Karsten Michael Tymann

Karsten Michael Tymann

M.Sc. Computer Science
Software Engineering,
NLP and AI

Posts by Collection

portfolio

publications

Detecting soiling on solar panels with datamining and machine learning

Published in Adjunct Proceedings of the 33rd edition of the EnviroInfo – the long standing and established international and interdisciplinary conference series on leading environmental information and communication technologies, 2019

The paper deals with detecting soiled solar panels just by means of collected data of a solar plant.

Recommended citation: Das Buch Rüdiger Schaldach, Karl-Heinz Simon, Jens Weismüller, Volker Wohlgemuth (eds.) - Environmental Informatics: Computational Sustainability: ICT methods to achieve the UN Sustainable Development Goals Adjunct Proceedings of the 33rd edition of the EnviroInfo – the long standing and established international and interdisciplinary conference series on leading environmental information and communication technologies (ISBN: 978-3-8440-6847-4) wurde im Shaker Verlag veröffentlicht. http://www.shaker.de/shop/978-3-8440-6847-4

GerVADER - A German Adaptation of the VADER Sentiment Analysis Tool for Social Media Texts

Published in Proceedings of the Conference on "Lernen, Wissen, Daten, Analysen - LWDA2019, 2019

GerVADER is a German adaptation of the sentiment classification tool VADER. VADER is a lexicon and rule-based approach in classifying sentences into positive, negative or neutral statements and puts a focus on social media texts.

Recommended citation: Karsten Michael Tymann, Matthias Lutz, Patrick Palsbröker and Carsten Gips: GerVADER - A German adaptation of the VADER sentiment analysis tool for social media texts. In Proceedings of the Conference "Lernen, Wissen, Daten, Analysen" (LWDA 2019), Berlin, Germany, September 30 - October 2, 2019. http://ceur-ws.org/Vol-2454/paper_14.pdf

EmoDex - An emotion detection tool composed of established techniques

Published in Proceedings of the Conference on "Lernen, Wissen, Daten, Analysen - LWDA2020", 2020

In this work we created an emotion analysis tool consisting of established models and techniques: Ekmanns and Plutchiks emotion models, WordEmbedding (GloVe), VADER sentiment analysis, emoji features and a RandomForest classifier. Additionally we composed a corpus based on existing corpora and with the help of distant supervision. As a result, our approach achieves an accuracy increase of up to 10% compared to other emotion analysis tools (ParallelDots and Twitter Emotion Recognition), while at the same time offering a broader set of emotion classes. In addition, adding a sentiment feature increased the accuracy by about 2%. We make the conclusion that a combination of features from multiple sources such as GloVe and VADER offer a good basis for a RandomForest classifier while only training on a very small set of texts (less than 70k sentences).

Recommended citation: Zhurakovskaya, Oxana & Steinkamp, Louis & Tymann, Karsten & Gips, Carsten. (2020). EmoDex - An emotion detection tool composed of established techniques. In Proceedings of the Conference "Lernen, Wissen, Daten, Analysen" (LWDA 2020), Online (Bonn), Germany, September 9 - 11, 2020. http://ceur-ws.org/Vol-2738/LWDA2020_paper_10.pdf

Native sentiment analysis tools vs. translation services - Comparing GerVADER and VADER

Published in Proceedings of the Conference on "Lernen, Wissen, Daten, Analysen - LWDA2020", 2020

VADER is a rule-based sentiment analysis tool for English texts with a social media focus. GerVADER is a German adaptation of VADER, which was developed following the steps of VADERs development process. VADER showed high F1 scores especially for the social media domain, whereas the German adaptation achieved much lower results within the same domain, although on other test data. In this work we examine the question of whether these differences are language-specific. Therefore we apply an improved version of GerVADER to German texts and compare the results with the application of VADER to the same texts that are automatically translated into English. The benchmarking showed, that the translation combined with VADER achieves up to 5% higher F1 scores in all test cases, which can be explained by the translation tools automatic fixing of flawed sentences. However, native language tools can still be viable, since it saves time and costs and does not need another dependency to a third party service.

Recommended citation: Tymann, Karsten & Steinkamp, Louis & Zhurakovskaya, Oxana & Gips, Carsten. (2020). Native sentiment analysis tools vs. translation services - Comparing GerVADER and VADER. In Proceedings of the Conference "Lernen, Wissen, Daten, Analysen" (LWDA 2020), Online (Bonn), Germany, September 9 - 11, 2020. http://ceur-ws.org/Vol-2738/LWDA2020_paper_9.pdf

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.