cjhutto/vaderSentiment

Incorrect results for sentences containing negations

Open

#71 建立於 2019年1月13日

在 GitHub 查看
 (1 留言) (0 反應) (0 負責人)Python (4,146 star) (975 fork)batch import
help wanted

描述

Hi

First of all, great work!

I've noticed some negations are analysed incorrectly. For instance,

Not the worst show I've seen-------- {'neg': 0.0, 'neu': 0.603, 'pos': 0.397, 'compound': 0.5096} Not the worst ever but definitely worst this year------ {'neg': 0.0, 'neu': 0.372, 'pos': 0.628, 'compound': 0.879}

The compound scores are incorrectly positive as these two sentences are neural at best.

貢獻者指南

技術棧
python
領域
machine learning
議題類型
bug
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
half day
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
stale
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
Python programmingSentiment analysis basics
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
65
研究方向
Investigate the vaderSentiment library's handling of negation in the lexicon and score computation. Look at the source code (e.g., vaderSentiment.py) to understand how negations like 'not' are processed. Modify the rule for negations to correctly flip the sentiment polarity for phrases like 'not the worst'. Test with given examples and ensure compound score is negative.