Asian Journal of Research in Computer Science

The use of touchscreens as an input method on Smartphones has become the norm in the mobile phone industry. This has changed the way keys are traditionally arranged on Smartphones devices. One of these changes is on the location of “back button”. Few studies, However, have been conducted to investigate the effect of key locations on users performance and experience.

Aims: In this paper, we investigate the effect of back button location on users’ experience of using Smartphones. We made a  comparison between Android Samsung Galaxy S3 and iPhone 5s.

Place and Duration of Study: Participants: Sokoto State University, Sokoto, Nigeria.  between May 2017 and July 2017.

Methodology: A total of 40 participants all Students of Sokoto State University participated in the study. 26 were males and 14 females. 30 owned Android Smartphones while the remaining  10 owned Apple iPhone. We used observation to observe how users navigate through Facebook and Gmail app on both iPhone and Android while paying attention to their use of the back button, the time it takes to locate these buttons and first point of reference. A semi-structured interview was also conducted on users, requesting them to compare how the difference in the location of the back button on iPhone and Android has affected their experience of navigating through both  Facebook and Gmail app.

Results: The study reveals that back button key location affects the user experience of using Smartphones, keys located at the bottom edges are easier to locate and those located at top edges are harder to locate and press.

Conclusion: Based on the results, the study concludes that designers of Smartphones user interfaces should strongly consider user preferences when deciding the location for back buttons and on Smartphones.

Smith-Waterman Algorithms (SWA) is becoming popular among researchers especially in the field of bioinformatics. The algorithm performance is better among other known alignment algorithms because of the high level of accuracy it exhibits. However, the algorithm performance is at low speed due to its computational complexity. Researchers are concerned with this problem and are looking for various ways to address the issue. Different approaches are adopted to improve the speed, such as the use of a systolic array to accelerate the algorithm, use of recursive variable expansion (RVE) method approach; some implemented the algorithm on software and hardware, etc. This paper used Residue Number System (RNS) approach to the algorithm of Smith-Waterman and carried out hardware implementation on Quartus II, 64-Bit version 12.1 (Cyclone II family) VHDL application software.

A new speech corpus of connected Bangla words derived from newspapers text corpus BdNC01 has recorded. This has been designed for various research activities related to speaker-independent Bangla speech recognition. The database consists of speech of 100 speakers, each of them uttered 52 sentences as connected words for training database. Another 50 new speakers were employed to speak all the list of speech to construct a test database. Every utterance was repeated 5 times in various days to avoid time variation of speaker property. A total of 62 hours of recording makes the corpus largest in its type, size and application area. This paper describes the motivation for the corpus and the processes undertaken in its construction. The paper concludes with the usability of the corpus.

In this work, we investigate the viability of the stacked generalization approach in predictive modeling of a direct marketing problem. We compare the performance of individual models created using different classification algorithms, and stacked ensembles of these models. The base algorithms we investigate and use to create stacked models are Neural Networks, Logistic Regression, Support Vector Machines (SVM), Naïve Bayes and Decision Tree (CART). These algorithms were selected for their popularity and good performance on similar tasks in previous studies. Using a benchmark experiment and statistical tests, we compared five single algorithm classifiers and 26 stacked ensembles of combinations these algorithms on two popular metrics: Area Under ROC Curve (AUC) and lift.  We will demonstrate a significant improvement in the AUC and lift values when the stacked generalization approach is used viz a viz the single-algorithm approach. We conclude that despite its relative obscurity in marketing applications, stacking holds great promise as an ensembling technique for direct marketing problems.

Botnet as it is popularly called became fashionable in recent times owing to it embedded force on network servers. Botnet has an exponential growth of about 170, 000 within network server and client infrastructures per day. The networking environment on monthly basis battle over 5 million bots. Nigeria as a country loses above one hundred and twenty five (N125) billion naira to network fraud annually, end users such as Banks and other financial institutions battle daily the botnet threats. The most worrisome part of the botmaster’s botnet is it propagation as an entity even when it is known to be large pool of malicious threats. The attacks leave end users (clients) to the risk of losing valuable credentials when connected to the affected infrastructure. It is on the above premise that this paper sort to expose the botnet method of propagation through proactive mechanism called Encapsulated Detection Mechanism (EDM) for botnet on Server Systems with further operations on conceptual framework, structural modules, usability and application of botnet. The mechanism uses one dimensional data stream evolutionary window approach of Distance Base Model (DBM) as an Outlier Analysis (OA). The Captcha, Username password and EDM Analyzer act as the front end of the data stream checker using Bot-Stream OutlieR Miner (B-STORM) algorithm and B-Exact Algorithm. The research work showed high level of data entering compliance efficiency on the server end network by neutralizing and mitigating botnet attack that falls short of the predefined data order within the networking signature.