Classification of guided expert systems according to their accuracy, domain resolution and range

Dan Ophir, Hila Samuel, Itzik Shmuelov, Iris Reychav

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations


An “Expert System” is based on “trial and error”. The configurations’ classification is difficult for the human user, however such categorization may be performed with the assistance of an human experts. The random configurations generated by the computer may be classified and for further computer expert system interrogation by the user. Imparting knowledge can be performed as follows: Transferring the knowledge – the instructor shows the collection of elements A={e1,e2,…, en} to the learning entity, and each value vi is assigned to the corresponding element ei (i=1,2,…n). Using the knowledge – the learning entity receives an element e, eU for consideration, where A U, and assigns to e its corresponding value, v. The learning entity scans the values; ei, finds the element ej, the closest to e, and joins it to the value vj. The computerized system is used as a model, enabling formalization and deduction for humans. In the computerized case, it is possible to observe a series of Expert Systems in the ascending orders of complexity, defined as follows: Binary case – the elements may receive only one of two values: ‘yes’ or ‘no’, without the possibility of attaching two different values to the same element. Ternary deterministic – the elements may receive one of three values ‘yes’, ‘no’, and ‘maybe’. Ternary nondeterministic – here the same element may appear with various values that are statistically supported. Canonization ofelements - here collection A is defined asA={A1, A2,…, Ak} whereAl(l=1,2,… k)represents: Al^e1l , e2l ,...,enl l ` the computerized system deduces, based on the elemente m l , the corresponding value, vl. Element-ranged canonization – here the collection of elements A is defined as follows: Al ^G1,G2,...,Gnl `(l=1,2,…k) A ={A1, A2,…, Ak, } G j >D j , E j @ is an interval of integer numbers; the computerized system deduces the corresponding value vl, relying on elements from the interval Aj. The following examples, graded according to their complexity and their level of uncertainty, represent the above Expert System’s classification: Determining correct blood transfusion; Teaching the Traffic Light rules in two modes: Deterministic Nondeterministic Identifying Deductive characters Recognizing Color Hues– based on the colors’ RGB ranges.

Original languageEnglish
Title of host publicationProceedings of the 19th European Conference on Knowledge Management, ECKM 2018
EditorsEleonora Di Maria, Enrico Scarso, Ettore Bolisani
Number of pages7
ISBN (Electronic)9781911218944
StatePublished - 2018
Event19th European Conference on Knowledge Management, ECKM 2018 - Padua, Italy
Duration: 6 Sep 20187 Sep 2018

Publication series

NameProceedings of the European Conference on Knowledge Management, ECKM
ISSN (Print)2048-8963
ISSN (Electronic)2048-8971


Conference19th European Conference on Knowledge Management, ECKM 2018


  • Expert system, knowledge transferring, SQL
  • Optical character recognition
  • Structured query language, determinism, OCR


Dive into the research topics of 'Classification of guided expert systems according to their accuracy, domain resolution and range'. Together they form a unique fingerprint.

Cite this