Article Per Year (5 Year)
p-Index From 2021 - 2026
0
P-Index
This Author published in this journals
All Journal International Journal of Artificial Intelligence Research
Darwish, Saad Mohamed
Unknown Affiliation
Computer Science & IT Electrical & Electronics Engineering

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Bio-inspired Expert System based on Genetic Algorithm for Printer Identification in Forensic Science Darwish, Saad Mohamed; ELgohary, Hany M
International Journal of Artificial Intelligence Research Vol 2, No 2 (2018): December 2018
Publisher : Universitas Dharma Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (777.591 KB) | DOI: 10.29099/ijair.v2i2.67

Abstract

Printer identification models are provided for the goal of distinguishing the printer that produced a suspicious imprinted document. Source identification of a published document can easily be a significant procedure intended for the forensic science. The arising problem is that the extraction of many features of the printed document for printer identification sometimes increases time and reduces the classification accuracy since a lot of the document features may come to be repetitive and non-beneficial. Distinct combinatorial collection of features will need to be acquired in order to preserve the most effective fusion to accomplish the maximum accuracy. This paper presents an intelligent machine learning algorithm for printer identification that adopts both of texture features formulated from gray level co-occurrence matrix of the printed letter ''WOO'' and genetic heuristic search to select the optimal reduced feature set. This integration aims to achieve high classification accuracy based on small group of discriminative features. For classification, the system utilizes k-nearest neighbors (KNN) to recognize the source model of the printer for its simplicity. Experimental results validate that the suggested system has high taxonomy accuracy and requires less computation time.
Quantum Inspired Genetic Programming Model to Predict Toxicity Degree for Chemical Compounds Darwish, Saad Mohamed
International Journal of Artificial Intelligence Research Vol 3, No 1 (2019): June 2019
Publisher : Universitas Dharma Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (27.314 KB) | DOI: 10.29099/ijair.v2i2.64

Abstract

Cheminformatics plays a vital role to maintain a large amount of chemical data. A reliable prediction of toxic effects of chemicals in living systems is highly desirable in domains such as cosmetics, drug design, food safety, and manufacturing chemical compounds. Toxicity prediction topic requires several new approaches for knowledge discovery from data to paradigm composite associations between the modules of the chemical compound; such techniques need more computational cost as the number of chemical compounds increases. State-of-the-art prediction methods such as neural network and multi-layer regression that requires either tuning parameters or complex transformations of predictor or outcome variables are not achieving high accuracy results.  This paper proposes a Quantum Inspired Genetic Programming “QIGP” model to improve the prediction accuracy. Genetic Programming is utilized to give a linear equation for calculating toxicity degree more accurately. Quantum computing is employed to improve the selection of the best-of-run individuals and handles parsimony pressure to reduce the complexity of the solutions. The results of the internal validation analysis indicated that the QIGP model has the better goodness of fit statistics and significantly outperforms the Neural Network model.
Distance and Fuzzy Classifiers Alliance: The Solution to Off-line Arabic Signature Verification System for Forensic Science Darwish, Saad Mohamed; Noori, Zainab H
International Journal of Artificial Intelligence Research Vol 2, No 2 (2018): December 2018
Publisher : Universitas Dharma Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1258.98 KB) | DOI: 10.29099/ijair.v2i2.66

Abstract

Signature of a person is one of the most popular and legally accepted behavioral biometrics that provides secure means for verification and personal identification in many applications such as financial, commercial and legal transactions. The objective of the signature verification system is to classify between genuine and forgery that is often associated with intrapersonal and interpersonal variability. Unlike other languages, Arabic has unique features; it contains diacritics, ligatures, and overlapping.  Because of lacking any form of dynamic information during the Arabic signature writing process, it will be more difficult to obtain higher verification accuracy. This paper addresses the above difficulty by introducing a novel Off-Line Arabic signature verification algorithm. Different from state-of-the-art works that adopt one-level of verification or multiple classifiers based on statistical learning theory; this work employs two-level of fuzzy set related verification. The level one verification depends on finding the total difference between the features extracted from the test signature and the mean values of each corresponding features in the training signatures (owning the same signature). Whereas, the level two verification relies on the output of the fuzzy logic module depending on the membership functions that has been created from the signature features in the training dataset for a specific signer. It is concluded from the experimental results that the verification system performs well and has the ability to reduce both False Acceptance Rate (FAR) and False Rejection Rate (FRR).
Design and Analysis of an Intelligent Integrity Checking Watermarking Scheme for Ubiquitous Database Access Darwish, Saad Mohamed; Selim, Hosam A.
International Journal of Artificial Intelligence Research Vol 3, No 1 (2019): June 2019
Publisher : Universitas Dharma Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (26.216 KB) | DOI: 10.29099/ijair.v3i1.65

Abstract

As a result of the highly distributed nature of ubiquitous database accessing, it is essential to develop security mechanisms that lend themselves well to the delicate properties of outsourcing databases integrity and copyright protection. Researchers have begun to study how watermarking computing can make ubiquitous databases accessing more confident work environments. One area where database context may help is in supporting content integrity. Initially, most of the research effort in this field was depending on distortion based watermark while the few remaining studies concentrated on distortion-free. But there are many disadvantages in previous studies; most notably some rely on adding watermark as an extra attributes or tuples, which increase the size of the database. Other techniques such as permutation and abstract interpretation framework require much effort to verify the watermark. The idea of this research is to adapt an optimized distortion free watermarking based on fake tuples that are embedded into a separate file not within the database to validate the content integrity for ubiquitous database accessing. The proposed system utilizes the GA, which boils down its role to create the values of the fake tuples as watermark to be the closest to real values. So that it's very hard to any attacker to guess the watermark. The proposed technique achieves more imperceptibility and security. Experimental outcomes confirm that the proposed algorithm is feasible, effective and robust against a large number of attacks.
Co-Authors ELgohary, Hany M Noori, Zainab H Selim, Hosam A.