As I helped my fourth-grade sister with her homework, I came across a question that asked her to complete the sentence "Mom is cooking in the kitchen, she...". This anecdote led me to ponder the well-established roles of women, particularly mothers and grandmothers, in the kitchen throughout history. To delve deeper into this topic, this research paper utilizes language content from the food.com website, an online community that shares recipes and cooking experiences, to substantiate the gender gaps in the kitchen. Specifically, the study investigates whether women cook more than men and the attitudes of food-makers and food-eaters. To analyze the collected data, two approaches are employed: the first attempt utilizes supervised machine learning for gender identification, but unfortunately does not yield satisfactory results. However, the second attempt utilizing the successful GPT response generation technique proves significantly more effective. Subsequently, sentiment analysis is conducted using traditional natural language processing methods. The findings of this analysis reveal that the conventional societal pattern of "women as food-makers and men as food-eaters" continues to persist in the modern era. Furthermore, both food-makers and food-consumers exhibit positive attitudes towards food; however, food-makers tend to emphasize qualities related to the culinary experience, as well as the sense of accomplishment and satisfaction derived from cooking, while food-consumers place greater value on the emotional aspects of feeling loved and cared for through food. Moreover, this study demonstrates the feasibility of incorporating GPT technology within the realm of big data social science research, despite encountering substantial difficulties and challenges along the way. As the field of AI continues to advance, with the generation of increasingly accurate responses and cleaner formatting, it is anticipated that such technology will prove to be a valuable asset in future research endeavors.