E-commerce commodity classification method based on deep learning model

The invention provides an e-commerce commodity classification method based on a deep learning model, belongs to the technical field of deep learning models and live broadcast e-commerce, and solves the problems of coordination, efficiency and cost of a single text or picture prediction scheme. Comprising the following steps: S1, acquiring original sample data, splitting text training data, and training a text classification model; s2, splitting invalid training data with unclear commodity title description, obtaining corresponding picture training data, and training a picture classification model; s3, obtaining commodity detail data of the live broadcast e-commerce in real time, and judging whether the text data is valid or not; s4, when the text data is valid, predicting commodity classification by using the text classification model; s5, when the text data is invalid, splitting the corresponding picture data, and predicting commodity classification by using a picture classification model; according to the i