this article adopts several machine-learning techniques including machine-learning method based on user " s feedbacks, machine-learning method based on observing, inductive method based on id3 algorithm, heuristic learning method, knowledge base technique . the system has good intelligence and agency 本文采用了以下幾種機(jī)器學(xué)習(xí)技術(shù):基于用戶反饋的機(jī)器學(xué)習(xí)技術(shù)、基于觀察記憶的機(jī)器學(xué)習(xí)方法、基于id3算法的歸納推理、啟發(fā)式學(xué)習(xí)方法、知識(shí)庫(kù)技術(shù),系統(tǒng)具有較好的自主性和代理性。
based on nt 4.0 operating system and the sql sever 2000 database, building the system needs using multiple languages, such as microsoft visual basic 6.0, etc . building the financial diagnosis system needs sufficiently utilize different advanced techniques of software engineering, such as object oriented, groupware technique, the well combination of the knowledge base technique and the data storehouse digging technique to develop expert system, which causes the system has many characteristics, such as simple operation, good stability, good maintainability, and so on 本系統(tǒng)開發(fā)基于nt4.0操作平臺(tái),采用sqlsever2000數(shù)據(jù)庫(kù)和microsoftvisualbasic6.0開發(fā)語(yǔ)言。系統(tǒng)開發(fā)充分利用現(xiàn)代軟件開發(fā)的各種先進(jìn)技術(shù)和理念,如面向?qū)ο蠹夹g(shù)、組件技術(shù)等;并利用知識(shí)庫(kù)技術(shù)與數(shù)據(jù)挖掘技術(shù)的良好結(jié)合來(lái)開發(fā)專家?guī)欤顾_發(fā)的系統(tǒng)具有操作簡(jiǎn)易、穩(wěn)定性好、可維護(hù)性好等特點(diǎn)。