What question did this study set out to answer?

To develop a nonparametric algorithm for hierarchical partitioning of feature space using mutual information.

July 1, 1982IEEE Transactions on Pattern Analysis and Machine Intelligence

Hierarchical Classifier Design Using Mutual Information

View Full Paper

Ask AI

Bookmark

To develop a nonparametric algorithm for hierarchical partitioning of feature space using mutual information.
Developed a nonparametric algorithm based on average mutual information for hierarchical partitioning.
Generated an efficient partitioning tree to minimize the probability of error.
Demonstrated the algorithm with examples including handprinted numeral recognition.
The algorithm effectively maximized average mutual information gain at each partitioning step.
Generated a confidence bound expression for the classifier's performance.
Showed practical effectiveness in multifeature multicategory pattern recognition problems, including numeral recognition.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Randomized trial evaluates a mutual information-based algorithm for hierarchical feature space partitioning in pattern recognition, indicating its effectiveness.

Cite This Study

Sethi et al. (1982) studied this question.

synapsesocial.com/papers/6a07edc47ad161a3abfe0a51 https://doi.org/10.1109/tpami.1982.4767278

View Full Paper

Ask AI

Bookmark

Implication

Randomized trial evaluates a mutual information-based algorithm for hierarchical feature space partitioning in pattern recognition, indicating its effectiveness.

Hierarchical Classifier Design Using Mutual Information

Key Points

Discussion

Implication

Cite This Study

Hierarchical Classifier Design Using Mutual Information

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider