A survey of automatic Arabic diacritization techniques

Abstract

In Modern Standard Arabic texts are typically written without diacritical markings. The diacritics are important to clarify the sense and meaning of words. Lack of these markings may lead to ambiguity even for the natives. Often the natives successfully disambiguate the meaning through the context; however, many Arabic applications, such as machine translation, text-to-speech, and information retrieval, are vulnerable due to lack of diacritics. The process of automatically restoring diacritical marks is called diacritization or diacritic restoration. In this paper we discuss the properties of the Arabic language and the issues that are related to the lack of the diacritical marking. It will be followed by a survey of the recent algorithms that were developed to solve the diacritization problem. We also look into the future trend for researchers working in this area.

A Hybrid Approach for the Morpho-Lexical Disambiguation of Arabic

“…The results presented in this paper open up many avenues for future research that focuses on the use of Alkhalil+. Azmi and Almajed [14] felt that hybrid-based schemes are a better choice, as they combine human knowledge with some intelligent techniques.…”

“…We completed our study by joining the opinions of Nelkel and Shieber [24] and Azmi and Almajed [14]: "the objective is to use enough of the optimal number of diacritical marks to avoid ambiguity in the written text. "…”

“…More recently, there have been several statistical, machine-learning approaches, such as the hidden Markov model (HMM), n-gram, statistical machine translation (SMT), and finite state transducers (FSTs). And the most current work in the area relies on hybrid approaches that combine rule-based and statistical modules [14]. Also, several systems 3 and tools have been developed for the resolution of the ambiguity for different levels of the analysis related to automatic diacritization for works such as [15][16][17][18][19][20][21][22].…”

Verification System for Quran Recitation Recordings

“…Holly Quran consists of 6236 verses. Each Sorra also differs from one another in the number of its verses [31].…”

[Editorial] Advanced Technologies and Applications for Security and Multimedia Computing

“…This disambiguation approach, which is based on the diacritics at the various analysis levels, was used to combine a linguistic approach with a multi-criteria decision, and the authors used both the Nelkel and Shieber [4] and Azmi and Almajed [5] algorithms to get the optimal number of diacritical marks.…”