Multiclass evaluation problem

sgmeiresgmeire Member Posts: 9 Contributor II
edited November 2018 in Help
Hi!

I´m using a cross-validation for a multiclass classification problem and I'm using a performance evaluation model for multiclass. I'd like to obtain the AUC measure but it is only possible for binary class. How can I do this? Can anybody help me?

Thank you,
Silvana.

Answers

  • wesselwessel Member Posts: 537 Maven
    Hey,

    I assume you referring to a nominal classification problem.
    As opposed to a bi-nominal classification problem.
    The term multi-class is sometimes used to refer to problems where an instances can belong to multiple classes, e.g. some instances belonging to both the sports and football class.

    AUC, area under ROC curve.
    ROC, Receiver Operating Characteristic, true positive rate plotted against the false positive rate.
    True positive rate, and true negative rate are only defined for bi-nominal classification problems.

    You could transform your problem into a bi-nominal classification problem.
    Lets say you have three classes A, B and C.
    You can then make 3 different datasets with classes A and not A, B and not B, C and not C.
    And then obtain the AUC score on each dataset.
    I doubt this is a useful measure though.

    Have you considered the Cohen's kappa coefficient as a measure of performance?

    Or maybe area under a learning curve?

  • mbeckmannmbeckmann Member Posts: 15 Contributor II
    Hi,

    This is a solution to implement a one against all strategy in order to obtain AUC results for multinomial problem. Nevertheless, this is painfull if we are dealing with databases.

    Please let me know if this could be solved with some few operators.

    Otherwise, you can count with me to collaborate in order to improve some operator.

    Thanks in advance for your help,
  • wesselwessel Member Posts: 537 Maven
    I always use the Weka experimenter if I want to compute AUC for mutli class problem.

    I don't think there is no easy way to get AUC for multi class problem in Rapid Miner.

    Strange :s

  • mbeckmannmbeckmann Member Posts: 15 Contributor II
    Hi there,


    I'm able to modify the AUC performance evaluation operator and put a check box like "use one against all strategy" , and implement this strategy in order to compute AUC performance for multinomial problem.s

    I'll let you know when done.

    Best regards,
  • mbeckmannmbeckmann Member Posts: 15 Contributor II
    Hi,

    There's no need to change any operator, or even migrate to weka in order to obtain AUC for polynomial classification problems. As Sebastian Land told me, the Rapidminer team aims to cover most of the use cases for data mining. Just use the Polynomial by Binomial Classification, and this operator will perform a 1-against-all strategy and other strategies.

    This topic has an example process showing how to use this operator:

    http://rapid-i.com/rapidforum/index.php?topic=2505.0

  • mbeckmannmbeckmann Member Posts: 15 Contributor II

    Hi all,

     

    There's a long time since I don't post anytime here. Last time I made this last comment, saying the polynomial by binomial classification will solve this problem, but in fact it didn't.

     

    Even right now with RM 7, it;s not possible to obtain a simple AUC or f-measure with a multiclass problem, for example, the Iris dataset, the most fundamental of machine learning problems.

     

    You know there's mathematical formulation to calculate other evaluation measurements for multiclass problems (called by RM as binomial and polynomial problems). In RM there is a clear distinction what operators can manipulate binomial and a polynomial classification problems, but this distinction should not exist anymore.

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist

    Hi,

     

    i am curious, how do you calculate AUC for a polynominal problem? And why don't you take logloss?

     

    Best,

    Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    In my experience, using AUC for multi-class classification problems is typically done by looking at separation of one class vs all others.  It would certainly be nice if RapidMiner did this automatically for polynominal labels, but it should be able to be done manually by remapping labels after the fact, I think.

     

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
  • mbeckmannmbeckmann Member Posts: 15 Contributor II

    Hi Martin,

     

    Thanks for your reply.

     

    I'm a RapidMiner fan, and I here to help some how. Sincerelly I thought this problem was solved in the newest versions, but I think it's not fair to say an operator does not work, without a deep research, so, my apologies. Nevertheless, so far I could not find a way to calculate AUC and f-measure for a multiclass problem, using for example the IRIS dataset, with 3 classes. What I could notice in other tools is: for f-measure, they make an average or a weighted average, and other choices of statistical mean for f-measure. For AUC basicall  they are following the formulation existing in the literature:

     

    Fawcett, T., 2006, An introduction to ROC analysis, Pattern Recognition Letters 27, 861-874.
    Hand, D.J., Till, R.J., 2001. A simple generalization of the area under the ROC curve to multiple class classification problems. Machine Learning 45 (2), 171-186.

     

    If you could point me out what operators should I use to do this in RapidMiner it will be great, as I'm evaluating this for a company.

     

    To overcome this problem in my personal research with multiclass problem I made an operator to integrate the RM predictions with the Weka evaluator, then I've got all the predictive results from Rapidminer, measured by the Weka classifier evaluator. However, just as a constructive thought, even if the solution for this is already there in RapidMiner, I think it's time to rethink how to do this rather than using the concept of binomial and polynomial, as others DM tools are progressing well without this.

     

     

     

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist

    Hi again,

     

    i need to disagree. I do think that the difference between polynominal and binominal is pretty clear. There are simply models like an SVM which can not cope with binominal data. That is a fact. LibSVM is simply using a internal wrapper to do one-vs-all and thus make a binominal algorithm runable on a polynominal data set. That is fully supported with Polynominal by Binominal Classification operator

     

    The point you raise, is that you would also like to be able to use a binominal performance measure for a polynominal problem. While i see that this is a possible approach, i would argue that logloss is a better measure. The missing operator in RM would be Polynominal by Binominal Performance, which would be a similar nested operator like the learner. Should not be hard to write.

     

    ~Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • mbeckmannmbeckmann Member Posts: 15 Contributor II

    Hi,

     

    I could make a process to calculate AUC and F-Measure for 3 classes with Iris dataset (Thanks Brian T. for the tip with Map), using the Polynomial by Binomial Classification and SVM, find below. You can also get the ROC plots by class. The Polynomial by Binomial Classification is working fine.

     

    The con with this process is: to calculate the performaces with Performance Binomial I need to remap each label agains all, to calculate the performance averages by class I transformed the performance to data, but it delivers what is wanted.

     

    In order to contribute to improve this scenario somehow, I'm preparing my Weka Classification Measure Operator to output performance (right now it returns data), then I'll offer that code to be integrated to Weka extension.

     

    Anyway I'm packaging my Weka Classification Measure Operator as is in the next days, and post it here.

     

    Regards,

     

    <?xml version="1.0" encoding="UTF-8"?><process version="7.3.001">
    <context>
    <input/>
    <output/>
    <macros/>
    context>
    <operator activated="true" class="process" compatibility="7.3.001" expanded="true"
    name="Process">
    <parameter key="logverbosity" value="init"/>
    <parameter key="random_seed" value="2001"/>
    <parameter key="send_mail" value="never"/>
    <parameter key="notification_email" value=""/>
    <parameter key="process_duration_for_mail" value="30"/>
    <parameter key="encoding" value="SYSTEM"/>
    <process expanded="true">
    <operator activated="true" class="retrieve" compatibility="7.3.001" expanded="true"
    height="68" name="Retrieve Iris" width="90" x="45" y="34">
    <parameter key="repository_entry" value="//Samples/data/Iris"/>
    </operator>
    <operator activated="true" class="split_data" compatibility="7.3.001" expanded="true"
    height="103" name="Split Data" width="90" x="179" y="85">
    <enumeration key="partitions">
    <parameter key="ratio" value="0.7"/>
    <parameter key="ratio" value="0.3"/>
    </enumeration>
    <parameter key="sampling_type" value="shuffled sampling"/>
    <parameter key="use_local_random_seed" value="false"/>
    <parameter key="local_random_seed" value="1992"/>
    </operator>
    <operator activated="true" class="polynomial_by_binomial_classification"
    compatibility="7.3.001" expanded="true" height="82" name="Polynomial by Binomial
    Classification" width="90" x="447" y="85">
    <parameter key="classification_strategies" value="1 against all"/>
    <parameter key="random_code_multiplicator" value="2.0"/>
    <parameter key="use_local_random_seed" value="false"/>
    <parameter key="local_random_seed" value="1992"/>
    <process expanded="true">
    <operator activated="true" class="support_vector_machine_libsvm"
    compatibility="7.3.001" expanded="true" height="82" name="SVM" width="90" x="112"
    y="34">
    <parameter key="svm_type" value="C-SVC"/>
    <parameter key="kernel_type" value="rbf"/>
    <parameter key="degree" value="3"/>
    <parameter key="gamma" value="0.125"/>
    <parameter key="coef0" value="0.0"/>
    <parameter key="C" value="2.0"/>
    <parameter key="nu" value="0.5"/>
    <parameter key="cache_size" value="80"/>
    <parameter key="epsilon" value="0.001"/>
    <parameter key="p" value="0.1"/>
    <list key="class_weights"/>
    <parameter key="shrinking" value="true"/>
    <parameter key="calculate_confidences" value="false"/>
    <parameter key="confidence_for_multiclass" value="true"/>
    </operator>
    <connect from_port="training set" to_op="SVM" to_port="training set"/>
    <connect from_op="SVM" from_port="model" to_port="model"/>
    <portSpacing port="source_training set" spacing="0"/>
    <portSpacing port="sink_model" spacing="0"/>
    </process>
    </operator>
    <operator activated="true" class="apply_model" compatibility="7.3.001" expanded="true"
    height="82" name="Apply Model (2)" width="90" x="313" y="289">
    <list key="application_parameters"/>
    <parameter key="create_view" value="false"/>
    </operator>
    <operator activated="true" class="multiply" compatibility="7.3.001" expanded="true"
    height="124" name="Multiply" width="90" x="112" y="442"/>
    <operator activated="true" class="map" compatibility="7.3.001" expanded="true"
    height="82" name="vir-all" width="90" x="112" y="748">
    <parameter key="attribute_filter_type" value="subset"/>
    <parameter key="attribute" value="label"/>
    <parameter key="attributes" value="label|prediction(label)"/>
    <parameter key="use_except_expression" value="false"/>
    <parameter key="value_type" value="attribute_value"/>
    <parameter key="use_value_type_exception" value="false"/>
    <parameter key="except_value_type" value="time"/>
    <parameter key="block_type" value="attribute_block"/>
    <parameter key="use_block_type_exception" value="false"/>
    <parameter key="except_block_type" value="value_matrix_row_start"/>
    <parameter key="invert_selection" value="false"/>
    <parameter key="include_special_attributes" value="true"/>
    <list key="value_mappings">
    <parameter key="Iris-setosa" value="all"/>
    <parameter key="Iris-versicolor" value="all"/>
    </list>
    <parameter key="consider_regular_expressions" value="false"/>
    <parameter key="add_default_mapping" value="false"/>
    </operator>
    <operator activated="true" class="performance_binominal_classification"
    compatibility="7.3.001" expanded="true" height="82" name="vir-all-perf" width="90"
    x="246" y="1003">
    <parameter key="main_criterion" value="first"/>
    <parameter key="accuracy" value="false"/>
    <parameter key="classification_error" value="false"/>
    <parameter key="kappa" value="false"/>
    <parameter key="AUC (optimistic)" value="false"/>
    <parameter key="AUC" value="true"/>
    <parameter key="AUC (pessimistic)" value="false"/>
    <parameter key="precision" value="false"/>
    <parameter key="recall" value="false"/>
    <parameter key="lift" value="false"/>
    <parameter key="fallout" value="false"/>
    <parameter key="f_measure" value="true"/>
    <parameter key="false_positive" value="false"/>
    <parameter key="false_negative" value="false"/>
    <parameter key="true_positive" value="false"/>
    <parameter key="true_negative" value="false"/>
    <parameter key="sensitivity" value="false"/>
    <parameter key="specificity" value="false"/>
    <parameter key="youden" value="false"/>
    <parameter key="positive_predictive_value" value="false"/>
    <parameter key="negative_predictive_value" value="false"/>
    <parameter key="psep" value="false"/>
    <parameter key="skip_undefined_labels" value="true"/>
    <parameter key="use_example_weights" value="false"/>
    </operator>
    <operator activated="true" class="map" compatibility="7.3.001" expanded="true"
    height="82" name="set-all" width="90" x="380" y="697">
    <parameter key="attribute_filter_type" value="subset"/>
    <parameter key="attribute" value="label"/>
    <parameter key="attributes" value="label|prediction(label)"/>
    <parameter key="use_except_expression" value="false"/>
    <parameter key="value_type" value="attribute_value"/>
    <parameter key="use_value_type_exception" value="false"/>
    <parameter key="except_value_type" value="time"/>
    <parameter key="block_type" value="attribute_block"/>
    <parameter key="use_block_type_exception" value="false"/>
    <parameter key="except_block_type" value="value_matrix_row_start"/>
    <parameter key="invert_selection" value="false"/>
    <parameter key="include_special_attributes" value="true"/>
    <list key="value_mappings">
    <parameter key="Iris-virginica" value="all"/>
    <parameter key="Iris-versicolor" value="all"/>
    </list>
    <parameter key="consider_regular_expressions" value="false"/>
    <parameter key="add_default_mapping" value="false"/>
    </operator>
    <operator activated="true" class="performance_binominal_classification"
    compatibility="7.3.001" expanded="true" height="82" name="set-all-perf" width="90"
    x="581" y="901">
    <parameter key="main_criterion" value="first"/>
    <parameter key="accuracy" value="false"/>
    <parameter key="classification_error" value="false"/>
    <parameter key="kappa" value="false"/>
    <parameter key="AUC (optimistic)" value="false"/>
    <parameter key="AUC" value="true"/>
    <parameter key="AUC (pessimistic)" value="false"/>
    <parameter key="precision" value="false"/>
    <parameter key="recall" value="false"/>
    <parameter key="lift" value="false"/>
    <parameter key="fallout" value="false"/>
    <parameter key="f_measure" value="true"/>
    <parameter key="false_positive" value="false"/>
    <parameter key="false_negative" value="false"/>
    <parameter key="true_positive" value="false"/>
    <parameter key="true_negative" value="false"/>
    <parameter key="sensitivity" value="false"/>
    <parameter key="specificity" value="false"/>
    <parameter key="youden" value="false"/>
    <parameter key="positive_predictive_value" value="false"/>
    <parameter key="negative_predictive_value" value="false"/>
    <parameter key="psep" value="false"/>
    <parameter key="skip_undefined_labels" value="true"/>
    <parameter key="use_example_weights" value="false"/>
    </operator>
    <operator activated="true" class="performance_to_data" compatibility="7.3.001"
    expanded="true" height="82" name="Performance to Data (2)" width="90" x="782" y="901"/>
    <operator activated="true" class="map" compatibility="7.3.001" expanded="true"
    height="82" name="ver-all" width="90" x="581" y="646">
    <parameter key="attribute_filter_type" value="subset"/>
    <parameter key="attribute" value="label"/>
