extract_sentiment_terms() currently labels the word "black" as 'negative' and labels "white" as 'positive' sentiment. I realize that in English and most languages, we have used "black" in the negative context (e.g., black sheep) and "white" in the positive context (e.g., white knight); however, this is due to historical and present-day racism.
All other colors and races/ethnicities are coded as neutral. "Red" is also used to describe negative sentiment (e.g., red tape); however, "red" is still classified as 'neutral.' Changing "black" and "white" to be neutral would be correcting infrastructural racism that affects one too many analyses. Additionally, given today's context in how "black" and "white" are used in the written language, I believe coding them to neutral would reduce the sentiment error rate.
extract_sentiment_terms() currently labels the word "black" as 'negative' and labels "white" as 'positive' sentiment. I realize that in English and most languages, we have used "black" in the negative context (e.g., black sheep) and "white" in the positive context (e.g., white knight); however, this is due to historical and present-day racism.
All other colors and races/ethnicities are coded as neutral. "Red" is also used to describe negative sentiment (e.g., red tape); however, "red" is still classified as 'neutral.' Changing "black" and "white" to be neutral would be correcting infrastructural racism that affects one too many analyses. Additionally, given today's context in how "black" and "white" are used in the written language, I believe coding them to neutral would reduce the sentiment error rate.
Would you please correct this?
Thank You, Chandni