字符相似度, 汉字字形/拼音/语义相似度(单字, 可用于数据增强, CSC错别字检测识别任务(构建混淆集)) Chinese character font/pinyin/semantic similarity (single character, can be used for data augmentation, CSC misclassified character detection and recognition tasks (building confusion sets))