C11

Question 1

problems with rule-based approach for extracting information (eg. name from a CV)

Answer 1

• gets complex to accommodate for exceptions
• coverage is limited

Question 2

typical machine learning pipeline for parsing CVs

Answer 2

• preprocessing/OCR
• detection of CV pages
• section segmentation
• item segmentation
• phrase segmentation

Question 3

how does Textkernel match CVs to vacancies?

Answer 3

transform CVs and vacancies into vectors, where relevant CVs are close to a job

calculate the distance between CV and job vectors to make a classification

Question 4

how does Textkernel match CVs to vacancies?

Answer 4

transform CVs and vacancies into vectors, where relevant CVs are close to a job

calculate the distance between CV and job vectors to make a classification