Welcome to Sign in | Help

Modele prea complexe in descoperirea de cunostinte

  •  07-30-2009, 9:55 AM

    Modele prea complexe in descoperirea de cunostinte

    Pentru cei care se ocupa cu descoperirea de cunostinte, nu exista o vorba: "Daca torturezi suficient de mult datele, acestea va marturisi aproape orice". Afirmatia este sprijinita matematic de o teorema a lui Boferroni, care afirma ca "as one performs an increasing number of statistical tests, the likelihood of getting an erroneous significant finding (Type I error) also increases". Se stie, de exemplu, în situatia data în "Principle of data mining": "One particularly humorous example of this type of prediction was provided by Leinweber (personal communication) who achieved almost perfect prediction of annual values of the well-known Standard and Poor 500 financial index as a function of annual values from previous years for butter production, cheese production, and sheep populations in Bangladesh and the United States."

    Ati întâlnit in practica situatia atunci când se utilizeaza un model prea complex si rezultatele au fost eronate? Puteti prezinta o astfel de situatie, împreuna cu modul de abordare pe care l-ati folosit?
View Complete Thread
Powered by Community Server (Commercial Edition), by Telligent Systems