Automatic tagging for online job listings

摘要

Methods and systems receive machine-readable input data describing a job listing or employment opportunity. The system scans the input data and software algorithms process it to predict specific information required by a particular html schema. To predict the information required by the schema, in one aspect, the system applies machine learning tools, for example, prediction models, which may be generated from training data selected from large job listing data collections, and specifically adapted and tuned for each field of interest, for example, job description, job location, and salary. This information is used to generate structured data, for example, html code, consistent with the schema for automatic insertion into a web page to comply with the schema and make the listing widely available to automated search processes.