Contribution to Semantic Analysis of Arabic Language

Abstract

We propose a new approach for determining the adequate sense of Arabic words. For that, we propose an algorithm based on information retrieval measures to identify the context of use that is the closest to the sentence containing the word to be disambiguated. The contexts of use represent a set of sentences that indicates a particular sense of the ambiguous word. These contexts are generated using the words that define the senses of the ambiguous words, the exact string-matching algorithm, and the corpus. We use the measures employed in the domain of information retrieval, Harman, Croft, and Okapi combined to the Lesk algorithm, to assign the correct sense of those proposed.

NGram Approach for Semantic Similarity on Arabic Short Text

“…However, such measures are rarely used to compare multi-word phrases [4]. Three broad categories of semantic similarity detection methods are used to determine the level of similarity between words: Dictionary/ontology-based methods consider knowledge bases to gather the semantic information that is compared when determining semantic similarity [14]. Meanwhile, corpus-based methods primarily use word frequencies to determine semantic similarity.…”

“…This is done based on statistics taken from extensive corpora. Finally, hybrid methods consider more than one information source to determine semantic similarity [14].…”

“…The statistical methods employed by corpus-based approaches have recently evolved. Thus, such approaches can follow one of two principal orientations [14]:…”

“…• The second orientation includes supervised and semisupervised approaches, by which an annotated training corpus with the appropriate classification models is applied. Supervised methods include probabilistic methods, and they typically employ the naive Bayes algorithm and follow the maximum entropy approach [14]. Which methods are followed when using such an approach depends on how similar the evaluated examples are.…”

Arabic Sentiment Classification Using Convolutional Neural Network and Differential Evolution Algorithm

“…Arabic is a Semitic language known by its morphology richness and different written and spoken forms such as modern standard Arabic (MSA) and its various dialects. Arabic morphology and structure complexity create many challenges such as the shortage of large datasets and limited tools to perform sentiment analysis [ 3 , 4 ]. Even deep neural networks (DNNs) [ 5 ] and convolutional neural networks (CNNs) [ 6 ] have shown promising and encouraging performance, little research on sentiment analysis using deep learning (DL) techniques has been done for Arabic language [ 7 – 9 ] while many researches have been done on other languages [ 10 – 12 ].…”