Filter by type:

Sort by year:

Development of an imputation technique - INI for software metric database with incomplete data

Conference Paper
Olanrewaju R. F , Ito W
Software metrics are numerical data that provides a quantitative basis for the development and validation of models, and effective measurement of the software development process. Gathering software engineering data can be expensive. Such precious and costly data cannot afford to be missing. However missing data is a common problem and software engineering database is not an exception. Though there are many algorithms to solve problem of incomplete data, unfortunately few have been developed in the field of Software Engineering. Missing data causes significant problem. With inaccurate data or missing data, it is very difficult to know how much a project will cost or worth. Missing data leads to loss of information, causes biasness in data analysis and hence results to inaccurate decision-making for project management and implementation. In this paper, an imputation technique for imputing missing data based on global-local Modified Singular Value Decomposition (MSVD) algorithm, INI was proposed. This technique was used for estimating missing data in a software engineering database (PROMISE). Its performance was evaluated and compared with two existing imputation techniques, Expectation Maximization (EM) and Mean Imputation (MI). Varying percentages of missings, (1%, 10%, 15%, and 20% 25%) were introduced in the original dataset in order to have an incomplete dataset for imputation. Simulations were carried for comparative purposes. Imputation Error (IE) was use as an evaluation criterion. Imputation Error (IE) was use as an evaluation criterion. Study results showed that, the only method that consistently outperformed other methods (EM and MI), guarantee a higher accuracy of imputed data, prompt and less bias at all level of missings is the global-local MSVD, INI. It maintained consistency at all level of missings compared to EM and MI. It was found that EM is not suitable for data with missing proportion greater than 20%. While MI lost in all count to EM and INI.

State-Of-The-Art Application Of Artificial Neural Network In Digital Watermarking And The Way Forward

Conference PaperJournal
Rashidah Funke Olanrewaju, Othman O. Khalifa, Andre Abdalla

Several high-ranking watermarking schemes using neural networks have been proposed in order to make the watermark stronger to resist attacks. The ability of Artificial Neural Network, ANN to learn, do mapping, classify, and adapt has increased the interest of researcher in application of different types ANN in watermarking. In this paper, ANN based approached have been categorized based on their application to different components of watermarking such as; capacity estimate, watermark embedding, recovery of watermark and error rate detection. We propose a new component of water marking, Secure Region, SR in which, ANN can be used to identify such region within the estimated capacity. Hence an attack-proof watermarking system can be achieved.

Mitigating Storage Security Threats in Mobile Phones

Journal
Sabahat Hussain, Rashidah F. Olanrewaju and Ahmad Fadzil Bin Ismail
Indian Journal of Science and Technology, Vol 11(16), DOI: 10.17485/ijst/2018/v11i16/121410, April 2018

This paper discusses the proposed system which provides the user the ability to run the application on Android phones for encrypting all types of files before they are stored. Methods/Statistical Analysis: In the proposed system, Advanced Encryption Standard (AES) is employed for encryption as well as decryption of the files stored by users in mobile phones. All types of files including docx, text, pdf, image, ppt, audio and video files can be encrypted and later regenerated as well. Findings: The use of AES in encrypting and decrypting data in this system provides good security as well as higher speed. An added advantage is that it is implementable on several platforms particularly in small gadgets like smartphones. As compared to the traditional computers, smartphones can be carried more easily and provide similar functionalities as that of a computer like data storage and processing, communication and other services including video call, wireless network, web browser, GPS, audio or video player. Application/Improvements: Data is transmitted, shared and stored for various purposes including production, banking, development, and research. Therefore, security is a must for information which can be provided by encryption using AES as proposed in this system.

