1 code implementation • FNP (COLING) 2020 • Felix Armbrust, Henry Schäfer, Roman Klinger
We answer this question by training and evaluating an end-to-end deep learning approach (based on BERT and GloVe embeddings) to predict the financial and environmental performance of the company from the “Management’s Discussion and Analysis of Financial Conditions and Results of Operations” (MD&A) section of 10-K (yearly) and 10-Q (quarterly) filings.
no code implementations • WASSA (ACL) 2022 • Valentino Sabbatino, Enrica Troiano, Antje Schweitzer, Roman Klinger
This raises the question if the association is purely a product of the learned affective imports inherent to semantic meanings, or is also an effect of other features of words, e. g., morphological and phonological patterns.
no code implementations • 26 Mar 2024 • Christopher Bagdon, Prathamesh Karmalker, Harsha Gurulingappa, Roman Klinger
Labeling corpora constitutes a bottleneck to create models for new tasks or domains.
no code implementations • 15 Mar 2024 • Aswathy Velutharambath, Amelie Wührl, Roman Klinger
As the argumentation stems from genuine belief, it may be unlikely to exhibit the linguistic properties associated with deception or lying.
no code implementations • 19 Feb 2024 • Amelie Wührl, Dustin Wright, Roman Klinger, Isabelle Augenstein
Distorted science communication harms individuals and society as it can lead to unhealthy behavior change and decrease trust in scientific institutions.
no code implementations • 5 Feb 2024 • Patrick Bareiß, Roman Klinger, Jeremy Barnes
This is particularly of interest when we have access to a multilingual large language model, because we could request labels with English prompts even for non-English data.
no code implementations • 2 Feb 2024 • Amelie Wührl, Yarik Menchaca Resendiz, Lara Grimminger, Roman Klinger
In a study with trained annotation experts we prompt them to find evidence for biomedical claims, and observe how they refine search queries for their evidence search.
no code implementations • 14 Dec 2023 • Maximilian Wegge, Roman Klinger
when funeral events are over-represented for instances labeled with sadness, despite the emotion of pride being more appropriate here.
no code implementations • 5 Sep 2023 • Roman Klinger
(1) Emotions are events; and this perspective is the fundament in natural language processing for emotion role labeling.
no code implementations • 9 Aug 2023 • Yarik Menchaca Resendiz, Roman Klinger
We evaluate the method on emotion-conditioned text generation with a focus on event reports and compare it to manually designed prompts that also act as the seed for the optimization procedure.
no code implementations • 26 Jul 2023 • Yarik Menchaca Resendiz, Roman Klinger
They put the assessment of the situation on the spot, for instance regarding the own control or the responsibility for what happens.
1 code implementation • 5 Jun 2023 • Aswathy Velutharambath, Roman Klinger
Verbal deception has been studied in psychology, forensics, and computational linguistics for a variety of reasons, like understanding behaviour patterns, identifying false testimonies, and detecting deception in online communication.
no code implementations • 26 May 2023 • Maximilian Wegge, Roman Klinger
We show that experiencer detection in text is a challenging task, with a precision of . 82 and a recall of . 56 (F1 =. 66).
no code implementations • 11 Apr 2023 • Amelie Wührl, Lara Grimminger, Roman Klinger
This mismatch can be mitigated by adapting the social media input to mimic the focused nature of common training claims.
no code implementations • 21 Oct 2022 • Maximilian Wegge, Enrica Troiano, Laura Oberländer, Roman Klinger
Emotion classification in NLP assigns emotions to texts, such as sentences or paragraphs.
no code implementations • ArgMining (ACL) 2022 • Amelie Wührl, Roman Klinger
To make user-generated content checkable by existing models, we propose to reformulate the social-media input in such a way that the resulting claim mimics the claim characteristics in established datasets.
