Combining Arabic Nested Noun Compound and Collocation Extraction Using Linguistic and Statistical Approach
Maryam Yaseen Al-Mashhadani, Luma Adnan Al-Sagban · IOSR Journal of Computer Engineering · 2016
Arabic multi-word expressions are the combinations of two or more terms that associated with each other as one concept.The process of extracting such expressions is challenging especially when the length of the combination is getting longer.Recently, researchers attempted to extract nested noun compounds which consists of two or more combinations of nouns.Such extraction process requires comprehensive analysis using linguistic and statistical approaches.However, the process of extraction in the state of the art have extended to include 4-gra and 5-gram candidates.This paper aims to combine the extraction of nested noun compound and collocation in order to extend the process of extraction to include 6-gram and 7-gram.For this manner, a linguistic approach comprises of various kinds of pattern has been used, as well as, three statistical measures have been utilized including NC-value, LLR and PMI.Results shown that the proposed method has the ability to extend the extraction to include longer candidates.