Experimental research of automatic active knowledge bases profiling and optimization based on machine learning methods on the example of array sorting problem

Paper discusses the problem of automatic determination of non-functional properties of operations and variables based on profiling in the active knowledge concept. Under the topic a wide range of issues outlining the direction of further research are identified. The solution architecture is proposed, the prototype of this solution is described. The viability of the main idea of this work is shown experimentally on the example of a simple computational model.