Abstract:
Demonstrative pronouns are key features in creating coherent texts in any language in general
and in Amharic in particular. Their presence signifies something in the text. Nevertheless, no
resolution model has been introduced to understand and analyze the grammatical and
functional features of these demonstrative pronouns. This work is therefore intended to design
anaphora resolution model for Amharic demonstrative pronouns. To this end, a machine
learning approaches (Logistic Regression, Random Forest Classifier, and Support Vector
Machine) are employed. We take corpora from the Holy Bible and Walta News indexed in
Walta Information Center. We use Python programming language as the main tool for
developing the prototype for anaphora resolution as applied to demonstrative pronouns in
Amharic. The accuracy of the model is tested for three algorithms.
Key words: anaphora resolution, Amharic demonstrative pronouns, Machine learning