain-specific textual preferences in solving Internet-based
               generic tasks among graduates/young professionals from three domains},
  abstract  = {Although Critical Online Reasoning (COR) is often viewed as a
               general competency (e.g. Alexander et al. 2016), studies have
               found evidence supporting their domain-specificity (Toplak et
               al. 2002). To investigate this assumption, we focus on commonalities
               and differences in textual preferences in solving COR-related
               tasks between graduates/young professionals from three domains.
               For this reason, we collected data by requiring participants to
               solve domain-specific (DOM-COR) and generic (GEN-COR) tasks in
               an authentic Internet-based COR performance assessment (CORA),
               allowing us to disentangle the assumed components of COR abilities.
               Here, we focus on GEN-COR to distinguish between different groups
               of graduates from the three disciplines in the context of generic
               COR tasks. We present a computational model for educationally
               relevant texts that combines features at multiple levels (lexical,
               syntactic, semantic). We use machine learning to predict domain-specific
               group membership based on documents consulted during task solving.
               A major contribution of our analyses is a multi-part text classification
               system that contrasts human annotation and rating of the documents
               used with a semi-automatic classification to predict the document
               type of web pages. That is, we work with competing classifications
               to support our findings. In this way, we develop a computational
               linguistic model that correlates GEN-COR abilities with properties
               of documents consulted for solving the GEN-COR tasks. Results
               show that participants from different domains indeed inquire different
               sets of online sources for the same task. Machine learning-based
               classifications show that the distributional differences can be
               reproduced by computational linguistic models.},
  pdf       = {https://www.texttechnologylab.org/wp-content/uploads/2022/04/On_latent_domain-specific_textual_preferences_in_solving_Internet-based_generic_tasks_among_graduates__young_professionals_from_three_domains.pdf}
}
Maxim Konca, Andy Lücking, Alexander Mehler, Marie-Theres Nagel and Olga Zlatkin-Troitschanskaia. April, 2022. Computational educational linguistics for `Critical Online Reasoning' among young professionals in medicine, law and teaching.
BibTeX
@misc{Konca:et:al:2022,
  author    = {Konca, Maxim and L{\"u}cking, Andy and Mehler, Alexander and Nagel, Marie-Theres
               and Zlatkin-Troitschanskaia, Olga},
  howpublished = {Presentation given at the AERA annual meeting, 21.-26.04. 2022, WERA symposium},
  month     = {04},
  title     = {Computational educational linguistics for `Critical Online Reasoning'
               among young professionals in medicine, law and teaching},
  year      = {2022},
  pdf       = {https://www.texttechnologylab.org/wp-content/uploads/2022/10/BRIDGE_WERA_AERA-2022_reduce.pdf}
}

2021

Maxim Konca, Alexander Mehler, Daniel Baumartz and Wahed Hemati. 2021. From distinguishability to informativity. A quantitative text model for detecting random texts.. Language and Text: Data, models, information and applications, 356:145–162.
BibTeX
@article{Konca:et:al:2021,
  title     = {From distinguishability to informativity. A quantitative text
               model for detecting random texts.},
  author    = {Konca, Maxim and Mehler, Alexander and Baumartz, Daniel and Hemati, Wahed},
  journal   = {Language and Text: Data, models, information and applications},
  volume    = {356},
  pages     = {145--162},
  year      = {2021},
  editor    = {Adam Paw{\l}owski, Jan Ma{\v{c}}utek, Sheila Embleton and George Mikros},
  publisher = {John Benjamins Publishing Company},
  doi       = {10.1075/cilt.356.10kon}
}

2020

Alexander Mehler, Wahed Hemati, Pascal Welke, Maxim Konca and Tolga Uslu. 2020. Multiple Texts as a Limiting Factor in Online Learning: Quantifying (Dis-)similarities of Knowledge Networks. Frontiers in Education, 5:206.
BibTeX
@article{Mehler:Hemati:Welke:Konca:Uslu:2020,
  abstract  = {We test the hypothesis that the extent to which one obtains information
               on a given topic through Wikipedia depends on the language in
               which it is consulted. Controlling the size factor, we investigate
               this hypothesis for a number of 25 subject areas. Since Wikipedia
               is a central part of the web-based information landscape, this
               indicates a language-related, linguistic bias. The article therefore
               deals with the question of whether Wikipedia exhibits this kind
               of linguistic relativity or not. From the perspective of educational
               science, the article develops a computational model of the information
               landscape from which multiple texts are drawn as typical input
               of web-based reading. For this purpose, it develops a hybrid model
               of intra- and intertextual similarity of different parts of the
               information landscape and tests this model on the example of 35
               languages and corresponding Wikipedias. In the way it measures
               the similarities of hypertexts, the article goes beyond existing
               approaches by examining their structural and semantic aspects
               intra- and intertextually. In this way it builds a bridge between
               reading research, educational science, Wikipedia research and
               computational linguistics.},
  author    = {Mehler, Alexander and Hemati, Wahed and Welke, Pascal and Konca, Maxim
               and Uslu, Tolga},
  doi       = {10.3389/feduc.2020.562670},
  issn      = {2504-284X},
  journal   = {Frontiers in Education},
  pages     = {206},
  title     = {Multiple Texts as a Limiting Factor in Online Learning: Quantifying
               (Dis-)similarities of Knowledge Networks},
  url       = {https://www.frontiersin.org/article/10.3389/feduc.2020.562670},
  pdf       = {https://www.frontiersin.org/articles/10.3389/feduc.2020.562670/pdf},
  volume    = {5},
  year      = {2020}
}