[ad_1]
Structure mining uncovers and highlights elements of a semi-structured data source, such as websites or simple databases, to understand how pieces interact and find information under certain tags. It can also predict what an object is based on user-written rules and is useful for comparing and improving efficiency.
Structure mining is a type of data mining in which a semi-structured data source is scanned and elements of its structure are uncovered and highlighted. A semi-structured data source is one that doesn’t use the traditional database structure of tables, but has a semantic element that separates information using tags and markers. Structure mining can be used to mine databases, websites, and many other forms of computer information to discover structure elements. It helps users understand how pieces interact with each other or how to find information under certain tags. This mining can also be used to predict what an object is, based on user-written rules.
There are many different types of data mining, and most involve mining a source that is structured in the traditional way. This includes any source that uses the tables and nodes typical of most databases. Only semi-structured data is used in structure mining. In this case, the data comes from websites or simple databases that have a structure but does not conform to the rules of traditional databases. The data needs tags or indicators that distinguish each element to be correctly extracted.
By reading the semi-structured data set, structure mining is able to find out how the structure interacts. For example, every website has a navigation pattern, and it’s this pattern that determines how pages interact. By extracting the structure, the user can find out how this navigation works, which can help create a similar navigation scheme.
Structure mining can also be used to find objects by writing rules to the mining program. For example, if there is a book dataset, the user can write a rule that all books with no index should be returned as fiction and those with an index should be returned as non-fiction. Most fiction books don’t have an index, so this rule predicts with high accuracy what the data is. This helps users when they are looking at a semi-structured set that has a method of organization but not one that fits what the user is looking for.
After locating the structure of the semi-structured drive, the user will typically compare it to another semi-structured drive. If the user has a corporate website, he can check out another corporate website for navigation and links and see how his website looks like. By comparing the extracted information, the user can find ways to increase the efficiency of the facility.
[ad_2]