Intelligent Content Environment
In a
day-to-day life, we are used to handling and absorbing a lot of data from
various sources, covering different areas of our personal and business life. Our
brains are well trained with processing this data and qualifying it as valuable
information, or disposing of it as irrelevant data.
In our
business processes, we are dealing with data in forms of computer documents and
other types of files. Since computers are already helping us write those
documents and exchange them with other users, why not try and ‘teach’ our
computers to also ‘understand’ those documents? By thorough examination and
analysis of the documents’ content, the computer would be able to recognize
some of its key parts and points. Based on those points, it could be able to
make a decision, whether some particular document contains some information,
valuable to us, or not, and act upon it.
Artificial
Intelligence is about ‘knowing’ and acting on that knowledge. Nowadays, there
is plenty of valuable information published on the internet. We have
encyclopedias, news and other media agencies, dedicated content stores, all
available to us for immediate use. We can only absorb a specific amount of
information in a given time, but computers, with all the storage options
available today, can store much more information than a single human’s brain
can. For a human, learning means reading books, attending lectures and
conversing with other humans, which takes us a lot of valuable time. On the
other hand, for a machine, learning could simply mean accessing some web sites
with a specific query, carefully processing and analyzing the returned content,
then classifying it by estimated relevance. Allowing computer system to expand
its ‘knowledge’ this way would prove an incredible benefit in the future
user-computer interaction. With storing and constantly expanding users’
profile, by remembering their input and tracking their behavior, it would make
that a very personal relationship. Imagine computer assistants, helping people
by instantly serving relevant information, when they need it, how they need it,
in a friendly, conversation-like form.
We need to
build a base for a powerful, but user-friendly Content Management System, which
would later develop into AI-powered Intelligent Content Environment. Such
system should be accessible from practically everywhere, anytime, and from
different devices or media. Its main features should include automatic data acquisition
from various (on-line) resources; intelligent content recognition of either
structured or non-structured data; complex data analysis using text patterns
and other data mining techniques; reliable storage of any standard or
user-defined data format; content versioning with full change history; advanced
classification and information cataloguing using context tags; cross-linking
and referencing various internal, as well as external pieces of information,
complex full-text and date-based (version) searching; customized content
publishing and delivering; granular customizable security with user roles;
intelligent analysis of users’ behavior, allowing system to associate users
with containing data and create appropriate responses. All features would be rounded
up in whole with the advanced workflow options for maximizing user experience
in the whole content management processing.
Andrej Tozon