2011 · 34 citations · 23 references
ComputingArabicComputer ScienceArtificial IntelligenceMachine LearningNatural Language ProcessingLetter RecognitionPattern RecognitionArtificial Immune SystemsCharacter ClassificationGray-level Co-occurrence MatricesHandwritten Arabic Recognition
Artificial Immune Systems (AIS) is an emerging bio-inspired computer science technique which embody the principles of biological immune systems for tackling complex real-world problems such as pattern recognition. Among the several immune-computing models, Artificial Immune Recognition System (AIRS) is one of the widely used for classification problems. Meanwhile, the issues related to writer identification are currently at the heart of numerous concerns in our modern day's society. Writer identification for Arabic text is receiving a renewed attention. Many popular machine learning techniques have been used in writer identification systems but only one limited attempt has been done with AIS. In this paper, we apply AIRS to perform Arabic writer identification based on a set of features extracted from Grey Level Co-occurrence Matrices. Some feature selection techniques are applied to improve computation time and accuracy results. Three traditional classifiers have also been used in our experiments for performance comparison. The obtained results show the promising ability of AIRS in Writer identification.
23
IFN/ENIT: database of handwritten arabic words
Mario Pechwitz, Samia Snoussi Maddouri, Volker Märgner et al. · 2002 · 428 citations
Personal identification based on handwriting
Huwida Said, Tieniu Tan, K.D. Baker · Pattern Recognition · 2000 · 336 citations
Immune system approaches to intrusion detection – a review
Natural Computing · 2007 · 201 citations · Full text