DUTH at SemEval-2020 Task 11: BERT with Entity Mapping for Propaganda Classification

2020-08-22 · SEMEVAL 2020 · Anastasios Bairaktaris, Symeon Symeonidis, Avi Arampatzis

This report describes the methods employed by the Democritus University of Thrace (DUTH) team for participating in SemEval-2020 Task 11: Detection of Propaganda Techniques in News Articles. Our team dealt with Subtask 2: Technique Classification. We used shallow Natural Language Processing (NLP) preprocessing techniques to reduce the noise in the dataset, feature selection methods, and common supervised machine learning algorithms. Our final model is based on using the BERT system with entity mapping. To improve our model's accuracy, we mapped certain words into five distinct categories by employing word-classes and entity recognition.

📄 PDF Abstract BibTeX arXiv:2008.09894

Code (1)

anasbair/SemEval2020-groups 공식 구현

Tasks

Articles BIG-bench Machine Learningfeature selectionGeneral Classification

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Feature Selection Feature selection, also known as variable selection, attribute selection or variable subset selection, is the process of selecting a subset of relevant features (variables,…

Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…

Weight Decay 설명 없음

Adam 설명 없음

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

WordPiece 설명 없음

Multi-Head Attention 설명 없음

DUTH at SemEval-2018 Task 2: Emoji Prediction in Tweets

2018-06-01 · SEMEVAL 2018 6 · Dimitrios Effrosynidis, Georgios Peikos, Symeon Symeonidis, Avi Arampatzis

This paper describes the approach that was developed for SemEval 2018 Task 2 (Multilingual Emoji Prediction) by the DUTH Team. First, we employed a combination of pre-processing techniques to reduce the noise of tweets a…

Information Retrieval Task 2

DUTH at SemEval-2017 Task 5: Sentiment Predictability in Financial Microblogging and News Articles

2017-08-01 · SEMEVAL 2017 8 · Symeon Symeonidis, John Kordonis, Dimitrios Effrosynidis, Avi Arampatzis

We present the system developed by the team DUTH for the participation in Semeval-2017 task 5 - Fine-Grained Sentiment Analysis on Financial Microblogs and News, in subtasks A and B. Our approach to determine the sentime…

Articles BIG-bench Machine Learning Feature Engineering Information Retrieval+2

DUTH at SemEval-2021 Task 7: Is Conventional Machine Learning for Humorous and Offensive Tasks enough in 2021?

2021-08-01 · SEMEVAL 2021 · Alexandros Karasakalidis, Dimitrios Effrosynidis, Avi Arampatzis

This paper describes the approach that was developed for SemEval 2021 Task 7 (Hahackathon: Incorporating Demographic Factors into Shared Humor Tasks) by the DUTH Team. We used and compared a variety of preprocessing tech…

Classification regression

DUTH at SemEval-2017 Task 4: A Voting Classification Approach for Twitter Sentiment Analysis

2017-08-01 · SEMEVAL 2017 8 · Symeon Symeonidis, Dimitrios Effrosynidis, John Kordonis, Avi Arampatzis

This report describes our participation to SemEval-2017 Task 4: Sentiment Analysis in Twitter, specifically in subtasks A, B, and C. The approach for text sentiment classification is based on a Majority Vote scheme and c…

BIG-bench Machine Learning General Classification Information Retrieval Sentiment Analysis+2

DUTH at SemEval-2019 Task 8: Part-Of-Speech Features for Question Classification

2019-06-01 · SEMEVAL 2019 6 · Anastasios Bairaktaris, Symeon Symeonidis, Avi Arampatzis

This report describes the methods employed by the Democritus University of Thrace (DUTH) team for participating in SemEval-2019 Task 8: Fact Checking in Community Question Answering Forums. Our team dealt only with Subta…

Community Question Answering Fact Checking feature selection General Classification+2