Papers by Stian Rødven Eide

2 papers
SenSALDO: Creating a Sentiment Lexicon for Swedish (L18-1)

Copied to clipboard

Challenge: sentiment analysis has seen an explosive expansion over the last decade or so . many theoretical and methodological questions remain unanswered and resource gaps unfilled .
Approach: They develop a sentiment lexicon for written (standard) Swedish using an existing dataset . they assign a real value sentiment score in the range [-1,1] and produce a label for it .
Outcome: The proposed sentiment lexicon is an open source resource from the Swedish Language Bank . it is based on an existing gold standard dataset and is available from Sprkbanken .
Generating a Gold Standard for a Swedish Sentiment Lexicon (L18-1)

Copied to clipboard

Challenge: Existing sentiment lexicons are compiled by (machine) translation from English resources, obscuring language-specific characteristics of sentiment-loaded vocabulary.
Approach: They propose a gold standard for sentiment annotation of Swedish terms using the SALDO lexicon and the Gigaword corpus.
Outcome: The proposed model is based on the free SALDO lexicon and the Gigaword corpus and is compared with existing models using human annotations.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations