Chatbot development help

I was developing Chatbot for telegram Where i used to scrap contents from websites using langchain webBaseLoader

But the problem is, the data was too rough (eg: one content title combines with another) and some times the entire data may not be useful or the contents are in non-English language

But i need only the contents to be in proper format as much as possible

Any better possible way, that can improve content scraping from websites?

I found, some of the API are available they provide better content scraping, but I’m student, so i can’t invest on those Free API was not enough for my purpose as well

Thankyou for everybody in advance ❤️