0
$\begingroup$

I am in the process of conducting a literature review for my thesis. Currently, I am struggling when it comes to developing a theoretical framework/methodology or to even correctly outline an approach to the given problem, as well as when it comes to finding relevant keywords and sources.

Problem description

When given a URL of a web page that contains a job posting, it can be assumed that there will be recurring key elements of information compared to any other such web page. The structure of the web page might differ in terms of page layout and source code, as well as in terms of the semantic structuring of the information present. There may be some information omitted that other job postings contain, and while a JobPosting schema for structured data on the Internet exists, it is not used by any (major) search engine for job postings.

I need to find a way to either automatically extract useful information from job posting web pages or automatically annotate these web pages for further processing using a natural language based approach. The information types that needs to be extracted or annotated includes:

  • Education Requirements
  • Eligibility to Work Requirement
  • Employer Overview
  • Employment Type
  • Experience Requirements
  • Job Benefits
  • Job Location
  • Qualifications
  • Responsibilities
  • Skills
  • Job Title

Possible approach

  1. Identify relevant content sections of the web page: Either only focusing on the text present or also with regards to the HTML structure of the web page
  2. Classifying the content based on keywords: Map content to their respective information type
  3. Annotate the information:
    1. In the form of a separate document such as RDFa, JSON-LD, etc. OR
    2. Modify the HTML code to inject the tags with the JobPosting schema and the previous classified information expressed in terms of the Microdata schema
  4. Now the information is "structured" and can be used in subsequent NLP tasks

Reviewed literature

I couldn't find anything that helps me validate/select a suitable approach for my problem. Possibly it is because I don't have the right approach or keywords in mind.

Questions

  1. Does anyone have suggestions for publications that I can consult to get a better understanding of my research problem?
  2. Could anyone comment on the possible approach I have listed to help me come up with a theoretical framework to solve the given problem?

Additionally, please let me know if you think I have forgotten a step in my approach. Any advice is greatly appreciated.

$\endgroup$
0

0

You must log in to answer this question.

Browse other questions tagged .