What is Latent Semantic Analysis (LSA)?
LSA and its applications.
Published in
3 min readFeb 10, 2020
Latent Semantic Analysis, or LSA, is one of the basic foundation techniques in topic modeling. It is also used in text summarization, text classification and dimension reduction. It is similar to the cosine similarity. For LSA, we generate a matrix by using the words present in the paragraphs of the document in the corpus. The rows of the matrix will represent the unique words present in each paragraph, and columns represent each paragraph.
The basic assumption for the LSA algorithm is that words that are closer in their meaning will occur in a similar excerpt of the text. Let us consider an example to understand this in detail.
#Define Documents
Document_A: Alpine snow winter boots.
Document_B: Snow winter jacket.
Document_C: Alpine winter gloves. Documents
==================================================================
Words Document_A Document_B Document_C
==================================================================
alpine 1 0 1
snow 1 1 0
winter 1 1 1
boots 1 0 0
jacket 0 1 0
gloves 0…
