vacancy_skills_data

3 datasets representing processed skill-sets for job advertisements obtained from HeadHunter online hiring platform (collected with open API https://dev.hh.ru/) for specialists in Information Technologies (in accordance with classifier https://github.com/hhru/api/blob/master/docs_eng/specializations.md). Description of main fields for vacancies is available via link https://github.com/hhru/api/blob/master/docs_eng/vacancies.md.<br><b>Datasets</b>:<br>1. "<i>vacancy_skill.csv</i>" - two column dataset, representing vacancy ID ("vacid") and processed skill name [in-demand skill] ("lv");<br>2. "<i>sh_soft_clusters.csv</i>" - three column dataset, representing initial formulations (translated) of "soft" skills ("V1"), the frequency of occurrence in the sample ("V2"), the etalon name [generalized categories of "soft" skills] ("ETALON");<br>3. "<i>jaccard_matrix.csv</i>" - dissimilarity square matrix between processed skill names (1,730 X 1,730) with Jaccard distances (computed by comparison of vacancy ID sets for each skill pair) [the first row and the first column contain skill names].