Technical Features  K-Infinity covers the complete range of functions required for knowledge-based enterprise applications. Here we would like to present you with a brief description of the most important features:

System of Rights  In larger applications, managing profiles and access rights can grow into enormous time and effort spent on administration. K-Infinity's intelligence, which incorporates the context of the Knowledge Network in the search, also benefits the distribution of information and monitoring of authorizations: Imagine, for example, that all employees who have been involved in projects for the manufacturing industry in south-east Asia are to be informed of an industry fair in Japan. Or that they should all receive access to an image brochure specially created for the Asian economic region. It is here that the Knowledge Network comes into play: With the help of rules, it uses the relationship between individuals, events and regions to deduce who should receive the information, and who may receive it. intelligent views has applied for international patents for its process of deducing access rights from the network.

Multilingualism  Multilingualism and synonyms are genuine parts of a Knowledge Network: K-Infinity already offers you the option of finding objects under different names or creating a multilingual Knowledge Network. The links in the Knowledge Network are facts that are valid independently of language – which means that a Knowledge Web can also be a connector between different languages. Thus a search operating with German search expressions could also return, say, an English text.

Documents  One of the main benefits of Knowledge Networks is the thematic development of documents: For every business object, the user can find the documents relevant for the particular topic independent of individual words, spellings and phrasing.

How do the documents get linked to the objects of the Knowledge Network? Automatic classification processes have come on a long way in the last few years. On the basis of just a few training documents, they are capable of deciding to which objects of the Knowledge Network (also known as "classifiers") a new document is to be assigned. For average quality requirements, the best processes can be operated fully automatically. If 100% quality is aimed at, for instance in the case of documents that are provided at cost, it is a logical step to think over the automatic classification and correct it if desired. Alongside the quality of allocation (Precision and Recall), the individual processes differ in the number of training documents required per Knowledge Network object and in the degree of selectivity, i.e. the ability to find the right "classifiers" for a document even from a number volume of objects with fine differences.

In the field of automatic classification, we at intelligent views have had very positive experiences with the technology of our partner brainbot technologies AG and have used it successfully in various projects. The brainbot classification engine delivers outstanding quality and selectivity, and only requires a very small number of training documents.