International Journal of Information Technology and Computer Science (IJITCS)

IJITCS Vol. 10, No. 10, Oct. 2018

Cover page and Table of Contents: PDF (size: 190KB)

Table Of Contents

REGULAR PAPERS

Homogeneous Densities Clustering Algorithm

By Ahmed Fahim

DOI: https://doi.org/10.5815/ijitcs.2018.10.01, Pub. Date: 8 Oct. 2018

Clustering based-density is very attractive research area in data clustering. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is the pioneer in this area. It can handle varied shaped and sized clusters. Also, it copes with noise and outliers efficiently, however it fails to handle clusters with varied densities due to the global parameter Eps. In this paper, we propose a method overcomes this problem, this method does not allow large variation in density within a cluster and use only two input parameters that will be called minpts and maxpts. They govern the minimum and maximum density of core objects within a cluster. The maxpts parameter will be used to control the value of Eps (neighborhood radius) in original DBSCAN. By allowing Eps to be varied from one cluster to another based on density of region this make DBSCAN able to handle varied density clusters and discover homogeneous clusters. The experimental results reflect the efficiency of the proposed method despite its simplicity.

[...] Read more.
Authorship Attribution for Bengali Language Using the Fusion of N-Gram and Naive Bayes Algorithms

By D. M. Anisuzzaman Abdus Salam

DOI: https://doi.org/10.5815/ijitcs.2018.10.02, Pub. Date: 8 Oct. 2018

This research shows the authorship attribution for three Bengali writers using both Naïve Bayes method and a new method proposed by us which performs better than Naïve Bayes for authorship attribution. Though a lot of works exist in the field of authorship attribution for other languages (especially English); the amount of work in this field for Bengali language is very low. For this experiment, we make our own dataset having 107380 words and 21198 unique words. For both methods, we pre-process our dataset to be compatible to work with the method experiments. For our dataset, Naïve Bayes gives an accuracy of 86% while our method gives an accuracy of 95%. The main inspiration behind our method is that every author has a nature to write some adjacent words and some single words repeatedly.

[...] Read more.
Text Summarization System Using Myanmar Verb Frame Resources

By May Thu Naing Aye Thida

DOI: https://doi.org/10.5815/ijitcs.2018.10.03, Pub. Date: 8 Oct. 2018

In today’s era, when the size of information and data is increasing exponentially, there is an upcoming need to create a concise version of the information available. Until now, humans have tried to create “summaries” of the documents. Especially, Myanmar Natural Language Processing does not have computerized text summarization. Therefore, this paper presents a summary generation system that will accept a single document as input in Myanmar. In addition, this work presents analysis on the influence of the semantic roles in summary generation. The proposed text summarization system involves three steps: first, the sentences are parsed using Part of Speech tagger with Myanmar Language Tool Knowledge Resource (ML2KR); secondly, pronouns in the original text are resolved using Myanmar Pronoun Resolution Algorithm (MPAR); thirdly, the sentences are labeled with semantic roles using Myanmar Verb Frame Resource (MVF), finally, extraction of the sentences containing specific semantic roles for the most relevant entities in text. After that, the system abstracts the important information in fewer words from extraction summary from single documents.

[...] Read more.
MF-NB Learning Based Approach for Recommendation System

By Hutashan V. Bhagat Shashi B. Sachin M. Baad

DOI: https://doi.org/10.5815/ijitcs.2018.10.04, Pub. Date: 8 Oct. 2018

The Multi Factor-Naive Bayes classifier based recommendation system is analyzed with respect to the traditional KNN classifier based recommendation system. The classification of the web usage data is done on the basis of the keyword name, keyword count, inbound links and age group of the users. Whereas, in traditional KNN the URL was the only factor that was considered for the purpose of classification. The performance evaluation is done in the terms of RMSE, Error Rate, Accuracy Rate and Precision. The MF-NB is observed to be outperforming the KNN classifier in all respective terms.

