Our customers have very rich content in databases and even in files and spreadsheets. The answer is AKM Loader from Ardans' AKM Software.
Of course, we start at the beginning, i.e. from an empty database, we build its skeleton with the right business terminology.
With Ardans Knowledge Maker®, the AKM Software, the two immediately available structuring elements are view trees (i.e. ontologies, taxonomies, classification, etc.), and models (i.e. templates, classes, frameworks, etc., which are made up of headings, etc.).
Based on these models, the user generates articles ( i.e. forms, instances, etc.) and fills in the content headings ( i.e. text, image, video, numerical value, xml, computer code, file, etc.) to obtain this knowledge element, which will be dated and validated by an expert ( i.e. business referent, or even an expert or community of experts).
The incremental way of filling in a knowledge base is therefore the result of an interview with an expert. When we gather expertise, we fill in these elements, which will ultimately be validated, linked together, hooked to view trees, and made accessible or not to users (individuals or even groups of users).
Another way is to feed the knowledge base "by block" from an existing IT source.
In organisations, there are already IT sources that are managed either in a very institutional way or in a very personal way. In the latter case, we often speak of "Shadow IT", as this is valuable business data that is present on personal workstations, created by operational needs on individual initiative.
When a knowledge management initiative discovers this type of information asset, the first reaction is to preserve it, then to understand how it was built, how it has grown over time in complete secrecy and what its usefulness is in the business process or on a day-to-day basis.
Once this reflection and analysis have been carried out, it is possible to understand how to exploit it and make it available at the heart of the knowledge base.
The aim is then to answer the question "how can all this information be injected into the knowledge base in a relevant way?".
The experts at Ardans Labs have developed a simple procedure which provides the AKM knowledge base manager with a means of describing how to proceed with the automatic loading of records. This involves defining, for a record, which model to use and how to place the content and in which section, with which rights, tags, links, classification, author, validator, etc.
Then all you have to do is ask for the load to be executed, and the injection is carried out via this AKM Loader module.
The diagram above shows how the AKM Loader module loads records into a knowledge base in blocks.
Already used internally for operational projects, a richer version of AKM Loader has been industrialised and has been available since September 2023.
Photo credit © Ardans - Alain Berger August 2020. The " Le trouve tout du Livre " bookshop at the Port du Somail next to the Canal du Midi.
