The rapid development of Internet public opinion has caused a series of information management problems. The contents of "Influencer chaos" and "being a talking point" have impacted the senses of public opinion, and even affected social security problems. Based on the Internet of Things (IoT) and big data, this work applies the Natural Language Processing (NLP) to the analysis of network public opinion. Additionally, it takes the content of microblog text format as the main collection target, constructs a big data collection tool, and establishes Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Deep Pyramid Convolutional Neural Network (DPCNN) based on Tensorflow and other deep learning models. It is also improved in combination with the characteristics of the model, and a new model is proposed. Finally, the performance of various models is compared and analyzed through experiments, and the path is proposed for the government to use big data to improve the ability of governing network public opinion and help social governance. The results show that the accuracy of the modified model is significantly improved compared with other models. In the most ideal case, the accuracy can be improved by nearly two percentage points, indicating that adding residual connection and Attention mechanism can make the model better extract the emotional features in the text and improve the emotional discrimination ability of the model. If the public opinion of online media under the big data and IoT is not effectively controlled, it will have great security risks to social governance. The proposed method is of great help to the analysis of online public opinion through the accurate analysis of microblog text.