Automated Amharic News Categorization Using Deep Learning Models

<table class="algorithm-group"><tr><td><table class="algorithm" id="alg1"><tr><td> </td><td><b>Input</b>: Raw document Doc</td></tr><tr><td> </td><td><b>Output</b>: Category of Doc</td></tr><tr><td> </td><td><b>Begin</b></td></tr><tr><td> </td><td> Preprocess document Doc (tokenization, normalization, stop word removal, and stemming)</td></tr><tr><td> </td><td> Create a list of distinct words in document Doc</td></tr><tr><td> </td><td> Use word embedding by transforming Doc into feature vector using fastText model</td></tr><tr><td> </td><td> Classify using CNN model</td></tr><tr><td> </td><td><b>End</b></td></tr></table></td></tr></table>

<div> The framework for the proposed system.</div>

Computational Intelligence and Neuroscience

alg1

Algorithm 1

Algorithm 1: Automated Amharic News Categorization Using Deep Learning Models 