Back to Index

Tutorial: Classification

This tutorial provides information for defining a workset by classification of its workparts. By classifying workparts that you know using characteristics you are interested in, you can compare how other workparts in the workset relate to each other.


Classification

  1. You can load the Workset in two ways.

    i) You can select the Classification toolset and click continue button, then, select and Load the Workset you want to work with, or ...

  2. ii) You can select both the Classification toolset and the Workset you want to examine, click continue button, then click on Load.

  3. Once the Workset is loaded, the Workparts become available. The text of the workpart can be viewed at any time on the Text Viewer on the right.

    This tool also has a concordance search function, available under the "options" pane. The concordance search function allows you to form three different types of queries. You can search by part of speech, by lemma or by spelling. For example, to search for all nouns in a work you would form the query "* (n)". The lemma search works in the same way as the selection pane. For example, to search for the lemma love as a noun, form the query "love (n)." Spelling is simplest; just type the word you're looking for.

    The concordance search also accepts the "*" and "|" operators used throughout MONK. You may not combine different types of searches. For example, searching spelling and lemma in one query (ex. "king (n*) | life") will not work. If you use multiple terms in a query, the concordance search will assume "or" statements between them.

    The advanced viewer also has a TEI header tab available under options. This gives a variety of contextual information about the document, including its provenance.

  4. To rate some of the Workparts click on the Click to Rate text. You can type your rating of the Workpart in the text box that appears. If you have already begun rating the Workparts, a dropdown of your previous ratings appear for you to choose from.

  5. Once you have rated the desired Workparts you can save the rating, or save as, to append to an earlier rating.

  6. You can Revert your rating to the last time the ratings were saved using the Revert button.

  7. You can remove a Workpart that you do not wist to work with or include in the analysis.

  8. You can also export the ratings as Tab Separated Values and import them later using the Import/Export button.

  9. Once you have rated the desired Workparts click on the Perform Analysis button. You can choose either a Naive Bayes analysis, or a Naive Bayes with Decision Tree analysis. Explanations on Naive Bayes and decision trees are available in the analytics articles. The basic difference between these two algorithms is that Naive Bayes calculates probabilities by feature, whereas the Naive Bayes with decision tree utilizes the classification output of the Naive Bayes algorithm.

    Depending on the size of the Workset you are analyzing, the results can be displayed immediately, or, may take a long time to process.

    If you are working with a large enough Workset that can take a few hours to analyze, you can type in your Email address so that the server can notify you when the analysis is over.

  10. If your Workset is not too large and the Analysis is completed, you can click on the continue button to see the results. You can also click on the Workpart itself to view the text in the Text Viewer.

  11. If you chose the Naive Bayes with Decision Tree algorithm, you can also access an image of the decision tree.

  12. If you were working with a large Workset and chose to come back to MONK at a later time, you will get an email notifying you when the analysis is completed. In that case, you can see the result in the MONK Workbench when you log back in. You can simply select and click continue or double-click the result to see the analysis results.


Go to the Index Page or back to the Top of this page