Word2Vec+LSTM多类别情感分类算法优化

Optimization of Word2Vec and LSTM Multi-Category Sentiment Classification Algorithm

随着网民的数量不断增加,用户上网产生的数据量也在成倍增多,随处可见各种各样的评论数据,所以构建一种高效的情感分类模型就非常有必要.本文结合Word2Vec与LSTM神经网络构建了一种三分类的情感分类模型:首先用Word2Vec词向量模型训练出情感词典,然后利用情感词典为当前训练集数据构建出词向量,之后用影响LSTM神经网络模型精度的主要参数来进行训练.实验发现:当数据不进行归一化,使用He初始化权重,学习率为0.001,损失函数选择均方误差,使用RMSProp优化器,同时用tanh函数作为激活函数时,测试集的总体准确率达到了92.28%.与传统的Word2Vec+SVM方法相比,准确率提高了大约10%,情感分类的效果有了明显的提升,为LSTM模型的情感分类问题提供了新的思路.

With the increasing number of netizens, the users on the Internet has doubled, and a variety of comment data can be seen everywhere. So, it is very necessary to construct an efficient emotional classification model. This study combined Word2 Vec with LSTM neural network to construct a three-class emotional classification model. Firstly,Word2 Vec word vector model is used to train the emotion dictionary. Then, we construct word vectors for the current training set data by using emotional dictionary. Then, this study used the main parameters that affecting the accuracy of LSTM neural network model to train the model. The experiment found that when the data are not normalized, using the weight of He is initialized, the learning rate is 0.001, the loss function is mean square error, the RMSProp optimizer is used, the training rounds are 30, and the accuracy of traditional Word2 Vec + SVM method improves by about 10%. The effect of affective classification promotes obviously, which provides a new way of thinking for LSTM model’s sentiment classification.