SSiW


Affective Norms: Abstractness, Arousal, Imageability, Valence


Affective Ratings for German

This resource contains a collection of 350,000 German lemmatised words, rated on the four affective attributes. All ratings were obtained via a supervised learning algorithm that automatically calculates a numerical rating for each word and each affective attribute.

See here on how to obtain the data.

Abstractness Ratings for Estonian

This resource contains a collection of 243,675 Estonian lemmatised words, rated on abstractness. These ratings were obtained via the same supervised learning algorithm as the German ratings, but based on translation seeds from English.

See here on how to obtain the data.

Abstractness Ratings for English

We combined a feed-forward neural network, existing ratings for English and word2vec cbow word vectors trained on a Google-internal news corpus with 100 billion tokens to propagate abstractness ratings to a vocabulary of 3 million words.

See here on how to obtain the data.


References:

Eleri Aedmaa, Maximilian Köper, Sabine Schulte im Walde (2018)
Combining Abstractness and Language-specific Theoretical Indicators for Detecting Non-Literal Usage of Estonian Particle Verbs
In: Proceedings of the NAACL 2018 Student Research Workshop (NAACL-SRW). New Orleans, LA.

Maximilian Köper, Sabine Schulte im Walde (2016)
Automatically Generated Norms of Abstractness, Arousal, Imageability and Valence for 350 000 German Lemmas
In: Proceedings of the 10th Conference on Language Resources and Evaluation (LREC). Portoroz, Slovenia.

Maximilian Köper, Sabine Schulte im Walde (2017)
Improving Verb Metaphor Detection by Propagating Abstractness to Words, Phrases and Individual Senses
In: Proceedings of the 1st Workshop on Sense, Concept and Entity Representations and their Applications (SENSE). Valencia, Spain.