No label defined (Q4028031)

[For the entire collection see Zbl 0747.00037.] The paper presents the current research at the Numerical Algorithms group (NAG) Ltd in numerical statistical Knowledge-Based Front-Ends (KBFE). NAG is researching and developing KBFE as a way to bridge the gap between the user's application domain and successful application of the package or library without discarding it. The goals of the KBFE project are: maintaining the library or package as the underlying algorithm source for resolving user tasks; capturing the encorporating knowledge about the problem domain into the KBFE (KBFE is much like an expert system); making the back-end's facilities easier to use and making them more extensive; enlarging the scope of potential library users beyond those with computing expertise; adapting the KBFE to a wider range of users (this allows the KBFE to intelligently choose, present, and adapt information to individual users in varying contexts; providing a more current user interface comparable to those on domain- specific packages-especially, Window Icon Mouse Pointer-based presentation on bit-mapped displays, including graphics.