Global AI and Data Science

These are the steps of indexing in Lucene given in our syllabus-:

The first step says that it is creating an index whereas the last step says that it's adding document to index.

What's the difference between these two? Can I get an example.

Here's what I think it should happen-:

1) Collect all words from each documents. Lists it like-;

doc1=>word1,word2,WORD3….wordn

doc2=>word1,WORD2,word3….wordn

And so on.

2) Analyse the words and remove various types of words as per analyzer, process them as per analyzer.

Say now what remains is-:

doc1=>word1,word3,...word(n-1)

doc2=>word2,...word(n-3)

3) Done. Now you can make inverted index as well by converting this to inverted index.