Secure Annihilation of Out-of-Band Authorization for Online Transactions

Journal
Sabahat Hussain , Burhan Ul Islam Khan *, Farhat Anwar , Rashidah Funke Olanrewaju
10.17485/ijst/2018/v11i5/121107

In this paper, an approach to online banking authorization using one-time passwords has been illustrated. Methods/Statistical Analysis: The algorithm presented in this paper provides an infinite as well as forward One-Time- Password (OTP) generation mechanism employing two Secure Hash Algorithms viz. SHA3 and SHA2, followed by dynamic truncation to produce human-readable OTP. An inimitable authentication scheme has been presented in which a unique initial seed is used for generating a series of OTPs on the user’s handheld gadget (i.e. a mobile phone). Findings: The proposed scheme demonstrated better results than the previous schemes of authorization after a security analysis was conducted on it. This is attributed to the eradication of cellular network within the authorization process. A high level of performance and efficiency in authentication and authorization was evident from the results that are vital for transacting online. Applications/Improvements: In the proposed system, the inherent features of the user’s device (mobile phone) are utilized to form the initial seed. The application of hash functions to that seed eliminates the necessity to send one time passwords to the users via Short Message Service and decreases the limitations posed by out-of-band systems, thus making it suitable to be employed in online banking and other financial organizations.

Snort-Based Smart and Swift Intrusion Detection System

Journal
Rashidah Funke Olanrewaju , Burhan Ul Islam Khan *, Athaur Rahman Najeeb , Ku Nor Afiza Ku Zahir , Sabahat Hussain
10.17485/ijst/2018/v11i4/120917

In this paper, a smart Intrusion Detection System (IDS) has been proposed that detects network attacks in less time after monitoring incoming traffic thus maintaining better performance. Methods/Statistical Analysis: The features are extracted using back-propagation algorithm. Then, only these relevant features are trained with the help of multi-layer perceptron supervised neural network. The simulation is performed using MATLAB. Findings: The proposed system has been verified to have high accuracy rate, high sensitivity as well as a reduction in false positive rate. Besides, the intrusions have been classified into four categories as Denial-of-Service (DoS), User-to-root (U2R), Remote-to-Local (R2L) and Probe attacks; and the alerts are stored and shared via a central log. Thus, the unknown attacks detected by other Intrusion Detection Systems can be sensed by any IDS in the network thereby reducing computational cost as well as enhancing the overall detection rate. Applications/Improvements: The proposed system does not waste time by considering and analysing all the features but takes into consideration only relevant ones for the specific attack and supervised learning neural network is used for intrusion detection. By the application of Snort before backpropagation algorithm, the latter has only one function to perform – detection of unknown attacks. In this way, the time for attack detection is reduced.

Intelligent Cooperative Adaptive Weight Ranking Policy via dynamic aging based on NB and J48 classifiers

Journal
Dua’a Mahmoud Al-Qudah, Rashidah Funke Olanrewaju, Amelia Wong Azman
10.11591/ijeei.v5i4.362

The increased usage of World Wide Web leads to increase in network traffic and create a bottleneck over the internet performance.  For most people, the accessing speed or the response time is the most critical factor when using the internet. Reducing response time was done by using web proxy cache technique that storing a copy of pages between client and server sides. If requested pages are cached in the proxy, there is no need to access the server. But, the cache size is limited, so cache replacement algorithms are used to remove pages from the cache when it is full. On the other hand, the conventional algorithms for replacement such as Least Recently Use (LRU), First in First Out (FIFO), Least Frequently Use (LFU), Randomised Policy, etc. may discard essential pages just before use. Furthermore, using conventional algorithms cannot be well optimized since it requires some decision to evict intelligently before a page is replaced. Hence, this paper proposes an integration of Adaptive Weight Ranking Policy (AWRP) with intelligent classifiers (NB-AWRP-DA and J48-AWRP-DA) via dynamic aging factor.  To enhance classifiers power of prediction before integrating them with AWRP, particle swarm optimization (PSO) automated wrapper feature selection methods are used to choose the best subset of features that are relevant and influence classifiers prediction accuracy.   Experimental Result shows that NB-AWRP-DA enhances the performance of web proxy cache across multi proxy datasets by 4.008%,4.087% and 14.022% over LRU, LFU, and FIFO while, J48-AWRP-DA increases HR by 0.483%, 0.563% and 10.497% over LRU, LFU, and FIFO respectively.  Meanwhile, BHR of NB-AWRP-DA rises by 0.9911%,1.008% and 11.5842% over LRU, LFU, and FIFO respectively while 0.0204%, 0.0379% and 10.6136 for LRU, LFU, FIFO respectively using J48-AWRP-DA.