Research on Text Generation of Movie Information Retrieval Based on Natural Language Processing Technology

The rapid development of the film industry and the complexity of related information pose challenges to effective film information retrieval. NLP technology has been widely used and is very suitable for film-related information extraction, processing and retrieval. Based on natural language processing technology, this paper researches film information retrieval text generation. First, this paper implements a movie information knowledge graph containing 5044 ontologies and 22,351 triples; then, this paper uses a dictionary-based named entity recognition method to extract key information from the retrieval problem and uses a rule-based method to generate the corresponding information. In order to obtain more objective evaluation information of related movies, this paper also uses the LSTM network model to conduct sentiment analysis of movie review information and adds this information to the knowledge graph to obtain richer movie information. The research in this paper can promote the intelligent retrieval of movie information.