Hello!
Hi! I'm Anders Björkelund.
I'm a research engineer in
the Computational Biology and
Biological Physics group at
the Department of Astronomy and
Theoretical Physics at Lund
University. Despite the apparent affiliation, I'm not a
phycisist but a computer scientist specialized in machine
learning. I am currently working in a VINNOVA project with the
hospital in Lund where we are applying machine learning methods to
clinical data with regard to heart disease.
I am originally from Lund, Sweden and did my undergraduate in
computer science at Lund University, where I subsequently also
worked both on NLP and knowledge representation and reasoning (i.e.,
old-school AI) for industrial robots. After that I spent a number of
years at the
Institute for Natural
Language Processing in Stuttgart, Germany where I also earned my
PhD degree. My work in Stuttgart in general and dissertation work in
particular revolved around predicting linguistic structure,
typically with approximate search and often the specific problem of
natural language parsing.
Table of contents
Selected Publications
- Anders Björkelund. Online Learning of Latent
Linguistic Structure with Approximate Search, 2019. Doctoral
dissertation, University of
Stuttgart. [UB
Uni Stuttgart]
- Anders Björkelund, Agnieszka Falenska, Xiang Yu, and Jonas
Kuhn. IMS at the CoNLL 2017 UD Shared Task: CRFs and
Perceptrons Meet Neural Networks. Proceedings of the
CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to
Universal Dependencies,
2017. [ACL
Anthology]
- Anders Björkelund, Agnieszka Faleńska, Wolfgang Seeker, and
Jonas Kuhn. How to Train Dependency Parsers with Inexact
Search for Joint Sentence Boundary Detection and Parsing of Entire
Documents. Proceedings of the 54th Annual Meeting of
the Association for Computational Linguistics (Volume 1: Long
Papers),
2016. [ACL Anthology]
[poster]
[code]
- Anders Björkelund and Joakim Nivre. Non-Deterministic
Oracles for Unrestricted Non-Projective Dependency
Parsing. Proceedings of the 14th International
Conference on Parsing Technologies,
2015. [ACL
Anthology]
[slides]
[code]
- Agnieszka Faleńska, Anders Björkelund, Özlem Çetinoğlu, and
Wolfgang Seeker. Stacking or Supertagging for Depdenency
Parsing -- What's the difference? Proceedings of the
14th International Conference on Parsing Technologies,
2015. [ACL Anthology]
- Anders Björkelund and Jonas Kuhn. Learning Structured
Perceptrons for Coreference Resolution with Latent Antecedents and
Non-local Features. Proceedings of the 52nd Annual
Meeting of the Association for Computational Linguistics (Volume
1: Long Papers),
2014. [ACL
Anthology]
[slides]
[code]
- Anders Björkelund, Özlem Çetinoğlu, Richárd Farkas, Thomas
Müller and Wolfgang Seeker. (Re)ranking Meets
Morphosyntax: State-of-the-art Results from the SPMRL 2013 Shared
Task. In Proceedings of the Fourth Workshop on
Statistical Parsing of Morphologically-Rich Languages,
2013. [ACL Anthology]
[slides]
[poster]
- Sameer Pradhan, Alessandro Moschitti, Nianwen Xue, Hwee Tou
Ng, Anders Björkelund, Olga Uryupina, Yuchen Zhang and Zhi
Zhong. Towards Robust Linguistic Analysis using
OntoNotes. In Proceedings of the Seventeenth
Conference on Computational Natural Language Learning,
2013. [ACL Anthology]
- Anders Björkelund and Richárd Farkas. Data-driven
Multilingual Coreference Resolution using Resolver
Stacking. In Proceedings of the Joint Conference on
EMNLP and CoNLL: Shared Task,
2012. [ACL Anthology]
[slides]
[poster]
[code]
- Anders Björkelund, Lisett Edström, Mathias Haage, Jacek Malec,
Klas Nilsson, Pierre Nugues, Sven Gestegård Robertz, Denis
Störkle, Anders Blomdell, Rolf Johansson, Magnus Linderoth, Anders
Nilsson, Anders Robertsson, Andreas Stolt and Herman
Bruyninkcx, On the Integration of Skilled Robot Motions
for Productivity in Manufacturing. In Proceedings of
IEEE International Symposium on Assembly in Manufacturing,
2011. [IEEE]
- Anders Björkelund and Pierre Nugues. Exploring
lexicalized features for coreference
resolution. In Proceedings of the 15th Conference on
Computational Natural Language Learning (CoNLL-2011): Shared
Task,
2011. [ACL Anthology]
[slides]
- Jacob Persson, Axel Gallois, Anders Björkelund, Love Hafdell,
Mathias Haage, Jacek Malec, Klas Nilsson, and Pierre
Nugues, A Knowledge Integration Framework for
Robotics. In Proceedings of the joint conference of
the 41st International Symposium on Robotics and the 6th German
Conference on Robotics,
2010. [IEEE]
- Anders Björkelund, Love Hafdell, and Pierre
Nugues, Multilingual semantic role
labeling. In Proceedings of the Thirteenth Conference
on Computational Natural Language Learning (CoNLL): Shared
Task,
2009. [ACL Anthology]
[slides]
[poster]
[code]
Open Source Software
- Dependency-based SRL system
I built a system for
dependency-based semantic role labeling during my master's
thesis. A cleaned up version of the code
is available on
Google Code, along with Bernd Bohnet's dependency
parsers. Note that as Google Code closed the code repositories
were taken down. All that remains are some tar balls for download
but they are nevertheless sufficient to access the code.
- Coreference Resolver -- old
With Richárd Farkas
I wrote up a coreference resolver for
the CoNLL 2012 Shared
Task (Björkelund and Farkas,
2012; download
source code). I used the same system (with some
modifcations) for the COLING 2012 paper Phrase-structures
and Dependencies for End-to-End Coreference Resolution
(Björkelund and Kuhn,
2012; download
source code).
- Coreference Resolver -- (somewhat) newer
In our
ACL 2014 paper, Jonas Kuhn and I describe a coreference system
based on latent antecedents and non-local features (Björkelund
and Kuhn, 2014;
download source code). It is clearly superior to the systems
above, and obtained the best results at the time on all CoNLL
2012 Shared Task data sets (Arabic, Chinese, and English).
- Transition-based Dependency Parser
For our paper
on non-deterministic oracles for transition-based parsers
(Björkelund and Nivre,
2015; download source
code), I implemented a transition-based dependency parser
with the oracles described in the paper. In subsequent work with
Agnieszka Falenska, Jonas Kuhn, and Wolfgang Seeker we extended
this parser to also handle sentence splitting (Björkelund et
al., 2016; download source
code).
Contact
ab (AT-symbol) bjorkelund.org
Last updated 2020-02-16