Problems with improper segmentation of mixed Chinese and English words

[nyzhyxydsh]

In the test, It is found that the word segmentation effect is poor in scenes containing English words，the test case is “openjdk,springboot框架”，The correct expected word segmentation result is “openjdk , springboot 框架”，but the actual word segmentation result is “o p e n j d k , s p r i n g b o o t 框架”，each character of the word is separated by a space.This can have very bad consequences in information retrieval tasks,is there any way to solve this problem?