Linköping Electronic Conference Proceedings
Not a member yet
1113 research outputs found
Sort by
Annotation Management Tool: A Requirement for Corpus Construction
Get PDFWe present an annotation management tool, SweLL portal, that has been developed for the purposes of the SweLL infrastructure project for building a learner corpus of Swedish (Volodina et al., 2019). The SweLL portal has been used for supervised access to the database, for data versioning, import and export of data and metadata, statistical overview, administration of annotation tasks, monitoring of annotation tasks and reliability controls. The portal was developed driven by visions of longitudinal sustainable data storage and was partially shaped by situational needs reported by the portal users, including project managers, researchers, and annotators
Help Yourself from the Buffet: National Language Technology Infrastructure Initiative on CLARIN-IS
Get PDFIn this paper we describe how a fairly new CLARIN member is building a broad collection of national language resources for use in language technology (LT). As a CLARIN C-centre, CLARIN-IS is hosting metadata for various text and speech corpora, lexical resources, software packages and models. The providers of the resources are universities, institutions and private companies working on a national LT infrastructure initiative, Language Technology Programme for Icelandic. All deliverables of the programme are published under open licences and are freely accessible for research as well as commercial use. We provide a broad overview of the available repositories and the core publishing guidelines
Reliability of Automatic Linguistic Annotation: Native vs Non-native Texts
Get PDFWe present the results of a manual evaluation of the performance of automatic linguistic annotation on three different datasets: (1) texts written by native speakers, (2) essays written by second language (L2) learners of Swedish in the original form and (3) the normalized versions of learner-written essays. The focus of the evaluation is on lemmatization, POS-tagging, word sense disambiguation, multi-word detection and dependency annotation. Two annotators manually went through the automatic annotation on a subset of the datasets and marked up all deviations based on their expert judgments and the guidelines provided. We report Inter-Annotator Agreement between the two annotators and accuracy for the linguistic annotation quality for the three datasets, by levels and linguistic features
Flexible Metadata Schemes for Research Data Repositories.The Common Framework in Dataverse and the CMDI Use Case
Get PDFIn this paper we present an approach called Common Framework, which addresses issues of interoperability and flexibility of metadata schemes as developed by specific scientific communities, and as later supported by domain and cross-domain data repositories. The approach was triggered by a very concrete use case, namely the question how to expose Component Metadata Infrastructure (CMDI) metadata, stored in computational linguistics datasets in the DANSEASY archive, for discovery services. The work in CLARIN to push further for the development of CMDI into a standard (ISO 24622-1:2015, ISO 24622-2:2019) forms part of the background of the use case. We used the Dataverse platform to deliver proof of concepts for various elements of the Common Framework, including the recommendation of standardised elements for Dataverse instances in CLARIN. At the core of the Common Framework is a design which envisions an interaction between different microservices, possibly also hosted by various service providers. Mechanisms of semantic mapping are used throughout a pipeline which starts at a set of existing metadata standards and values at a digital research data repository (Extraction) and their analysis. This leads to an alignment of these metadata standards with others standards (Transformation) and proposes enrichments to be used by other service providers but also to be imported back to the original source (Load). Some modules applied along this pipeline are discussed in detail, together with the challenges this specific use case entails. At the same time, we also stress generic aspects, as we are convinced that this approach can also be applied in other settings, other archival platforms and other domain specific metadata schemes. The high-level goal of this exploration is to explore ways to make research data collections FAIR (Findable, Accessible, Interoperable and Re-usable), and in particular interoperable and re-usable, while preserving the rigour of domain specific indexing practices
The DECODE Database of Historical Ciphers and Keys: Version 2
Get PDFWe report recent developments of the DECODE database aimed for the systematic collection and annotation of encrypted sources: ciphertexts, keys and related documents. We released a new, more functional graphical user interface, revised some metadata features and enlarged the collection and tripled its size
Physical modeling of daily electric appliances for education by Modelica
