International Journal of Digital Enterprise Technology (3 papers in press)
Solving Problems on a Knowledge model of Operators and Application
by Hien Nguyen, Nhon Do, Vuong Pham, Katsumi Inoue
Abstract: Knowledge of operators is a popular form of knowledge domains, especially computational knowledge domains. This knowledge domain is useful to build the intelligent problem solver for knowledge domains about mathematics and physics. There are many methods for representing this knowledge domain, such as: rule-based systems, computational networks, ontology. However, these current methods are not effective for representation practical applications. In this paper, we present a mathematical approach for building a knowledge model of operators, called Ops-model. The foundation of this model includes: concepts, operators, and inference rules. Each concepts of this model is a class of objects with the behaviours for solving problems on themselves. This model refers to both unary and binary operators and their properties: commutation, association, identity. Based on the structure of Ops-model, general problems on this model are studied, such as: Reducing an expression, Prove an equality of expressions. The algorithms for solving these problems are designed. They are also proved their effectiveness. Moreover, Ops-model has been applied to specify a part of knowledge domain about Vector Algebra in high school. It is used for constructing a program for solving some problems on this knowledge domain. The solutions of this program are step-by-step, readable and suitable with the learners level. This program is also tested and evaluated by the high-school students. It is useful for supporting students to learn this subject.
Keywords: knowledge representation; intelligent problem solver; automated reasoning; knowledge engineering.
A Forensic Evidence Recovery from Mobile Device Applications
by John K. Alhassan, Agbejule Gbolahan, Ismaila Idris, Shafi’i Muhammad Abdulhamid, Victor O. Waziri
Abstract: In recent past, there is a lot of research advancements in mobile forensics tools. This is so due to increase usage of mobile phones in storage of companys information, law enforcement, mobile online transactions, and also negatively by criminals due to increased computational capabilities. Mobile forensics device continues to remain a difficult task due to poor user data retrieval techniques for evidence and procedural resolutions. Recently, third party applications assume a veritable feet because its supported by majority of mobile devices platforms; thereby making it easy to extract information of its users for future criminal audit. This paper proposes an evidence data retrieval method from InstagramApp using two networks based platforms (that is, pure peer-to-peer (PPP) and special cluster peer (SCP) based),whose concept is to manage mobile device communication and generate multiple copies of users data/information to be dumped across three servers.The forensic test results evaluation was carried out for these forensic models; PPP and SCP developed to securely extract data from mobile devices revealed in either cases considered that, SCP outperformed PPP in terms of the time taken to fulfill forensic auditors requests, throughput and broadband utilization were 42.82% to 57.18%, 56.81% to 43.19% and 35.41% to 64.53% respectively.
Keywords: Mobile Forensic; Evidence Recovery; Pure Peer-to-Peer; Special Cluster Peer; Mobile Device.
Modeling and Time Evaluation of Optical Disc and Retinal Lesions
by Jan Kubicek, Marek Penhaker, Martin Augustynek, Jakub Slonka, Veronika Kovarova
Abstract: In the clinical ophthalmology, the retinal image analysis is a routine procedure with a target of the time considering the retinal lesions over the time. Unfortunately, there is an absence of the clinical software instruments providing a precise tracking the retinal lesions from the image records. Thus, the retinal image analysis is carried out by skilled physicians, but without any objective software feedback. We aim to propose a procedure having ambitions to an automatic and autonomous extraction of the retinal lesion from the retinal images, and its time evaluation. The retinal lesions are considered on a base of the optical disc which is simultaneously segmented from the retinal images. It is clinically supposed that the optical disc has stable geometrical features over the time contrarily geometrical features of the retinal lesions are time-developed. The proposed methodology for a time modeling of the retinal lesions comprises three essential procedures. The optical disc and the retinal lesions segmentation on a base of the time evolving curves ensure indication areas of these retinal objects. Consequently, the binary classification is used with a target of an extraction of a respective model of the optical disc and the retinal lesions.
Keywords: Retinal lesions; optical disc; modeling; active snake model; binarization; retinal image; RetCam 3; Fundus camera; image processing; image segmentation; geometrical features; initial contour.