الفهرس 
An enhanced approach for hierarchical Arabic text classification and keyphrase extraction
ABSTRACTOnly 14 pages are availabe for public view
 |  | from | 148 |  |  |
| | | from | 148 | | |
Abstract
Multi-label classification (MLC) is concerned with learning from examples where each document is associated with a set of labels (categories) in opposite to traditional single-label. Classification where an example or document typically is assigned a single label (Category). MLC problems appear in many areas, including text Classification and categorization, protein function classification, and multimedia semantic Annotation. The religious domain has become an interesting and challenging area for machine learning and natural language processing. A 2fatwa3 in the Islamic religion represents the legal opinion or interpretation that a qualified scholar (mufti) can give on issues or case related to the Islamic law. It is similar to the issue of legal opinions from courts in common-law systems. In This research, a multi-Label hierarchical classification system is introduced to automatically route incoming fatwa requests to the most relevant mufti. Each fatwa is associated with multiple categories by mufti where the categories can be organized in a hierarchy. The results of fatwa requests routing have confirmed the effective and efficient predictiveperformance of hierarchical ensembles of multi-label classifierstrained using the HOMER method and its variations comparedto binary relevance, which simply trains a classifier for eachlabel independently. This research also aKey Phrase Extraction and title generation system is introduced to automatically generate Key-Phrase that represent fatwa requests Idea and main topic depending on the relevant category. The key phrase generation depends on the fatwa category (class). Each fatwa class hasa lexicon of words (feature vector). Eachword contributesin feature vector that represent theclass by percentage. We take class that fatwa requests relevant to generate the key phrase by an enhancedHybrid Approach for Arabic Text Key Phrase Extraction. Both results in the proposed model for text classificationreferred to a high degree accuracy of the technique (HOMER) in the multi label classification,where the fatwa classification achieve 78% Micro-averaged F-Measure,76.5 %Micro-averaged Precision, 77% Micro-averaged Recall,in determining thefatwa class. Also theresults in the proposed model for Key Phrase extraction referred to a high degree accuracy,where KP extraction achieves87.32%Accuracy, in determining thefatwa key phrases