i-manager's Journal on Computer Science (JCOM)


Volume 6 Issue 4 December - February 2019

Research Paper

Enhanced Query Expansion Algorithm: Framework for Effective Ontology Based Information Retrieval System

Enesi Femi Aminu* , Ishaq Oyebisi Oyefolahan**, Muhammad Bashir Abdullahi***, Muhammadu Tajudeen Salaudeen****
* Lecturer,Department of Computer Science, Federal University of Technology, Minna, Nigeria.
** Senior Lecturer,Department of Information and Media Technology, Federal University of Technology, Minna, Nigeria.
*** Head,Department of Computer Science, Federal University of Technology, Minna, Nigeria.
**** Senior Lecturer,Department of Crop Production, Federal University of Technology, Minna, Nigeria.
Aminu,E.F., Oyefolahan,I.O., Abdullahi,M.B., Salaudeen,M.T.(2019) Enhanced Query Expansion Algorithm: Framework for Effective Ontology Based Information Retrieval System,i-manager's Journal on Computer Science, 6(4),1-11. https://doi.org/10.26634/jcom.6.4.15721

Abstract

The strength of an Information Retrieval System lies on its ability to retrieve relevant information or documents according to user's intent by considering a high level of precision and a low level of irrelevant recall of results. A recent development to actualize this dream is the application of ontology. Therefore, Ontology-Based Information Retrieval is becoming an interesting area in the current research trend of ontology and semantic web. However, the sufficiency of developing domain ontology alone to efficiently and effectively take care of information retrieval becomes a research issue. Thus, to address the research gap, a technique called Query Expansion has been identified as a veritable tool. Query Expansion is a process of expanding initial user's query term(s) with the aid of a technology such as wordNet to return relevant results according to user's intent. But returns of query results using the existing wordNet is challenging in normal or inflected terms, such as synonyms or polysemy (word mismatch). Therefore, this paper proposes improved query expansion algorithm as framework to effectively and efficiently develop ontology based information retrieval system.

Research Paper

Comparative Study of Various Machine Learning Algorithms for Tweet Classification

Umar Abubakar* , Sulaimon A. Bashir**, Muhammad Bashir Abdullahi***, Olawale S. Adebayo****
*,**,*** Department of Computer Science, Federal University of Technology, Minna, Nigeria.
**** Department of Cyber Security Science, Federal University of Technology, Minna, Nigeria.
Abubakar, U., Bashir, S. A., Abdullahi, M. B., Adebayo, O. S.(2019 Comparative Study of Various Machine Learning Algorithms for Tweet Classification,i-manager's Journal on Computer Science, 6(4),12-24. https://doi.org/10.26634/jcom.6.4.15722

Abstract

Twitter is a social networking platform that has become popular in recent years. It has become a versatile information dissemination tool used by individuals, businesses, celebrities, and news organizations. It allows users to share messages called tweets with one another. These messages can contain different types of information from personal opinions of users, advertisement of products belonging to all kinds of businesses to the news. Tweets can also contain messages that are racist, bigotry, offensive, and of extremist views as shown by research. Manual identification of such tweets is impossible as hundreds of millions of tweets are posted every day and hence a solution to automate the identification of these types of tweets through classification is required for the Twitter administrators or an intelligence and security analyst. This paper presents a comparative study of traditional machine learning algorithms and deep learning algorithms for the task of tweet classification to detect different categories of abusive languages with the aim to determine which algorithm performs best in detecting abusive language that is prevalent on social media. Two approaches for building feature vectors were explored. Feature vectors based on the bag-of-words method and feature vectors based on word embeddings. These two methods of feature representation were evaluated in this paper using tweet messages representing five abusive language categories. The experiments show that the deep learning algorithms trained with word embeddings outperformed all the other machine learning algorithms that were trained with feature vectors based on the bag-of-words approach.

