The Bidirectional Long-Short-Term Memory Neural Network based Word Retrieval for Arabic Documents
Journal Title: Transactions on Machine Learning and Artificial Intelligence - Year 2015, Vol 3, Issue 1
Abstract
The reflow from Arabic document image collections is a challenging task. This is partly due to the insolubility of the Arabic script. Because of the peculiarity of the whole body of the Arabic words, namely connectivity between the characters, thereby the segmentation of An Arabic word is very arduous and also the variability of the handwritten styles and shapes as well as deceleratation in the print therefore Arabic document repositories are not liable to indexation and reflow. In this paper, we are proposing an idea for reflow coherent Arabic document in response to an appropriate Arabic query word. We present a novel approach at reflow on an Arabic document image, using a Bidirectional Long-Short-Term Memory Neural Network. The designed to take relating to information into account, these networks can maintain Arabic word images that cannot be durable segmented into individual Arabic characters. The partitioning Arabic word, we easier the problem and receive elevated reflow rates. The proposed capable reflow scheme avoids unambiguous recognition of Arabic characters. An experimental evaluation on a dataset of Arabic word images conjunct from handwritten notebook show good precision even in the impendence of printing transformation and deceleratation. The reflow Arabic word performance is comparison with baseline methods. These results encourage the development of real world systems for word reflow for Arabic documents
Authors and Affiliations
Dr. Yusuf Perwej
Keywords
Related Articles
- EP ID EP278943
- DOI 10.14738/tmlai.31.863
- Views 86
- Downloads 0
Evaluation of Tools and Techniques for the Generation of Warning Alerts: A Survey Paper
Quality assurance is a key factor for the improvement of an organizational behaviour. It is quite challenging to enhance an organizational performance without realising internal errors and mistakes done by its employees....
Temperature, Precipitation and Relative Humidity Fluctuation of Makkah Al Mukarramah, Kingdom of Saudi Arabia (1985-2016)
The study presents the temperature, rainfall and relative humidity fluctuation of Makkah Al Mukarramah, Saudi Arabia for a time period of 1985-2016 in terms of general climatology, climate change, seasonal pattern and ex...
Phishing Websites Detection Using Data Mining Classification Model
Phishing is a significant security threat to the Internet; it is an electronic online identity theft in which the attackers use spoofing techniques like fake websites that mimic legal websites to trick users into reveali...
Structural Optimization of Deep Belief Network by Evolutionary Computation Methods including Tabu Search
This paper proposes structural optimization method of a Deep Belief Network (DBN) which consists of multiple Restricted Boltzmann Machines (RBMs) and a single Feedforward Neural Network (FNN) using several kinds of evolu...
Detection of the Onset of Diabetes Mellitus by Bayesian Classifier Based Medical Expert System
Expert systems play an important role in medical diagnosis research. Researches are still being conducted for building expert systems capable of diagnosing different diseases. Diabetes mellitus is one of the diseases tha...