Dosar practică sisteme expert

Kamran Parsaye was one of the first researchers to apply artificial intelligence techniques to large scale database systems, and become known for coining the term Intelligent Database in a book of the same title in 1989. This book is written in an easy to read style and offers practical examples for each new topic presented. Emphasis is less on the intricacies of programming language, more on explanation. Defines what expert systems are, and discusses knowledge representation and interface. Chapters also cover logic, two-valued inference, inexact and semi-exact reasoning, advanced tools and topics, and draw on studies of human cognition to motivate technical definitions. Each chapter has an introduction and a summary, and provides suggestions for further reading. Knowledge is stored in frames with slot names and values. In backward chaining, the system has a goal to prove and it works backwards to prove it gaining the evidence with each iteration. The system attempts to match conclusions in the database until it finds one. In addition the system may prove the rule premise. Lastly, the AI system may try to backtrack within a premise. If a clause fails, it goes to the previous clause and tries to find new bindings for variables in the clause. In the event that everything fails, it finds a new rule structure. This work is a treasure chest of information for anyone seeking to design a workable AI system utilizing sound programming and logic principles.

Reviewed by:

II.REFERAT STIINTIFIC

IMPORTANTA TEHNOLOGIEI SISTEMELOR EXPERT IN CONTABILITATE

Abstract

Expert system are applicable accounting areas where an expert exists, definitions of facts and decision rules are clear, and the system application boundaries are known. They are computer products which simulates the performance of human expert and that are use in management accounting.

Key words: expert system, management accounting, technologies

I.INTRODUCERE

Sistemele expert sunt aplicabile in zonele de contabilitate unde exista un expert, faptele si regulile de decizie fiind clare iar limitele sistemului de aplicare fiind cunoscute. Ele nu pot fi flosite atunci cand datele sun incomplete sau ambigue.

II. SISTEME EXPERT

2.1 Delimitari teoretice

"Un sistem expert este un sistem care emuleaza abilitatea de a lua decizii a expertului uman. Termenul emuleaza inseamna ca sistemul este menit sa actioneze, in toate privintele ca expertul uman. Emularea fiind ceva mai mult decat simularea care cere doar sa se actioneze prin imitarea conditiilor relizate. Sistemele expert actioneaza foarte bine in domenii bine delimitate".

Modalitatea in care sistemul expert utilizeaza cunostintele cu care este dotat, releva o importanta deosebita deoarece un sistem expert trebuie sa dispuna atat de cunostinte cat si de mijloacele de utilizare efectiva pentru a fi considerat suficient de calificat intr-un anumit domeniu.

Structura unui sistem expert cuprinde urmatoarele componente de baza:

o Baza de cunostinte, pentru stocarea tutuor pieselor de cunoastere specific unui domeniu aplicativ;

o Motorul de inferente, un program care contine cunoasterea de control, pocedurala su operatorie, care exploateaza baza de cunostinte;

o Interfata de dialog cu utilizatorii care dispun si de un limbaj de exprimare a cunoasterii achizitionata de la expertii umani.

Un sistem expert trebuie sa asigure si functii complementare de achizitie cunoasteii si explicare a prototipului comportament. In acest caz, el va fi dotat cu doua component suplimentare si anume:

o Modulul de achizitionare a cunoasterii

o Modulul explicativ

In figura nr.1.1 este prezentata structura de baza a unui sistem expert cu cele cinci componente ale sale.

fapte

eurisci

explica efectueaza colecteza cunoasterea rationamente rationamente

consultatii

sfaturi

concluzii

explicatii

decizii

UTILIZATORI

Figura 1.1 Structura de baza a sistemelor expert

Sursa[1]

2.2 Domenii de aplicatie ale sistemeloe expert in contabilitate

Principalele domenii de aplicatie ale sistemelor expert in contabilitate sunt:auditing, planificarea impozitelor intreprinderii, determinarea impozitului pe profit, planificare financiara, contabilitate financiara si contbilitate de gestiune. In SUA, EY/Decision Support, Expertest, AuditPlanning, Advisor sunt numai unele dintre sistemele expert folosite in asistarea planificarii si derularii programelor de audit. Risk Advisor ,CCR/36 Advisor sunt sisteme expert menite sa identifice si sa evalueze riscul auditingului, VATIA si Loan Probe se utilizeaza in auditul unui domeniu specific.