#  Can neural network model tell you if the title is a letter title 

 



 Here we share [our model](https://huggingface.co/cbdb/ClassicalChineseLetterClassification?text=%E4%B8%8A%E4%B8%9E%E7%9B%B8%E5%BA%B7%E6%80%9D%E5%85%AC%E6%9B%B8) to predict whether the title in the table of contents in Ming literary collections is a letter title or not. For example, the model can predict with almost perfect accuracy that 上丞相康思公書 is a letter title but 劉先生墓誌銘 is not.

 In collaboration with [the Center for Chinese Studies 漢學研究中心](/csa-ming-letters-project) and our research assistant [Katherine Enright](https://github.com/KatherineMEnright), the China Biographical Database project created 438,000 records of training data for letter titles. We fine-tuned a transformer model with these training data. Queenie Luo queenieluo\[at\]g.harvard.edu wrote the code. For further information, please see [THIS LINK](https://huggingface.co/cbdb/ClassicalChineseLetterClassification?text=%E4%B8%8A%E4%B8%9E%E7%9B%B8%E5%BA%B7%E6%80%9D%E5%85%AC%E6%9B%B8).