Classifying Amharic webnews

Abstract

We present work aimed at compiling an Amharic corpus from the Web and automatically categorizing the texts. Amharic is the second most spoken Semitic language in the World (after Arabic) and used for countrywide communication in Ethiopia. It is highly inflectional and quite dialectally diversified. We discuss the issues of compiling and annotating a corpus of Amharic news articles from the Web. This corpus was then used in three sets of text classification experiments. Working with a less-researched language highlights a number of practical issues that might otherwise receive less attention or go unnoticed. The purpose of the experiments has not primarily been to develop a cuttingedge text classification system for Amharic, but rather to put the spotlight on some of these issues. The first two sets of experiments investigated the use of Self-Organizing Maps (SOMs) for document classification. Testing on small datasets, we first looked at classifying unseen data into 10 predefined categories of news items, and then at clustering it around query content, when taking 16 queries as class labels. The second set of experiments investigated the effect of operations such as stemming and part-of-speech tagging on text classification performance. We compared three representations while constructing classification models based on bagging of decision trees for the 10 predefined news categories. The best accuracy was achieved using the full text as representation. A representation using only the nouns performed almost equally well, confirming the assumption that most of the information required for distinguishing between various categories actually is contained in the nouns, while stemming did not have much effect on the performance of the classifier.

The impact of indexing approaches on Arabic text classification

“…Extensive research studies, which considered the impact of using different indexing approaches (full-word, stem, and root), exist for English [23–25] and other languages [26–29]. In the case of Arabic, there are research studies on classifying Arabic documents: some have been based only on the stem [30 –32]; others have been based only on the root form [33, 34].…”

Multi-label Emotion Classification on Social Media Comments using Deep learning

“…Ethiopians utilize Amharic, a Semitic language, to communicate across the country. It is the world's second most spoken Semitic language (after Arabic), and one of the top ve on the African continent (Asker et al, 2009). Despite its enormous number of speakers, little computational linguistic resources have been produced for it, and little has been done to make viable higher-level Internet or computer-based applications available to persons who only speak Amharic (Asker et al, 2009).…”