WISDOM-2017-Creating Domain-Specific Sentiment Lexicons via Text Mining

[BrambleXu]

Summary:

通过文本挖掘的技巧来生成domain-specific lexicon

Resource:

• pdf
• [code](
• [paper-with-code](

Paper information:

• Author:
• Dataset:
• keywords:

Notes:

The coverage is the proportion of vocabulary covered by the lexicon 这里的coverage定义就比较简单了。直接拿test set里所有的entity在lexicon里进行检索，找到的概率越高，说明coverage越高。（找到的entity / test set里的所有entity）。这不就是recall嘛

Model Graph:

Result:：

Thoughts:

Next Reading: