Hobby has become a crucial component of human identity, providing them with both social and emotional satisfaction. Numerous persons express their viewpoints about their pastimes on various social media platforms. Twitter is often seen as a disruptive innovation in the field of sports communication because of its capacity to organize groups swiftly, therefore generating innovative channels for the sharing of viewpoints, information, and business transactions. Twitter, as a social media service, allows users to disclose personal information better so that it can be analyzed using Named Entity Recognition (NER) with the entity tag model found and categorized manually. This study aims to create a NER model for identifying human hobbies using Indonesian-language tweet data. The first steps in developing the NER model are inputting data sources, then creating the Indonesian NER Model, and the last step is model evaluation. This model uses the Python platform and Spacy library to create the NER model manually. The test result of the NER model, which was developed manually, has fifteen categories of label tags with an F1 score average of 92,31%. Furthermore, it is necessary to study models for other sentence patterns, considering that each part of the dataset has a different sentence pattern.