header

NBC : Classifier in supervised learning
Tag:
Edit edit   Starstar

Program Information

Name: NBC
Domain: Machine learning
Functionality: Classifier in supervised learning
Input:
     T: The training data (Type: List) D: The testing data (Type: List)
Output:
     L: The label of testing data (Type : List)

Reference

              Application of Metamorphic Testing to Supervised Classifiers http://dx.doi.org/10.1109/QSIC.2009.26 Testing and validating machine learning classifiers by metamorphic testing http://dx.doi.org/10.1016/j.jss.2010.11.920 Testing Approach for Dynamic Web Applications Based on Automated Test Strategies http://dx.doi.org/10.1007/978-3-319-03095-1_43

MR Information


MR1


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Value of any subset of features in Tf = k * correspoding value in Ts + b, (k != 0), Value of any subset of features in Df = k * correspoding value in Ds + b, (k != 0),
    Output relation: Lf = Ls

MR2


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Class labels in Tf = Permutation of class labels in Ts.
    Output relation: Lf = Permutation of Ls

MR3


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Attributes of Tf = Permutation of attributes of all the samples of Ts, Attributes of Df = Permutation of attributes of all the samples of Ds
    Output relation: Lf = Ls

MR4


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts with a new uninformative attribute, Df = Ds with the same one
    Output relation: Lf = Ls

MR5


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts with a new informative attribute, Df = Ds with the same one
    Output relation: Lf = Ls

MR6


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts + any number of examples in Ts which have label Ls , Df = Ds
    Output relation: Lf = Ls

MR7


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts + any number of examples in Ts which do not have label Ls , concatenate an arbitrary symbol “*” to the class labels of the duplicated examples , Df = Ds
    Output relation: Lf = Ls

MR8


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts with some classes which is not Label Ls removed , Df = Ds
    Output relation: Lf = Ls

MR9


    Source input: <Ts,Ds> ; Source output: Ls
    Follow-up input: <Tf, Df > ; Follow-up output: Lf
    Input relation: Tf = Ts permuting the values of any attribute among examples sharing the same class ,Df = Ds
    Output relation: Lf = Ls

Related

Insert title here