| abstract |
The present application provides a method for generating a search term, which uses a machine learning model to determine the retention of each word in a search term, and the machine learning model is trained by a sample of search terms having a target word feature. In the existing method for generating search terms, the target word feature can reflect the importance of the word to the query result, and the machine learning model trained using the target word feature can more accurately select the word selection result in the search term, thereby generating more Accurately retrieve items to avoid recalling irrelevant query results. In addition, the present application also provides related devices for generating a search term to ensure the implementation and application of the above method in practice. |