Research on the Application of Natural Language Processing Technology in E-commerce

In recent years, more and more people are keen on online shopping and publish their evaluations on the Internet. In the shopping platform, analyzing these comments can help merchants grasp the advantages and disadvantages of the products in a timely manner, and make timely improvements, thereby improving the quality of the products. It also provides a reference for consumers to fully understand product information. Therefore, the method of using the existing natural language processing technology to process and analyze the emotional tendency of these texts has become one of the current research hotspots. In this paper, combining deep learning theory and natural language processing technology, based on the text classification method based on convolutional neural network, designs a segmented convolutional neural network text sentiment analysis model based on feature fusion. This model can extract the main features of sentences in segments. In the experimental results, it shows that compared with the traditional text CNN, this method has a significant improvement in indicators such as accuracy and recall in sentiment analysis tasks.