    <parameter key="attributes" value="label|prediction(label)"/>
    <parameter key="use_except_expression" value="false"/>
    <parameter key="value_type" value="attribute_value"/>
    <parameter key="use_value_type_exception" value="false"/>
    <parameter key="except_value_type" value="time"/>
    <parameter key="block_type" value="attribute_block"/>
    <parameter key="use_block_type_exception" value="false"/>
    <parameter key="except_block_type" value="value_matrix_row_start"/>
    <parameter key="invert_selection" value="false"/>
    <parameter key="include_special_attributes" value="true"/>
    <list key="value_mappings">
    <parameter key="Iris-virginica" value="all"/>
    <parameter key="Iris-setosa" value="all"/>
    </list>
    <parameter key="consider_regular_expressions" value="false"/>
    <parameter key="add_default_mapping" value="false"/>
    </operator>
    <operator activated="true" class="performance_binominal_classification"
    compatibility="7.3.001" expanded="true" height="82" name="ver-all-perf" width="90"
    x="581" y="799">
    <parameter key="main_criterion" value="first"/>
    <parameter key="accuracy" value="false"/>
    <parameter key="classification_error" value="false"/>
    <parameter key="kappa" value="false"/>
    <parameter key="AUC (optimistic)" value="false"/>
    <parameter key="AUC" value="true"/>
    <parameter key="AUC (pessimistic)" value="false"/>
    <parameter key="precision" value="false"/>
    <parameter key="recall" value="false"/>
    <parameter key="lift" value="false"/>
    <parameter key="fallout" value="false"/>
    <parameter key="f_measure" value="true"/>
    <parameter key="false_positive" value="false"/>
    <parameter key="false_negative" value="false"/>
    <parameter key="true_positive" value="false"/>
    <parameter key="true_negative" value="false"/>
    <parameter key="sensitivity" value="false"/>
    <parameter key="specificity" value="false"/>
    <parameter key="youden" value="false"/>
    <parameter key="positive_predictive_value" value="false"/>
    <parameter key="negative_predictive_value" value="false"/>
    <parameter key="psep" value="false"/>
    <parameter key="skip_undefined_labels" value="true"/>
    <parameter key="use_example_weights" value="false"/>
    </operator>
    <operator activated="true" class="performance_to_data" compatibility="7.3.001"
    expanded="true" height="82" name="Performance to Data" width="90" x="782" y="799"/>
    <operator activated="true" class="union" compatibility="7.3.001" expanded="true"
    height="82" name="Union" width="90" x="983" y="901"/>
    <operator activated="true" class="performance_to_data" compatibility="7.3.001"
    expanded="true" height="82" name="Performance to Data (3)" width="90" x="782" y="1003"/>
    <operator activated="true" class="union" compatibility="7.3.001" expanded="true"
    height="82" name="Union (2)" width="90" x="1117" y="1003"/>
    <operator activated="true" class="select_attributes" compatibility="7.3.001"
    expanded="true" height="82" name="Select Attributes" width="90" x="1050" y="748">
    <parameter key="attribute_filter_type" value="subset"/>
    <parameter key="attribute" value=""/>
    <parameter key="attributes" value="Value|Criterion"/>
    <parameter key="use_except_expression" value="false"/>
    <parameter key="value_type" value="attribute_value"/>
    <parameter key="use_value_type_exception" value="false"/>
    <parameter key="except_value_type" value="time"/>
    <parameter key="block_type" value="attribute_block"/>
    <parameter key="use_block_type_exception" value="false"/>
    <parameter key="except_block_type" value="value_matrix_row_start"/>
    <parameter key="invert_selection" value="false"/>
    <parameter key="include_special_attributes" value="false"/>
    </operator>
    <operator activated="true" class="generate_attributes" compatibility="7.3.001"
    expanded="true" height="82" name="AUC FIX Values" width="90" x="983" y="544">
    <list key="function_descriptions">
    <parameter key="Value" value="if(Criterion==&quot;AUC&quot; &amp;&amp;
    Value==0,1,Value)"/>
    </list>
    <parameter key="keep_all" value="true"/>
    </operator>
    <operator activated="true" class="aggregate" compatibility="7.3.001" expanded="true"
    height="82" name="Aggregate" width="90" x="983" y="340">
