Discovering Job Knowledge from Web Data
The project DISKOW aims at creating a Job Knowledge Base (JKB) prototype, based on existing open source Business Intelligence platform in order to cover most important factors in this regard such as required job knowledge for specific job.
Traditionally, to get more insight into labor demand or supply, researchers and policymakers have relied on interviews, trade publications, surveys, and vacancies. While these traditional data sources have some clear advantages, they are also characterized by limitations that can be addressed by using web-based data instead. Web is a gold mine for job knowledge discovery. Linked open data, job announcements, social media, job search engines, forums, wikis, data streams, and interlinked information are few examples of such valuable job-related sources on the net.
In this regard, the main problem is not the availability of data and how to retrieve them, but how to clean, explore, visualize and interpret such huge volume of various web data. With an aim to streamline this process and make such data suitable for further exploitation (e.g. by specialised mobile apps) an open Job Knowledge Base (JKB) is proposed within the scope of the DISKOW project that can be used by employers, employees, job seekers, labor market experts and policy makers. A JKB contains different types of information such as responsibilities and roles, required competences (described using existing standards, such as eCF, EQF, etc.) that could be used to develop training and identify priorities, wage information, geographical and demographic trends, cultural issues, demands of the job markets in different domains, job announcement information and rates, job popularity and other useful statistics. The project DISKOW aims at creating a JKB prototype, based on existing open source Business Intelligence platform in order to cover most important factors in this regard such as required job knowledge for specific job.