0 ×

Text Classifier Predictor

Palladian Nodes for KNIME Workbench version 1.8.0.201907271536 by palladian.ws; Philipp Katz, Klemens Muthmann, David Urbansky.

This node uses the dictionary build with the corresponding learner node and predicts the categories for uncategorized documents by looking up the relevance scores in the dictionary and assigning the categories with the highest probability.

This classifier won the first Research Garden competition where the goal was to classify product descriptions into eight different categories. See press release.

The complementary Naïve Bayes scoring has been described in "Tackling the Poor Assumptions of Naive Bayes Text Classifiers"; Jason D. M. Rennie; Lawrence Shih; Jaime Teevan; David R. Karger; 2003. This way, not the term counts in the regarded class, but the counts from all other classes are regarded for each class prediction. This leads to better classification accuracy.

Options

Text input
The column containing the text which will be used for classification.
Append columns with class distribution
Activate to append a column for each class holding the probability value.

Advanced

Scoring formula
Allows to customize the scoring. Usually, the "Palladian" option provides a good tradeoff between speed and accuracy, while the complementary Naïve Bayes scorer may provide a better accuracy in some cases (lots of classes, usually).

Input Ports

The model data of the trained classifier.
Input with text documents to categorize.

Output Ports

Documents with assigned categories based on the supplied model.

Best Friends (Incoming)

Best Friends (Outgoing)

Installation

To use this node in KNIME, install Palladian for KNIME from the following update site:

KNIME 4.0
Wait a sec! You want to explore and install nodes even faster? We highly recommend our NodePit for KNIME extension for your KNIME Analytics Platform.

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.