There are few challenges I am facing when building a resume recommendation for a particular job positing.
Let's say we convert the resume into a vector on n-dimensions and job description also as an n-dimension vector and in order to see how similar, we can use any similarity metrics like cosine.
Now, for me the biggest problem with such approach is not able to provide more importance to the job title required. Some times, for a cloud engineer position I am getting java developer resume recommended in top 10 just because some of the skills/keywords are overlapping between two so their embeddings becomes similar.
I want to provide more weightage to the job title as well. What are some possible things I can do to make my recommendations consider or put bit emphasis on job title.
Note:- A common job title lookup in resume will fails because people write job titles in multiple ways. (java engineer (or) java developer) | (cloud engineer or aws engineer) etc.,
How can I overcome this issue?