Internet contains large amount of data which user want to retrieve with the help of search input query. But the result return from the web has multiple dynamic output records. Hence, there is need of flexible information extraction system to convert web pages into machine process able structure which is essential for much application. This, essential information need to be extracted & annotated automatically which is challenge in data mining. In this paper, the author's survey on different HTML structure based technique to scrap data from web pages.