Problem Statement

In the last session we talked about the existing biases in the natural language processing $($NLP$)$ libraries. We also identified some of these biases through the Word Analogy python file with the focus on gender bias.

Required Analysis

Question $1$: Discuss other possible sources of bias that might exist. Question $2$: Use the Word Analogy python code $($which I shared in Week $1)$ to generate some word analogies that show some sort of bias in the NLP libraries $($Don$\text{'}$t limit yourself to gender bias and try to be creative$).$