Research On Hotel Intelligent Recommendation System Based On LSTM Model Analysis

In recent years,the research direction of text analysis is very hot.As English is simpler than Chinese,in the text analysis,foreign English text analysis is more mature than Chinese text analysis.In the analysis of Chinese text,there are still many problems to be studied.Chinese and English are different from English.It needs participle processing,and the Chinese is extensive and profound.The words have different meanings in different sentences.The sentences in different environments have different meanings,and the continuous updating of the Internet words,so the Chinese text is studied.There's a lot of space for research.Nowadays,deep learning is the most popular and most advanced research method.The concept of deep learning originated from artificial neural network,which originated from image recognition.But in a few years,deep learning has been extended to various fields of machine learning,and Natural Language Processing has also developed very much.Neural network learning has achieved good results in many research directions,such as the AlphaGo victory over human career go players.This is the first artificial intelligence program that defeating the world champion of weiqi.Its main principle is deep learning;in image classification and processing,the Convolutional Neural Network(CNN).Also obtained very good research results;and now the Internet chat robot is also the use of deep learning algorithm in Natural Language Processing research results.In the field of Hotel Review and analysis,the current research results show that there is still a lot of improved space for the results of the review of hotel reviews by machine learning.LSTM(Long Short-Term Memory)is a long and short term memory network,a kind of time recurrent neural network,for RNN(Recurrent neural Network).Recurrent neural network)an improved method of neural network,which has a good effect on the classification of long text.The hotel review is a combination of short text and long text.It will have good research effect on the research field ofthe LSTM model in the classification of hotel reviews.LSTM-RNN(long and short time memory network)is also a good method in the current language model.It is an important improvement model of recurrent neural network.It can better combine contextual links for emotional analysis.In this paper,the most popular LSTM model in Natural Language Processing is used to classify the text of comments.LSTM selects ADAM optimization algorithm,logarithmic loss function,regulates dropout to prevent over fitting.Finally,it combines the CNN network layer to optimize the LSTM model,and gets a good classification effect.