Python, NLP - 查找包含给定单词列表的顶部文档

Question

I am new to learning NLP. I am trying to do an exercise of finding the best matching resume.

For example, I have a list of skills that I am looking for like ['java', 'python', 'SQL', 'API', ...], and a set of documents. I want to create a model to find the document that is the best match with these skills. Similar to resume matching.

I started with this tutorial - Extracting words from pdf as a reference

I was able to extract the text from pdf, remove stop words, perform lemmatization, compute the number of times these keywords appear in each document and I am not sure how to go head from here.

Could anyone provide me with what the next steps should be? Any tutorials or references would be helpful as well.

Answer 1

if you assume that "the best matching", is the resume with the "largest intersection" with the set skills, then you would have (python):

import numpy as np
D=[["I","know","python"],["I","know","java"]] # list of Documents
skils=["java"] # list of skils
I=[len(list(set(skils) & set(d))) for d in D]
R = sorted(range(len(I)), key=lambda k: I[k]) # rank of intersection with the skillset
best_resume=R[0]
print (R)

I hope it can be useful. good luck.