Before using the service, please read the preliminary information containing a description of steps that enable access to the CLARIN-PL developer interface.
Websty is a tool for analyzing and visualizing the similarity of documents.
The service can be run:
type - a method for determining vectors describing documents:
authorship - authorship analysis100base - 100 the most frequent lemmas100orths - classic authorship attribution (100 most frequent words from the text)1000base - classic content similarity (1000 most frequent lemmas)multilingual-e5-base - default valuelang - language of the text, crucial for TF-IDF based methods
pl - default valuechunk_size - the size (in bytes) of the fragments that the input files are divided into
20000 - default valuemetric - a method for determining distance in similarity analysis for clustering and visualization (UMAP):
cosine - default valueeuclideanmanhattanchebyshevminkowskicanberrabraycurtishaversinemahalanobiswminkowskiseuclideancorrelationn_neighbors - The number of neighbors in the UMAP method
15 - default valuemin_dist - the minimum distance in the UMAP method
0.1The service can be run in the Windows system with default values using the following LPMN query: ['websty']
[['websty']] - input data in the form of a compressed directory (.zip)
Corpus
JSON, JSONL and HTML files.
(C) CLARIN-PL