Calculate Candidate Compatibility Percentage for a Job Position (NLP) part 2 | by Patrizia Castagno | Jan, 2024

This half is the continuation of: Calculate Candidate Compatibility Percentage for a Job Position (NLP)

It’s important to recollect the truth that machines lack an inherent understanding of characters, phrases, or sentences. Their performance is restricted to processing numerical information solely. Consequently, textual knowledge ought to bear encoding into numerical codecs for enter or output in any machine.

On this case, utilizing Perform Encoding is essential as a result of it entails transforming important textual content material proper right into a numerical/vector illustration, preserving the context and relationships between phrases and sentences. This permits a machine to know patterns in any textual content material and discern the context of sentences. Inside the realm of Perform Encoding, various methods exist. For this express case, we’ve acquired opted to utilize the Bag of Phrases (BoW) methodology.

What’s Bag of Phrases (BoW)?

Bag of Phrases, usually abbreviated as BoW, serves as a simplified illustration also used in pure language processing (NLP) and knowledge retrieval. Inside the BoW model, a doc is portrayed as an unordered set of phrases, disregarding grammar and phrase order, whereas meticulously monitoring the frequency of each phrase’s incidence.

Primarily, BoW solely focuses on the presence or absence of phrases, with out concern for his or her sequential affiliation.

How does BoW perform?

It constructs a vector based totally on the presence (1) or absence (0) of a phrase. Consequently, the following encodings are extraordinarily sparse and multidimensional. BoW’s counting mechanism tallies the occurrences of each phrase throughout the doc. As an example, suppose you may need the following two fast sentences:

1. “I like programming.”
2. “Programming is satisfying.”

Now, let’s create a main Bag of Phrases illustration:

1. Create a vocabulary by itemizing all distinctive phrases from the sentences: [“I”, “love”, “programming”, “is”, “fun”].
2. Symbolize each sentence as a vector, indicating the presence or absence of each phrase:

• Sentence 1: [1, 1, 1, 0, 0] (I: 1, love: 1, programming: 1, is: 0, satisfying: 0)
• Sentence 2: [0, 0, 1, 1, 1] (I: 0, love: 0, programming: 1, is: 1, satisfying: 1)