A Hybrid Approach to Pronominal Anaphora Resolution in Arabic

Abstract

One of the challenges in natural language processing is to determine which pronouns to be referred to their intended referents in the discourse. Performing anaphora resolution is considered as an important task for a number of natural language processing applications such as information extraction, question answering and text summarization. Most of the earlier works of anaphora resolution have been applied to English and other languages. However, the work done in Arabic is not sufficiently studied. In this study, a hybrid approach that combines different architectures for resolving pronominal anaphora in Arabic language is presented. The hybrid model adopted the strategy based on the combination of a rule-based and machine learning approach. The collection of anaphora and respective possible antecedents was identified in a rule-based manner with morphological information taken into account. In addition, the selection of the most probable candidate as the antecedent of the anaphor was done by machine learning based on a k-Nearest Neighbor (k-NN) approach. In this study, the appropriate features to be used in this task were determined and their effect on the performance of anaphora resolution was investigated. Experiments of the proposed method were performed using the corpus of the Quran annotated with pronominal anaphora. The experimental results indicate that the proposed hybrid approach is completely reasonable and feasible for Arabic pronominal anaphora resolution.

Neural Coreference Resolution for Arabic

“…Similarly Björkelund and Kuhn (2014) adopted a tree representation approach to cluster mentions, but improved the learning strategy and introduced non-local features to capture more information about coreference relations. There have been other research studies related to anaphora resolution (Trabelsi et al, 2016;Bouzid et al, 2017;Beseiso and Al-Alwani, 2016;Abolohom and Omar, 2015), but they only considered pronominal anaphora. Aloraini and Poesio (2020) also considered a specific type of pronominal anaphora, zero-pronoun anaphora.…”

Anaphora Resolution for Myanmar Text Using K-Nearest Neighbor Algorithm

“…It used Quranic corpus annotated with antecedent references of pronoun and the Arabic Statistical POS Tagger and it considered the search limit of 17 sentences. They presented that the proposed hybrid approach is completely reasonable and feasible for Arabic pronominal anaphora resolution [3]. An anaphora resolution system for Russian language [4] uses the formal rules to obtain a number of features, the Extra Trees machine learning algorithm and the Balance Cascade algorithm for working with imbalanced learning sets and a neural network algorithm SyntaxNet to analyze the syntactic context.…”