1 code implementation • COLING 2022 • Flor Miriam Plaza-del-Arco, María-Teresa Martín-Valdivia, Roman Klinger
This raises the question how to prompt a natural language inference model for zero-shot learning emotion classification.
no code implementations • 10 Jun 2022 • Enrica Troiano, Laura Oberländer, Roman Klinger
We analyze the suitability of appraisal theories for emotion analysis in text with the goal of understanding if appraisal concepts can reliably be reconstructed by annotators, if they can be predicted by text classifiers, and if appraisal concepts help to identify emotion categories.
no code implementations • LREC 2022 • Isabelle Mohr, Amelie Wührl, Roman Klinger
The corpus consists of 300 tweets, each annotated with medical named entities and relations.
no code implementations • LREC 2022 • Amelie Wührl, Roman Klinger
Named entity recognition and relation extraction are methods to structure information that is available in unstructured text.
no code implementations • LREC 2022 • Enrica Troiano, Laura Oberländer, Maximilian Wegge, Roman Klinger
In addition, we link them to the event they found salient (which can be different for different experiencers in a text) by annotating event properties, or appraisals (e. g., the perceived event undesirability, the uncertainty of its outcome).
no code implementations • 24 Feb 2022 • Valentino Sabbatino, Enrica Troiano, Antje Schweitzer, Roman Klinger
This raises the question if the association is purely a product of the learned affective imports inherent to semantic meanings, or is also an effect of other features of words, e. g., morphological and phonological patterns.
1 code implementation • NAACL 2022 • Emīls Kadiķis, Vaibhav Srivastav, Roman Klinger
We do this by testing how well the pre-trained models perform on the \alpha{}nli task when just comparing sentence embeddings with cosine similarity to what the performance that is achieved when training a classifier on top of these embeddings.
no code implementations • WASSA (ACL) 2022 • Anne Kreuter, Kai Sassenberg, Roman Klinger
We investigate this approach for personality profiling, and evaluate BERT classifiers fine-tuned on such psychometric test items for the big five personality traits (openness, conscientiousness, extraversion, agreeableness, neuroticism) and analyze various augmentation strategies regarding their potential to address the challenges coming with such a small corpus.
no code implementations • WASSA (ACL) 2022 • Anna Khlyzova, Carina Silberer, Roman Klinger
The emotions of anger and sadness are best predicted with a multimodal model, while text alone is sufficient for disgust, joy, and surprise.
no code implementations • 29 Oct 2021 • Enrica Troiano, Aswathy Velutharambath, Roman Klinger
With this paper, we aim at providing a comprehensive discussion of the styles that have received attention in the transfer task.
no code implementations • 21 Sep 2021 • Flor Miriam Plaza-del-Arco, Sercan Halat, Sebastian Padó, Roman Klinger
The recognition of hate speech and offensive language (HOF) is commonly formulated as a classification task to decide if a text contains HOF.
no code implementations • KONVENS (WS) 2021 • Felix Casel, Amelie Heindl, Roman Klinger
It states that emotions are a coordinated process of various subcomponents, in reaction to an event, namely the subjective feeling, the cognitive appraisal, the expression, a physiological bodily reaction, and a motivational action tendency.
no code implementations • KONVENS (WS) 2021 • Bao Minh Doan Dang, Laura Oberländer, Roman Klinger
Emotion stimulus extraction is a fine-grained subtask of emotion analysis that focuses on identifying the description of the cause behind an emotion expression from a text passage (e. g., in the sentence "I am happy that I passed my exam" the phrase "passed my exam" corresponds to the stimulus.).
1 code implementation • ICLR 2022 • Sean Papay, Roman Klinger, Sebastian Padó
However, the CRF's Markov assumption makes it impossible for CRFs to represent distributions with \textit{nonlocal} dependencies, and standard CRFs are unable to respect nonlocal constraints of the data (such as global arity constraints on output labels).
Ranked #6 on Semantic Role Labeling on OntoNotes
no code implementations • NAACL (BioNLP) 2021 • Amelie Wührl, Roman Klinger
We aim to fill this research gap and annotate a corpus of 1200 tweets for implicit and explicit biomedical claims (the latter also with span annotations for the claim phrase).
no code implementations • EACL (WASSA) 2021 • Lara Grimminger, Roman Klinger
Further, we annotate if the tweet is written in an offensive style.
no code implementations • EACL (WASSA) 2021 • Enrica Troiano, Sebastian Padó, Roman Klinger
When humans judge the affective content of texts, they also implicitly assess the correctness of such judgment, that is, their confidence.
no code implementations • EACL (WASSA) 2021 • Jan Hofmann, Enrica Troiano, Roman Klinger
Appraisal theories explain how the cognitive evaluation of an event leads to a particular emotion.
no code implementations • COLING 2020 • Enrica Troiano, Roman Klinger, Sebastian Pad{\'o}
Machine translation provides powerful methods to convert text between languages, and is therefore a technology enabling a multilingual world.
no code implementations • COLING (PEOPLES) 2020 • Laura Oberländer, Kevin Reich, Roman Klinger
Is it a particular target (everybody loves X) or a stimulus (doing X makes everybody sad)?
