Gregor Wiedemann

Bio: Gregor Wiedemann is an academic researcher from University of Hamburg. The author has contributed to research in topics: Topic model & Information extraction. The author has an hindex of 13, co-authored 53 publications receiving 690 citations. Previous affiliations of Gregor Wiedemann include Leipzig University.

Applying LDA Topic Modeling in Communication Research: Toward a Valid and Reliable Methodology

Abstract: Latent Dirichlet allocation (LDA) topic models are increasingly being used in communication research. Yet, questions regarding reliability and validity of the approach have received little attentio...

Does BERT Make Any Sense? Interpretable Word Sense Disambiguation with Contextualized Embeddings

Abstract: Contextualized word embeddings (CWE) such as provided by ELMo (Peters et al., 2018), Flair NLP (Akbik et al., 2018), or BERT (Devlin et al., 2019) are a major recent innovation in NLP. CWEs provide semantic vector representations of words depending on their respective context. Their advantage over static word embeddings has been shown for a number of tasks, such as text classification, sequence tagging, or machine translation. Since vectors of the same word type can vary depending on the respective context, they implicitly provide a model for word sense disambiguation (WSD). We introduce a simple but effective approach to WSD using a nearest neighbor classification on CWEs. We compare the performance of different CWE models for the task and can report improvements above the current state of the art for two standard WSD benchmark datasets. We further show that the pre-trained BERT model is able to place polysemic words into distinct 'sense' regions of the embedding space, while ELMo and Flair NLP do not seem to possess this ability.

Opening up to Big Data: Computer-Assisted Analysis of Textual Data in Social Sciences

Abstract: Two developments in computational text analysis may change the way qualitative data analysis in social sciences is performed: 1. the availability of digital text worth to investigate is growing rapidly, and 2. the improvement of algorithmic information extraction approaches, also called text mining, allows for further bridging the gap between qualitative and quantitative text analysis. The key factor hereby is the inclusion of context into computational linguistic models which extends conventional computational content analysis towards the extraction of meaning. To clarify methodological differences of various computer-assisted text analysis approaches the article suggests a typology from the perspective of a qualitative researcher. This typology shows compatibilities between manual qualitative data analysis methods and computational, rather quantitative approaches for large scale mixed method text analysis designs. URN: http://nbn-resolving.de/urn:nbn:de:0114-fqs1302231

Does BERT Make Any Sense? Interpretable Word Sense Disambiguation with Contextualized Embeddings

Abstract: Contextualized word embeddings (CWE) such as provided by ELMo (Peters et al., 2018), Flair NLP (Akbik et al., 2018), or BERT (Devlin et al., 2019) are a major recent innovation in NLP. CWEs provide semantic vector representations of words depending on their respective context. Their advantage over static word embeddings has been shown for a number of tasks, such as text classification, sequence tagging, or machine translation. Since vectors of the same word type can vary depending on the respective context, they implicitly provide a model for word sense disambiguation (WSD). We introduce a simple but effective approach to WSD using a nearest neighbor classification on CWEs. We compare the performance of different CWE models for the task and can report improvements above the current state of the art for two standard WSD benchmark datasets. We further show that the pre-trained BERT model is able to place polysemic words into distinct 'sense' regions of the embedding space, while ELMo and Flair NLP do not seem to possess this ability.

A Primer in BERTology: What We Know About How BERT Works

Abstract: Transformer-based models have pushed state of the art in many areas of NLP, but our understanding of what is behind their success is still limited. This paper is the first survey of over 150 studies of the popular BERT model. We review the current state of knowledge about how BERT works, what kind of information it learns and how it is represented, common modifications to its training objectives and architecture, the overparameterization issue and approaches to compression. We then outline directions for future research.

A Primer in BERTology: What we know about how BERT works

Abstract: Transformer-based models have pushed state of the art in many areas of NLP, but our understanding of what is behind their success is still limited. This paper is the first survey of over 150 studies of the popular BERT model. We review the current state of knowledge about how BERT works, what kind of information it learns and how it is represented, common modifications to its training objectives and architecture, the overparameterization issue and approaches to compression. We then outline directions for future research.

SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval).

Abstract: We present the results and the main findings of SemEval-2019 Task 6 on Identifying and Categorizing Offensive Language in Social Media (OffensEval). The task was based on a new dataset, the Offensive Language Identification Dataset (OLID), which contains over 14,000 English tweets, and it featured three sub-tasks. In sub-task A, systems were asked to discriminate between offensive and non-offensive posts. In sub-task B, systems had to identify the type of offensive content in the post. Finally, in sub-task C, systems had to detect the target of the offensive posts. OffensEval attracted a large number of participants and it was one of the most popular tasks in SemEval-2019. In total, nearly 800 teams signed up to participate in the task and 115 of them submitted results, which are presented and analyzed in this report.

SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (OffensEval 2020)

Abstract: We present the results and the main findings of SemEval-2020 Task 12 on Multilingual Offensive Language Identification in Social Media (OffensEval-2020). The task included three subtasks corresponding to the hierarchical taxonomy of the OLID schema from OffensEval-2019, and it was offered in five languages: Arabic, Danish, English, Greek, and Turkish. OffensEval-2020 was one of the most popular tasks at SemEval-2020, attracting a large number of participants across all subtasks and languages: a total of 528 teams signed up to participate in the task, 145 teams submitted official runs on the test data, and 70 teams submitted system description papers.

The ERP response to the amount of information conveyed by words in sentences (vol 140, pg 1, 2015)

Abstract: Reading times on words in a sentence depend on the amount of information the words convey, which can be estimated by probabilistic language models. We investigate whether event-related potentials (ERPs), too, are predicted by information measures. Three types of language models estimated four different information measures on each word of a sample of English sentences. Six different ERP deflections were extracted from the EEG signal of participants reading the same sentences. A comparison between the information measures and ERPs revealed a reliable correlation between N400 amplitude and word surprisal. Language models that make no use of syntactic structure fitted the data better than did a phrase-structure grammar, which did not account for unique variance in N400 amplitude. These findings suggest that different information measures quantify cognitively different processes and that readers do not make use of a sentence's hierarchical structure for generating expectations about the upcoming word.