Get PDFModelica is very useful to make physical models in various engineering fields such as mechanical, electrical, thermal, fluid systems, etc. This capability of Modelica is also useful to educate students and engineers about many physical areas using simulation. The authors are posting serialized articles in a technical magazine about physical modeling of daily electric appliances by Modelica to educate readers about both physics and Modelica language in Japan. This paper introduces some examples of physical modeling of various appliances such as electric minicar, dryer and speaker by Modelica
Bringing Automatic Scoring into the Classroom – Measuring the Impact of Automated Analytic Feedback on Student Writing Performance
Get PDFWhile many methods for automatically scoring student writings have been proposed, few studies have inquired whether such scores constitute effective feedback improving learners’ writing quality. In this paper, we use an EFL email dataset annotated according to five analytic assessment criteria to train a classifier for each criterion, reaching human-machine agreement values (kappa) between .35 and .87. We then perform an intervention study with 112 lower secondary students in which participants in the feedback condition received stepwise automatic feedback for each criterion while students in the control group received only a description of the respective scoring criterion. We manually and automatically score the resulting revisions to measure the effect of automated feedback and find that students in the feedback condition improved more than in the control group for 2 out of 5 criteria. Our results are encouraging as they show that even imperfect automated feedback can be successfully used in the classroom
Electrification of an Entrainment Calciner in a Cement Kiln System – Heat Transfer Modelling and Simulations
Get PDFCarbon capture and storage may be applied to reduce the CO2 emissions from a cement plant. However, this often results in complex CO2 capture solutions. To simplify the capturing process, an alternative is to electrify the cement calciner. This study covers the feasibility of electrifying an existing calciner by inserting electrically heated rods in the calciner. An existing entrainment calciner in a Norwegian cement plant is used as a case study. A model is developed to quantify the aspects concerning the feasibility of the calciner. The model first estimates the possible area of inserted rods in the available space. A mass and energy balance is then performed to estimate the heat duty of the heating rods. Further, a radiation heat transfer model is included to identify the feasibility of transferring heat from the rods to the raw meal. Finally, the model includes the design of the heating rod to estimate the required number of heating elements. The results indicate that it is technically feasible to electrify the calciner. The total heat duty of the calciner is 77 MW, with 68 MW for meal preheating and calcining, and 9 MW for gas preheating. 2570 heating rods are required, operating at 1150 °C in the gas preheating zone and 1050 °C in the meal preheating and calcining zone. The feasible heat flux is 26-34 kW/m² for gas preheating, 35-80 kW/m² for meal preheating and 30-50 kW/m² for calcination. However, some challenges related to recuperating the heat from the gas and maintenance of the system must be studied further
On solving Fault Detection Problem and Risk Estimation Monitoring with Deep Neural Networks and Postprocessing
Get PDFIn this study, we consider fault prediction problem and production process risk monitoring based on observational data. We consider case, when there are no variables, by which one could classify the situation preceding to the fault. We propose an approach that is based on a specific auxiliary risk variable and modifications of the modeling accuracy estimation criterion, so the fault detection problem is reduced to supervised learning problem. We use deep learning and examine different model architectures. Trained model produces the risk estimations for new observations, then we use postprocessing to interpret the estimations to decision-maker. This work confirms that data-driven risk estimation can be integrated into digital services to successfully manage plant operational changes and support plant prescriptive maintenance. This was demonstrated with data from a commercial circulating fluidized bed firing various biomass and residues but is generally applicable to other production plants
Simulation and Impact of different Optimization Parameters on CO2 Capture Cost
Get PDFThe influence of different process parameters/factors on CO2 capture cost, in a standard amine based CO2 capture process was studied through process simulation and cost estimation. The most influential factor was found to be the CO2 capture efficiency. This led to investigation of routes for capturing more than 85% of CO2. The routes are by merely increasing the solvent flow or by increasing the absorber packing height. The cost-efficient route was found to be by increasing the packing height of the absorber. This resulted in 20% less cost compared to capturing 90% CO2 by increasing only the solvent flow. The cost optimum absorber packing height was 12 m (12 stages). The cost optimum temperature difference in the lean/rich heat exchanger was 5 °C. A case with a combination of the two cost optimum parameters achieved a 4% decrease in capture cost compared to the base case. The results highlight the significance of performing cost optimization of CO2 capture processes