| 1. |
- Alfter, David, 1986, et al.
(författare)
-
Interconnecting lexical resources and word alignment: How do learners get on with particle verbs?
- 2019
-
Ingår i: Linköping Electronic Conference Proceeding, No. 167, NEAL Proceedings of the 22nd Nordic Conference on Computational Linguistics (NoDaLiDa), September 30-October 2, Turku, Finland / Editor(s): Mareike Hartman and Barbara Plank. - Linköping university : Linköping University Electronic Press, Linköpings universitet. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present a prototype for an online exercise aimed at learners of English and Swedish that serves multiple purposes. The exercise allows learners of the aforementioned languages to train their knowledge of particle verbs receiving clues from the exercise application. The user themselves decide which clue to receive and pay in virtual currency for each, which provides us with valuable information about the utility of the clues that we provide as well as the learners willingness to trade virtual currency versus accuracy of their choice. As resources, we use list with annotated levels from the proficiency scale defined by the Common European Framework of Reference (CEFR) and a multilingual corpus with syntactic dependency relations and word annotation for all language pairs. From the latter resource, we extract translation equivalents for particle verb construction together with a list of parallel corpus examples that can be used as clues in the exercise.
|
|
| 2. |
- Alfter, David, 1986, et al.
(författare)
-
LEGATO: A flexible lexicographic annotation tool.
- 2019
-
Ingår i: Linköping Electronic Conference Proceedings, No. 167, NEAL Proceedings of the 22nd Nordic Conference on Computational Linguistics (NoDaLiDa), September 30-October 2, Turku, Finland Editor(s): Mareike Hartman and Barbara Plank. - Linköping university : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)abstract
- This article is a report from an ongoing project aiming at analyzing lexical and grammatical competences of Swedish as a Second language (L2). To facilitate lexical analysis, we need access to metalinguistic information about relevant vocabulary that L2 learners can use and understand. The focus of the current article is on the lexical annotation of the vocabulary scope for a range of lexicographical aspects, such as morphological analysis, valency, types of multi-word units, etc. We perform parts of the analysis automatically, and other parts manually. The rationale behind this is that where there is no possibility to add information automatically, manual effort needs to be added. To facilitate the latter, a tool LEGATO has been designed, implemented and currently put to active testing.
|
|
| 3. |
- Alfter, David, 1986, et al.
(författare)
-
Lärka: From Language Learning Platform to Infrastructure for Research on Language Learning
- 2019
-
Ingår i: Linköping Electronic Conference Proceedings. - Linköping : Linköping University Press. - 1650-3686 .- 1650-3740. - 9789176850343
-
Konferensbidrag (refereegranskat)abstract
- Lärka is an Intelligent Computer-Assisted Language Learning (ICALL) platform developed at Språkbanken, as a flexible and a valuable source of additional learning material (e.g. via corpusbased exercises) and a support tool for both teachers and L2 learners of Swedish and students of (Swedish) linguistics. Nowadays, Lärka is being adapted into a building block in an emerging second language research infrastructure within a larger context of the text-based research infrastructure developed by the national Swedish Language bank, Språkbanken, and SWE-CLARIN. Lärka has recently received a new responsive user interface adapted to different devices with different screen sizes. Moreover, the system has also been augmented with new functionalities. These recent additions aim at improving the usability and the usefulness of the platform for pedagogical purposes. The most important development, though, is the adaptation of the platform to serve as a component in an e-infrastructure supporting research on language learning and multilingualism. Thanks to Lärka’s service-oriented architecture, most functionalities are also available as web services which can be easily re-used by other applications.
|
|
| 4. |
- Bernardy, Jean-Philippe, 1978, et al.
(författare)
-
A Wide-Coverage Symbolic Natural Language Inference System
- 2019
-
Ingår i: Proceedings of the 22nd Nordic Conference on Computational Linguistics, 30 September – 2 October, 2019, Turku, Finland / Mareike Hartmann, Barbara Plank (Editors). - University of Linköping : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)
|
|
| 5. |
- Bernardy, Jean-Philippe, 1978, et al.
(författare)
-
Predicates as Boxes in Bayesian Semantics for Natural Language
- 2019
-
Ingår i: Proceedings of the 22nd Nordic Conference on Computational Linguistics (NoDaLiDa 2019), 30 September-2 October, 2019, Turku, Finland / Mareike Hartmann, Barbara Plank (Editors). - Linköping : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present a Bayesian approach to natural language semantics. Our main focus is on the inference task in an environment where judgments require probabilistic reasoning. We treat nouns, verbs, adjectives, etc. as unary predicates, and we model them as boxes in a bounded domain. We apply Bayesian learning to satisfy constraints expressed as premises. In this way we construct a model, by specifying boxes for the predicates. The probability of the hypothesis (the conclusion) is evaluated against the model that incorporates the premises as constraints.
|
|
| 6. |
- Ek, Adam, 1990, et al.
(författare)
-
Language Modeling with Syntactic and Semantic Representation for Sentence Acceptability Predictions
- 2019
-
Ingår i: Proceedings of the 22nd Nordic Conference on Computational Linguistics, 30 September – 2 October, 2019, Turku, Finland / Mareike Hartmann, Barbara Plank (Editors). - University of Linköping : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we investigate the effect of enhancing lexical embeddings in LSTM language models (LM) with syntactic and semantic representations. We evaluate the language models using perplexity, and we evaluate the performance of the models on the task of predicting human sentence acceptability judgments. We train LSTM language models on sentences automatically annotated with universal syntactic dependency roles (Nivre, 2016), dependency depth and universal semantic tags (Abzianidze et al., 2017) to predict sentence acceptability judgments. Our experiments indicate that syntactic tags lower perplexity, while semantic tags increase it. Our experiments also show that neither syntactic nor semantic tags improve the performance of LSTM language models on the task of predicting sentence acceptability judgments.
|
|
| 7. |
- Ermolaev, Evgheni, et al.
