机撰文稿的智能识别系统设计与实现

Design and Implement of Intelligent Recognition System for Machine Written Manuscripts

随着自媒体时代的到来,互联网新闻、网络互动社区言论成为民众舆论的主力军。该系统针对网络虚假舆论被恶意引导、传播等问题,设计一款机撰文稿的智能识别系统。首先使用Python设计网络文稿数据采集功能,然后利用TensorFlow深度学习框架训练出能分辨正负面情感的文本识别模型,最后实现业务逻辑模块和数据大屏展示,对用户请求的文稿进行判断甄别,该系统为现阶段的舆情分析提供一种新思路和新手段。

With the coming of we media era, Internet news and online interactive community speech have become the main force of public opinion. In this paper, aiming at the problem that the network false public opinion is maliciously guided and spread, an intelligent recognition system of machine written manuscripts is designed. Firstly, Python is used to design the data collection function of network manuscripts, and then tensorflow deep learning framework is used to train a text recognition model that can distinguish positive and negative emotions. Finally, the business logic module and data large screen display are implemented to judge and screen the manuscripts requested by users. This system provides a new idea and new means for the current public opinion analysis.