1 code implementation • Joint Conference on Lexical and Computational Semantics 2020 • Laura Oberländer, Roman Klinger
"), and (3) as clause classification ("Does this clause contain the emotion stimulus?").
no code implementations • EMNLP 2020 • Sean Papay, Roman Klinger, Sebastian Padó
Span identification (in short, span ID) tasks such as chunking, NER, or code-switching detection, ask models to identify and classify relevant spans in a text.
1 code implementation • WS 2020 • David Helbig, Enrica Troiano, Roman Klinger
We propose the task of emotion style transfer, which is particularly challenging, as emotions (here: anger, disgust, fear, joy, sadness, surprise) are on the fence between content and style.
no code implementations • COLING 2020 • Jan Hofmann, Enrica Troiano, Kai Sassenberg, Roman Klinger
Automatic emotion categorization has been predominantly formulated as text classification in which textual units are assigned to an emotion from a predefined inventory, for instance following the fundamental emotion classes proposed by Paul Ekman (fear, joy, anger, disgust, sadness, surprise) or Robert Plutchik (adding trust, anticipation).
1 code implementation • LREC 2020 • Thomas Haider, Steffen Eger, Evgeny Kim, Roman Klinger, Winfried Menninghaus
Thus, we conceptualize a set of aesthetic emotions that are predictive of aesthetic appreciation in the reader, and allow the annotation of multiple labels per line to capture mixed emotions within their context.
no code implementations • LREC 2020 • Valentino Sabbatino, Laura Bostan, Roman Klinger
Obituaries contain information about people's values across times and cultures, which makes them a useful resource for exploring cultural history.
no code implementations • LREC 2020 • Laura Bostan, Evgeny Kim, Roman Klinger
Most research on emotion analysis from text focuses on the task of emotion classification or emotion intensity regression.
no code implementations • 6 Sep 2019 • Deniz Cevher, Sebastian Zepf, Roman Klinger
We use off-the-shelf tools for emotion detection in audio and face and compare that to a neural transfer learning approach for emotion recognition from text which utilizes existing resources from other domains.
no code implementations • WS 2019 • Evgeny Kim, Roman Klinger
Our analysis shows that stories written by humans convey character emotions along various non-verbal channels.
1 code implementation • 24 Jun 2019 • Jeremy Barnes, Roman Klinger
As expected, the choice of annotated source language for projection to a target leads to better results for source-target language pairs which are similar.
no code implementations • 6 Jun 2019 • Evgeny Kim, Roman Klinger
Our analysis shows that stories written by humans convey character emotions along various non-verbal channels.
no code implementations • ACL 2019 • Enrica Troiano, Sebastian Padó, Roman Klinger
Sentiment analysis has a range of corpora available across multiple languages.
no code implementations • WS 2019 • Laura Bostan, Roman Klinger
Adjective phrases like "a little bit surprised", "completely shocked", or "not stunned at all" are not handled properly by currently published state-of-the-art emotion classification and intensity prediction systems which use pre-dominantly non-contextualized word embeddings as input.
no code implementations • NAACL 2019 • Evgeny Kim, Roman Klinger
The development of a fictional plot is centered around characters who closely interact with each other forming dynamic social networks.
no code implementations • NAACL 2019 • Robert McHardy, Heike Adel, Roman Klinger
We therefore propose a novel model for satire detection with an adversarial component to control for the confounding variable of publication source.
no code implementations • EMNLP 2018 • Heike Adel, Laura Ana Maria Bostan, Sean Papay, Sebastian Pad{\'o}, Roman Klinger
As a result, comparability of models across tasks is missing and their applicability to new tasks is limited.
no code implementations • WS 2018 • Roman Klinger, Orphée De Clercq, Saif M. Mohammad, Alexandra Balahur
Here, for the first time, we propose a shared task where systems have to predict the emotions in a large automatically labeled dataset of tweets without access to words denoting emotions.
no code implementations • 31 Aug 2018 • Florian Strohm, Roman Klinger
We select an appropriate scope detection method for modifiers of emotion words, incorporate it in a document-level emotion classification model as additional bag of words and show that this approach improves the performance of emotion classification.
no code implementations • 9 Aug 2018 • Evgeny Kim, Roman Klinger
Emotions are a crucial part of compelling narratives: literature tells us about people with goals, desires, passions, and intentions.
no code implementations • COLING 2018 • Laura-Ana-Maria Bostan, Roman Klinger
Based on this aggregation, we perform the first cross-corpus classification experiments in the spirit of future research enabled by this paper, in order to gain insight and a better understanding of differences of models inferred from the data.
1 code implementation • COLING 2018 • Jeremy Barnes, Roman Klinger, Sabine Schulte im Walde
Our analysis shows that our model performs comparably to state-of-the-art approaches on domains that are similar, while performing significantly better on highly divergent domains.
no code implementations • COLING 2018 • Evgeny Kim, Roman Klinger
We aim at filling this gap and present a publicly available corpus based on Project Gutenberg, REMAN (Relational EMotion ANnotation), manually annotated for spans which correspond to emotion trigger phrases and entities/events in the roles of experiencers, targets, and causes of the emotion.
no code implementations • ACL 2018 • Matthias Hartung, Hendrik ter Horst, Frank Grimm, Tim Diekmann, Roman Klinger, Philipp Cimiano
Supervised machine learning algorithms require training data whose generation for complex relation extraction tasks tends to be difficult.
1 code implementation • 12 Jun 2018 • Jeremy Barnes, Roman Klinger, Sabine Schulte im Walde
Our analysis shows that our model performs comparably to state-of-the-art approaches on domains that are similar, while performing significantly better on highly divergent domains.
1 code implementation • ACL 2018 • Jeremy Barnes, Roman Klinger, Sabine Schulte im Walde
Sentiment analysis in low-resource languages suffers from a lack of annotated corpora to estimate high-performing models.
Cross-Lingual Sentiment Classification Machine Translation +5
no code implementations • 26 Sep 2017 • Hendrik ter Horst, Matthias Hartung, Roman Klinger, Matthias Zwick, Philipp Cimiano
In the context of personalized medicine, text mining methods pose an interesting option for identifying disease-gene associations, as they can be used to generate novel links between diseases and genes which may complement knowledge from structured databases.
no code implementations • WS 2017 • Jeremy Barnes, Roman Klinger, Sabine Schulte im Walde
We show that Bi-LSTMs perform well across datasets and that both LSTMs and Bi-LSTMs are particularly good at fine-grained sentiment tasks (i. e., with more than two classes).
no code implementations • RANLP 2017 • Camilo Thorne, Roman Klinger
In this paper, we discuss this task and propose different approaches for confidence estimation and a pipeline to evaluate such methods.
no code implementations • WS 2017 • Maximilian K{\"o}per, Evgeny Kim, Roman Klinger
Our submission to the WASSA-2017 shared task on the prediction of emotion intensity in tweets is a supervised learning method with extended lexicons of affective norms.
no code implementations • WS 2017 • Matthias Hartung, Roman Klinger, Franziska Schmidtke, Lars Vogel
Social media are used by an increasing number of political actors.
no code implementations • WS 2017 • Hendrik Schuff, Jeremy Barnes, Julian Mohme, Sebastian Pad{\'o}, Roman Klinger
There is a rich variety of data sets for sentiment analysis (viz., polarity and subjectivity classification).
no code implementations • WS 2017 • Evgeny Kim, Sebastian Pad{\'o}, Roman Klinger
Literary genres are commonly viewed as being defined in terms of content and stylistic features.
no code implementations • 21 Jul 2016 • Janik Jaskolski, Fabian Siegberg, Thomas Tibroni, Philipp Cimiano, Roman Klinger
The popularity of distance education programs is increasing at a fast pace.
no code implementations • LREC 2016 • Mario S{\"a}nger, Ulf Leser, Steffen Kemmerer, Peter Adolphs, Roman Klinger
This corpus consists of 1, 760 annotated application reviews from the Google Play Store with 2, 487 aspects and 3, 959 subjective phrases.
no code implementations • WS 2014 • Benjamin Paassen, Andreas St{\"o}ckel, Raphael Dickfelder, Jan Philip G{\"o}pfert, Nicole Brazda, Tarek Kirchhoffer, Hans Werner M{\"u}ller, Roman Klinger, Matthias Hartung, Philipp Cimiano
no code implementations • LREC 2014 • Roman Klinger, Philipp Cimiano
Contributing to this situation, this paper describes the Bielefeld University Sentiment Analysis Corpus for German and English (USAGE), which we offer freely to the community and which contains the annotation of product reviews from Amazon with both aspects and subjective phrases.