You shall know a word by the company it keeps
From Algolit
Revision as of 11:00, 24 October 2017 by Manetta (talk | contribs) (Created page with "__NOTOC__ ''5 landscapes'' {| |- | Type: || Algolit extension |- | Datasets: || Frankenstein, AnarchFem, Wikipedia harassments in comments, Pirated Deep Learn...")
Revision as of 11:00, 24 October 2017 by Manetta (talk | contribs) (Created page with "__NOTOC__ ''5 landscapes'' {| |- | Type: || Algolit extension |- | Datasets: || Frankenstein, AnarchFem, Wikipedia harassments in comments, Pirated Deep Learn...")
5 landscapes
| Type: | Algolit extension |
| Datasets: | Frankenstein, AnarchFem, Wikipedia harassments in comments, Pirated Deep Learning textbooks, Tristes Tropiques |
| Technique: | calculating semantic similarity with word-embeddings |
| Developed by: |
- Semantic similarities
- distance in the graph
- is the product of a code study to an existing word-embedding tutorial script of word2vec.
- Based on 5 different datasets, as an exercise to read their content in semantic clusters.
5 landscapes, 5 datasets. Each of the graphs contain the words 'human', 'view' and 'power', which are highlighted in red.
graph 1: Frankenstein dataset
Includes the book Frankenstein.
Error creating thumbnail: Unable to save thumbnail to destination
graph 2: Anarch Feminist dataset
Includes 3 books (...)
Error creating thumbnail: Unable to save thumbnail to destination
graph 3: Claude Levi-Strauss dataset
Includes the book Tristes Tropiques.
Error creating thumbnail: Unable to save thumbnail to destination
graph 4: Deep Learning textbooks dataset
Includes the books (...).
Error creating thumbnail: Unable to save thumbnail to destination
graph 5: Harassing comments dataset
Includes examples of harassment on Talk page comments from Wikipedia.
Error creating thumbnail: Unable to save thumbnail to destination