The techniques used in file automatic classification include:
Catalog By Example, Statistical, Rule-Based
.
Catalog By Example
It is the technology used by the most mainstream manufacturers, most emphasizes artificial intelligence, fuzzy computing and other technologies to conduct a textual analysis, automatically generate keywords, and automatic classification according to intra. "Model Training" is made by the user's small amount of classification sample data, and the model of classification behavior is established. When there is a new file in the future, when it enters the system, according to the established model, each file can be automatically attributed to one or more categories.
However, in addition to information technology such as artificial intelligence and nerve treatment, it is often possible to settle a large amount of data through more obvious rules. In addition to greatly enhance system performance, it is also possible to make the system with maximum elasticity, quickly adapt to the needs of enterprises or websites, which is
Rule-based
The maximum features and advantages of file automatic classification system.
Further, it can also
Catalog By Example
and
Rule-based
Two technical integration, that is, when "Model Training", the model rules, organize various rules that can be checked, revised, and the rules of recession, reducing the rules of the recession, and reducing the management Set a large number of ruled time and cost
