nlp-project-acurracy-analysis-of-bow-and-tf-idf.
𝐏𝐫𝐨𝐣𝐞𝐜𝐭 𝐇𝐢𝐠𝐡𝐥𝐢𝐠𝐡𝐭: 𝐀 𝐒𝐮𝐫𝐩𝐫𝐢𝐬𝐢𝐧𝐠 𝐓𝐰𝐢𝐬𝐭 𝐢𝐧 𝐓𝐞𝐱𝐭 𝐂𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧 𝐀𝐜𝐜𝐮𝐫𝐚𝐜𝐲! Contrary to the common assumption that TF-IDF generally outperforms Bag of Words due to its sophisticated weighting mechanism, my findings showed that the Bag of Words model actually delivered higher accuracy in this specific case! BOW- accuracy: 0.9856502242152466 , whereas TF-IDF has accuracy of accuracy: 0.979372197309417