One to Many Comparison - LSA at CU Boulder

This interface allows you to compare the similarity of multiple texts within a particular LSA space, with word2vec word embeddings, or using the BERT language model. One designated text is compared to all other texts.

To compute the similarity of a particular text to many other texts, enter the one text as the first line in the text box and each of the others below it. Use a blank line to separate each text. Then press the 'Submit Texts' button. The system will compute a cosine similarity score for each submitted text compared to all submitted texts.

Select the word embedding method:

Select an embedding space:

Select the comparison type:

Number of factors to use (leave blank for maximum factors available):

Show vector lengths

Enter texts here (separated by a blank line):