[目的/意義]構建在線評論的產品特征提取及情感分類模型,可以為產品設計人員進行產品優化改進提供決策支持。[方法/過程]提出了基于卷積神經網絡算法的產品特征提取及情感分類模型。模型采用卷積神經網絡進行短文本評論情感分類,以情感分類標簽標注相應評論中提取的產品特征詞,并利用詞向量對產品特征詞聚類。通過爬取的筆記本電腦和手機評論對模型進行訓練和測試。[結果/結論]結果表明,模型能夠實現有效的產品特征提取及高準確率情感分類,是在線評論分析的有效模型。

[Purpose/significance] This paper constructs the product feature extraction and sentiment classification model for online reviews,which can provide decision support for product designers to improve and optimize products. [Method/process]The paper proposes the product feature extraction and sentiment classification model based on convolution neural network algorithm. The model uses the convolution neural network to classify the short online reviews,tags the result labels on the product feature words,which are extracted from the corresponding reviews,and adopts the word vector to cluster the product feature words. The laptop and cell phone reviews crawled online are used to train and test the model. [Result/conclusion]The results show that the model can effectively realize online reviews analysis with sound product feature extraction and high accuracy rate of sentiment classification.