Key points are not available for this paper at this time.
本文提出了一种简单的无监督学习算法,用于将评论分类为推荐(点赞)或不推荐(点踩)。评论的分类由包含形容词或副词的短语的平均语义倾向预测。当短语具有良好关联(如“细微差别”)时,语义倾向为正;当短语具有不良关联(如“非常随意”)时,语义倾向为负。本文中,短语的语义倾向计算为该短语与“excellent”的互信息减去该短语与“poor”的互信息。若短语的平均语义倾向为正,则评论被分类为推荐。该算法在410条来自Epinions的评论上评估,涵盖汽车、银行、电影和旅行目的地四个不同领域,平均准确率达74%。准确率从汽车评论的84%到电影评论的66%不等。
Building similarity graph...
Analyzing shared references across papers
Loading...
Peter D. Turney(周三)研究了这个问题。
www.synapsesocial.com/papers/6a07a00db2d9a7d54307ad50 — DOI: https://doi.org/10.48550/arxiv.cs/0212032
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:
Peter D. Turney
National Research Council Canada
Building similarity graph...
Analyzing shared references across papers
Loading...