    <parameter key="use_default_aggregation" value="false"/>
    <parameter key="attribute_filter_type" value="all"/>
    <parameter key="attribute" value=""/>
    <parameter key="attributes" value=""/>
    <parameter key="use_except_expression" value="false"/>
    <parameter key="value_type" value="attribute_value"/>
    <parameter key="use_value_type_exception" value="false"/>
    <parameter key="except_value_type" value="time"/>
    <parameter key="block_type" value="attribute_block"/>
    <parameter key="use_block_type_exception" value="false"/>
    <parameter key="except_block_type" value="value_matrix_row_start"/>
    <parameter key="invert_selection" value="false"/>
    <parameter key="include_special_attributes" value="false"/>
    <parameter key="default_aggregation_function" value="average"/>
    <list key="aggregation_attributes">
    <parameter key="Value" value="average"/>
    <parameter key="Value" value="standard_deviation"/>
    </list>
    <parameter key="group_by_attributes" value="Criterion"/>
    <parameter key="count_all_combinations" value="false"/>
    <parameter key="only_distinct" value="false"/>
    <parameter key="ignore_missings" value="true"/>
    </operator>
    <connect from_op="Retrieve Iris" from_port="output" to_op="Split Data"
    to_port="example set"/>
    <connect from_op="Split Data" from_port="partition 1" to_op="Polynomial by Binomial
    Classification" to_port="training set"/>
    <connect from_op="Split Data" from_port="partition 2" to_op="Apply Model (2)"
    to_port="unlabelled data"/>
    <connect from_op="Polynomial by Binomial Classification" from_port="model"
    to_op="Apply Model (2)" to_port="model"/>
    <connect from_op="Apply Model (2)" from_port="labelled data" to_op="Multiply"
    to_port="input"/>
    <connect from_op="Multiply" from_port="output 1" to_op="ver-all" to_port="example set
    input"/>
    <connect from_op="Multiply" from_port="output 2" to_op="set-all" to_port="example set
    input"/>
    <connect from_op="Multiply" from_port="output 3" to_op="vir-all" to_port="example set
    input"/>
    <connect from_op="vir-all" from_port="example set output" to_op="vir-all-perf"
    to_port="labelled data"/>
    <connect from_op="vir-all-perf" from_port="performance" to_op="Performance to Data
    (3)" to_port="performance vector"/>
    <connect from_op="set-all" from_port="example set output" to_op="set-all-perf"
    to_port="labelled data"/>
    <connect from_op="set-all-perf" from_port="performance" to_op="Performance to Data
    (2)" to_port="performance vector"/>
    <connect from_op="Performance to Data (2)" from_port="example set" to_op="Union"
    to_port="example set 2"/>
    <connect from_op="ver-all" from_port="example set output" to_op="ver-all-perf"
    to_port="labelled data"/>
    <connect from_op="ver-all-perf" from_port="performance" to_op="Performance to Data"
    to_port="performance vector"/>
    <connect from_op="Performance to Data" from_port="example set" to_op="Union"
    to_port="example set 1"/>
    <connect from_op="Union" from_port="union" to_op="Union (2)" to_port="example set 1"/>
    <connect from_op="Performance to Data (3)" from_port="example set" to_op="Union (2)"
    to_port="example set 2"/>
    <connect from_op="Union (2)" from_port="union" to_op="Select Attributes"
    to_port="example set input"/>
    <connect from_op="Select Attributes" from_port="example set output" to_op="AUC FIX
    Values" to_port="example set input"/>
    <connect from_op="AUC FIX Values" from_port="example set output" to_op="Aggregate"
    to_port="example set input"/>
    <connect from_op="Aggregate" from_port="example set output" to_port="result 1"/>
    <connect from_op="Aggregate" from_port="original" to_port="result 2"/>
    <portSpacing port="source_input 1" spacing="0"/>
    <portSpacing port="sink_result 1" spacing="0"/>
    <portSpacing port="sink_result 2" spacing="0"/>
    <portSpacing port="sink_result 3" spacing="0"/>
    </process>
    </operator>
    </process>

     

     

  • 994kaca994kaca Member Posts: 5 Contributor I
    Hi there, i have problem with classificiaton of products. I have to classify products in 9 different classes. The objective is to build a predictive model which is able to distinguish between main product categories. I have dataset with 93 features an 61878 examples, can anybody help?
Sign In or Register to comment.