[...] Read more.
Fredkin Circuit in Nanoscale: A Multilayer Approach

By Md. Abdullah-Al-Shafi Ali Newaz Bahar

DOI: https://doi.org/10.5815/ijitcs.2018.10.05, Pub. Date: 8 Oct. 2018

Nanotechnologies, exceedingly Quantum-dot Cellular Automata (QCA), presents a notable perception for upcoming nanocomputing. Feature extent of circuits is moving to sub-micron point that produces the sophisticated device intricacies. In this work, QCA is considered as an application technique for reversible logic. A multi-layer reversible Fredkin circuit is proposed with QCA nanotechnology. The accomplishment of the outlined circuit is substantiated with five existing Fredkin gate, which exhibits from 71.20% to 37.50% improvement in term of cell intricacy. The proposed design uses 55 cells concerning only 0.03 μm2 area and latency is 0.75. The power consumption by the proposed circuit is also presented in this literature. The proposed design has been realized with QCADesigner version 2.0.3.

[...] Read more.
On the Technologies and Systems for Student Attendance Tracking

By Zoran Kotevski Natasa Blazheska-Tabakovska Andrijana Bocevska Tome Dimovski

DOI: https://doi.org/10.5815/ijitcs.2018.10.06, Pub. Date: 8 Oct. 2018

Manual student attendance tracking, by calling student names from a check list or taking students’ signs on a paper, has gone into history. Nowadays, modern technologies have already enabled the development of various automatized attendance tracking systems. These technologies include: Radio-Frequency Identification (RFID), Biometric (fingerprint, face or voice recognition), Barcode identification and Bluetooth communication technologies, that are implemented over an IP infrastructure as a platform. But, all these technologies perform in a different manner and exhibit certain functional limitations considering the given implementation. The main motivation for this research was to explore the possibilities for overcoming the issues of current systems for attendance tracking, considering the limitations of the technologies employed. Hence, the core contribution of this research can be considered as a fourfold, i.e. i) it presents the most frequently used technologies in the development of attendance tracking systems, ii) it reviews a range of existing student attendance tracking systems, iii) it defines criteria for performance evaluation of the technologies employed in student attendance tracking, from a perspective of educational institutions, and iv) it evaluates the mostly used technologies according to the predefined functional criteria. As a summary of the evaluation it provides directions for future development of a student attendance tracking system that would address the explored issues and functional limitations.

[...] Read more.
Assessing Users Satisfaction and Experience on Academic Websites: A Case of Selected Nigerian Universities Websites

By Arnold A. OJUGO Andrew O. EBOKA

DOI: https://doi.org/10.5815/ijitcs.2018.10.07, Pub. Date: 8 Oct. 2018

Webometric seeks to assess the convenience as well as experience of users, as they interact with web-based system. Businesses use websites as means to present their deliverables to a larger audience. The aim of which is to change, refocus and reshape a user’s image of an organization. Today, it has been extended to facilitate activities such as recruitment etc. We investigate usability of selected Nigerian Universities based on expert review guideline with a focus on the Federal University of Petroleum Resources Effurun (FUPRE), compare the scores achieved in each criterion among selected university websites. Result shows common strengths, common weakness, unique strengths and unique weakness of these university websites. Some websites were found to have unique characteristics which separated them from the other ones. Most university websites successfully ensured that the trust and credibility section as well as their homepages received the highest scores in their usability analysis. Some of the sites on the contrary, struggled to ensure good search usability and form, and data entry criteria, as many of the websites did not even include the mentioned essential functions in their web design. Findings suggest that even the top ranked universities have websites which is lacking in some different website usability areas.

[...] Read more.
RDF Link Generation by Exploring Related Links on the Web of Data

By Kumar Sharma Ujjal Marjit Utpal Biswas

DOI: https://doi.org/10.5815/ijitcs.2018.10.08, Pub. Date: 8 Oct. 2018

Interlinking RDF resources is a vital aspect of the Semantic Web technology. It is the basis of Linked Data that provides interlinked datasets on the web. One of the principles of Linked Data is interlinking resources from different data sources on the web. Data interlinking is a critical and challenging problem that every Linked Data generation applications face. Various approaches have been evolved for resolving this problem, but, for more massive datasets, it becomes almost indefinite time while linking similar or related resources. Linking RDF resources is like the problem of entity matching, record matching or duplicate resource detection. More or less they attempt to point to the same problem, but the RDF link generation is the task of finding related resources on the web. In this article, we present an approach for generating RDF links using the similarity measure between two RDF resources and by exploring associated relationships of the matched resources. The idea is to find related resources and link them with an RDF resource that is being generated.

[...] Read more.