This paper presents an investigation of three different natural language processing (NLP) models for intent classification on the dataset from the Airline Travel Information System (ATIS). The ATIS dataset is a widely used benchmark dataset in the field of NLP for its practical relevance in the airline reservation domain. The three models used in this study are Support Vector Machines (SVM), Bidirectional Encoder Representations from Transformers (BERT) and Long Short-Term Memory (LSTM). Our results show that BERT outperformed both SVM and LSTM, achieving an accuracy of 98.5%. This indicates that contextualized word embeddings can effectively capture the complex nuances and variations in the language that is present in the ATIS dataset. The results also highlight the importance of selecting appropriate NLP models for specific tasks and datasets. Our findings have practical implications for the airline industry, as accurate intent classification can lead to improved customer service and operational efficiency. Overall, this study demonstrates The ATIS dataset was used to test the efficacy of NLP models for intent classification, and the analysis offers information on the advantages and disadvantages of various models.