What is Vader Sentiment Analysis? Vader Sentiment Analysis is a Natural Language Processing technique used to determine the sentiment of various sentence strings. This analysis becomes useful in deciphering the emotions behind text within various media. This report follows basic rules and does not decipher more complex strings that include emoticons or other complex elements. This program was implemented using C and utilized memory allocation. Improvements in the future can be made by storing the words from the lexicon file in a hashmap.

Rules Implemented within program All Capital cases: In the case of a lexicon word having all capital characters, the program will multiply the sentiment score and boost it by a factor of 1.5. For non lexicon words, all capital cases are neutral. Punctuation Cases: If there is an exclamation mark directly after a lexicon word, the sentiment score is boosted by a factor of 1.292. If there are multiple exclamation marks, the logic follows 1+n(0.292) where n is the number of exclamation marks. This number however is capped at 3 so additional exclamation marks have no impact on the sentiment score of the word. Exclamation marks following non-lexicon words are neutral and have no effect on the total score. Negation Cases: If a negation word such as “not” or “isn’t” is detected in the sentence, the next lexicon word detected will be multiplied by a factor of-0.5 to negate the sentiment and slightly suppress it. Non-lexicon words following negations remain neutral. NOTE: negation words are counted as neutral within the program. Intensifiers: Intensifier words such as “very” and “uber” add a boost to the sentiment value. This boost is an addition of 0.293 to the sentiment score. If the intensifier is an all capital case, the boost is boosted by 1.5. Intensifiers have no effect on non-lexicon words. NOTE: intensifier words are counted as non-neutral in the program