Research Paper

Development of a Road Surface Condition Monitoring and Database System

H. Bello-Salau* , A. M. Aibinu**, A. J. Onumanyi***, S. Ahunsi, ****, E. N. Onwuka*****, J. J. Dukiya******
* Lecturer,Department of Computer Engineering, Ahmadu Bello University Zaria, Nigera.
** Professor,Department of Mechatronics Engineering, Federal University of Technology, Minna, Niger State.
*** Lecturer,Department of Telecommunication Engineering, Federal University of Technology, Minna, Niger Statee.
**** Graduate,Department of Telecommunication Engineering, Federal University of Technology, Minna, Niger Statee.
***** Professor,Department of Telecommunication Engineering, Federal University of Technology, Minna, Niger Statee.
****** Professor,Department of Transport Management, Federal University of Technology, P.M.B. 65, Minna, Niger State.
Bello-Salau, H., Aibinu, A. M., Onumanyi, A.J., Ahunsi, S. Onwuka E.N., Dukiya,J.J.(2019) Development of a Road Surface Condition Monitoring and Database System,i-manager's Journal on Computer Science, 6(4),25-33. https://doi.org/10.26634/jcom.6.4.15723

Abstract

This paper proposes a road surface condition monitoring device. The design features the use of a programmed accelerometer sensor deployed to respond to vehicular vibrations as a function of the vehicle's acceleration due to gravity (g-force). Furthermore, a database was created and hosted online to store the traces acquired over the different test surfaces. The test results show that the proposed system successfully sensed the utilized road surfaces, and effectively logged the acquired traces into the created database. This device will be beneficial to road maintenance agencies for road surface monitoring, and it can be installed in both manned and unmanned vehicles to enhance road navigation. In addition, the stored traces can be freely accessed and used by researchers working in related areas.

Research Paper

On the Use of Extreme Learning Machines for Detecting Anomalies in Students’ Results

Hamza O. Salami * , Mohammed O. Yahaya**
* Lecturer,Department of Computer Science, Federal University of Technology, Minna, Niger State, Nigeria.
** Lecturer,Department of Computer Science and Engineering, University of Hafr Al Batin, Saudi Arabia.
Salami, H.O., Yahaya, M.O.(2019) On the Use of Extreme Learning Machines for Detecting Anomalies in Students’ Results,i-manager's Journal on Computer Science, 6(4),34-42. https://doi.org/10.26634/jcom.6.4.15724

Abstract

Examinations are means of assessing the knowledge or skills that students have acquired, after having been taught over a period of time. Anomalies in student results are noteworthy observations that require additional clarifications. Manual detection of anomalies in results leads to human errors and wastage of manpower. This paper describes how extreme learning machines can be used to automatically detect anomalies in student results. The results show that using extreme learning machines almost always produces better or equal results compared to decision trees.

Research Paper

Performance Evaluation of Cultural Artificial Bee Colony and Cultural Artificial Fish Swarm Algorithm

Busayo Hadir Adebiyi* , Ahmed Tijani Salawudeen**, Risikat Folashade O. Adebiyi***
*-***Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria.
Adebiyi, B.H., Salawudeen, A.T., Adebiyi,R.F.(2019) Performance Evaluation of Cultural Artificial Bee Colony and Cultural Artificial Fish Swarm Algorithm,i-manager's Journal on Computer Science, 6(4),43-50. https://doi.org/10.26634/jcom.6.4.15725

Abstract

The introduction of Computational Intelligence (CI) algorithms in the area of optimizations have been given significant attention in science and engineering allied disciplines. This is because they always find answers to a problem while maintaining perfect stability among its components. However, these algorithms sometimes suffer from premature convergence and fitness stagnation, which usually originates from the lack of explorative search capability of its perturbation operator. This paper presents a comparative performance of a Cultural Artificial Bee Colony (called the CABCA) algorithm and Cultural Artificial Fish Swarms Algorithm (called the mCAFAC). In both algorithms (CABCA and mCAFAC), the normative and situational knowledge is employed to guide the direction and step size of the population (ABC and AFSA). Four variants of each ABC and AFSA were developed using different configurations of cultural knowledge in Matlab/Simulink environment. A collection of twelve optimization benchmark functions was used to test the performance, and it was found that the modified algorithms (CABCA and mCAFAC) outperformed their respective standard (ABC and AFSA) algorithms.