I’m not sure I understand the problem. Do you have a string value s that contains HTML tags, and do you want to use things like INSTR and LIKE on s, but ignore the tags?
If so, you can use REGEXP_REPLACE starting in Oracle 10:
REGEXP_REPLACE (s, ‘<[^>]+>’, NULL)
returns a copy of s, without ‘<’s and anything that follows up to the next ‘>’.