(författare)
-
Model selection for waste conversion efficiency and energy demands in a pilot for large-scale larvae treatment
- 2019
-
Ingår i: Linköping electronic conference proceedings. - : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. ; , s. 118-124
-
Konferensbidrag (refereegranskat)abstract
- Implementations of conversion systems for segregated food waste into larval biomass are reliant on stability of production environment and predictable outcomes. However, the knowledge is currently lacking for large scale implementation modelling allowing to adjust the process for reaching a stable production quality and predicting the treatment capacity and output product quantity. This study contributed to the development of such knowledge and investigated models for prediction of larvae mass gain in the conversion process, food waste degradation due to larval and microbial activity, heat production based on metabolic activity in the conversion process. These models were used to evaluate the amount of heat produced and compared to the demand for water removal for achieving a minimal total solids content (50%) in the treatment residue for easy larvae separation by sieving. Based on these models’ predictions and measurements of the conversion efficiency it was established that, assuming no heat loss from the system, the heat generated by the process was sufficient to achieve a desirable total solids content in the residue after 14 days of treatment. An exponential heat production from waste degradation can be expected, and thus, for wetter food waste, the treatment period can be expanded to achieve the desired residue total solids content.
|
|
| 8. |
- Fotedar, Sunney, 1989, et al.
(författare)
-
Mathematical optimization of the tactical allocation of machining resources for an efficient capacity utilization in aerospace component manufacturing
- 2019
-
Ingår i: Proceedings of the 10th Aerospace Technology Congress. - : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789175190068 ; , s. 183-188
-
Konferensbidrag (refereegranskat)abstract
- In the aerospace industry, with low volumes and many products, there is a critical need to efficiently use available manufacturing resources. Currently, at GKN Aerospace, resource allocation decisions that in many cases will last for several years are to some extent made with a short-term focus so as to minimize machining time, which results in a too high load on the most capable machines, and too low load on the less capable ones. This creates an imbalance in capacity utilization that leads to unnecessary queuing at some machines, resulting in long lead times and in an increase in tied-up capital. Tactical resource allocation on the medium to long-range planning horizon (six months to several years) aims to address this issue by allocating resources to meet the predicted future demand as effectively as possible, in order to ensure long range profitability. Our intent is to use mathematical optimization to find the best possible allocations.
|
|
| 9. |
- Hoang, M., et al.
(författare)
-
Aspect-Based Sentiment Analysis using BERT
- 2019
-
Ingår i: Proceedings of the 22nd Nordic Conference on Computational Linguistics, 30 September–2 October, 2019, Turku, Finland / Mareike Hartmann, Barbara Plank (Editors). - Sweden : Linköping University Electronic Press. - 1650-3686 .- 1650-3740. - 9789179299958
-
Konferensbidrag (refereegranskat)
|
|
| 10. |
- Jonsson, Isak, 1990
(författare)
-
Experimental aero- and thermal investigation for a next generation engine exit module.
- 2019
-
Ingår i: Linköping Electronic Conference Proceedings. - 1650-3686 .- 1650-3740. - 9789175190068 ; 162
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Beyond 2020, new efforts are required to improve the engine efficiency and fuel burn to meet the Advisory Council for Aeronautics Research in Europe (ACARE) goals for the year 2035 and 2050. The year 2050 targets aim for a 75% reduction in CO2 emissions, a 90% reduction in NOx emissions and a 65% reduction of the perceived noise relative to engine and aircraft performance of year 2000. It is commonly agreed that the geared turbofan engines have the potential to make a significant step towards the above targets. These engines will have a higher overall pressure ratio which will result in increased operating temperatures. They also have higher by-pass ratios, i.e., larger fan with relatively smaller engine cores, which has consequently a wider operating envelope for the LPT. For the turbine rear structures (TRS) this development implies that designs are needed that: can withstand higher temperatures and temperature variations, and furthermore, can operate under more severe LPT off-design conditions. GKN Aerospace in Trollhättan, with a dominating market position on the TRS for commercial aeroengines, is addressing these challenges in this program. A unique facility for experimental testing of TRS at engine-realistic flow conditions is currently available at Chalmers' Laboratory of Thermal and Fluids Science and will be used for testing of novel TRS designs for future more efficient aero-engines. The unique design of the facility includes an open test section which permits investigation of the complete TRS assemblies consisting of a TRS and a core exhaust nozzle. The facility is equipped with a 1.5 stage shrouded low-pressure turbine (LPT) providing realistic inflow to the tested TRS. The main objective at Chalmers is to gain an understanding of the flow mechanism affecting the performance of the LPT-OGV and to validate designs that take advantage of the acquired knowledge. One expected outcome of the project is the most comprehensive aerothermal validation database for TRS designs. For this new measurement methods have been developed and implemented. At the time of writing hotwire and PIV are used to investigate boundary layer development and IR thermography for steady-state heat transfer but further extensive testing is planned within the project. The most up to date results from the test campaign will accompany the already published scientific work in this presentation.
|
|
