skillNer.utils.Utils¶

class skillNer.utils.Utils(nlp, skills_db)¶

Methods

`__init__`(nlp, skills_db)
`compute_w_ratio`(skill_id, matched_tokens)
`get_clusters`(co_oc)
`get_corpus`(text, matches)	create a corpus matrix which will be used in future computations.
`grouper`(iterable, dist)
`make_one`(cluster, len_)
`one_gram_sim`(text_str, skill_str)
`process_n_gram`(matches, text_obj)	apply on conflicted matches to choose which ones to keep
`retain`(text_obj, span, skill_id, sk_look, corpus)	add doc here
`split_at_values`(lst, val)