You are on page 1of 7

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/337960752

Modeling Techniques for Knowledge Representation of Expert System: A


Survey

Article  in  Journal of Applied Computer Science & Mathematics · December 2019


DOI: 10.4316/JACSM.201902006

CITATIONS READS

3 369

4 authors, including:

Muhammad Lawan Jibril Etemi Garba


Federal University, Kashere Federal University of Technology, Yola
38 PUBLICATIONS   135 CITATIONS    27 PUBLICATIONS   37 CITATIONS   

SEE PROFILE SEE PROFILE

Gregory Msksha Wajiga


Federal University of Technology, Yola
32 PUBLICATIONS   172 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Machine Learning Driven Approach for Waterborne disease Warning System View project

COVID-19 Predictive Models View project

All content following this page was uploaded by Muhammad Lawan Jibril on 03 January 2020.

The user has requested enhancement of the downloaded file.


Journal of Applied Computer Science & Mathematics, Issue 2/2019, vol.13, No. 28, Suceava

Modeling Techniques for Knowledge Representation of Expert System: A Survey


1
L. J. MUHAMMAD, 2E. J. GARBA, 3N. D. OYE, 4G. M. WAJIGA
1
Mathematics and Computer Science Department, Federal University of Kashere, Gombe State
2,3,4
Computer Science Department, Modibbo Adama University of Technology, Yola
1
mljtech@gmail.com, 2e.i.garba@mautech.edu.ng, 3oyenath@yahoo.co.uk,
4
gwajiga@gmail.com

Abstract –Knowledge representation is one of the most procedures to solve problems that was difficult enough to
desirable things to make system intelligent. Every expert system acquire significant human expertise for their solutions. It can
may only be an intelligent if its intelligence is equivalent to the also be defined as a computer system that uses human
intelligence of human being for a particular domain. In knowledge to solve problems that normally require human
expert system development, a good solution depends on
intelligence. The expert systems represent the expertise
a good knowledge representation modeling technique
chosen. So, failure to choose appropriate technique can be a knowledge as data or rules within the computer [9]. These
major problem in the later stages of an expert system rules and data can be called upon when needed to solve
development, because some critical information cannot be problems. Expert systems are only those systems where the
encoded within the chosen technique. This study reviewed some knowledge for the definition of the system (regardless of the
various modeling techniques for knowledge presentation of importance of its role in the performance, or the particular
expert system, identified and discussed the pros and cons of method of representation) comes from a human whose ability
each technique. to solve problems is well beyond the common, up to the rank
Keywords: Artificial Intelligence, Expert System, of unusual expertise [12]. According [12] Expert System is
Knowledge Base, Knowledge transfer, Knowledge inference
defined as an intelligence system which uses extracted
I. INTRODUCTION knowledge from past domain expert decision making
reasoning in form of rules to solve problems that ordinarily
Artificial intelligence (AI) is one of the branches of require human expertise for their solution, and has the
computer science which enables computers to perform the capability to update its rule-base as new knowledge is
types of intelligent actions normally attributed to human discovered.
being[22]. Such actions include the ability to recognize objects
and scenes, understand conversational and written language, II. DEVELOPMENT PROCESS OF EXPERT SYSTEM
physically manipulate objects, and solve problems in complex
Development process of expert system is the activities
domains such as medicine, engineering, and battle
involve for knowledge transfer from human expert to
management [23]. It is a highly specialized technique of
knowledge base of the expert system and then to the user in
computer science that used to model and imitates the human
form of knowledge. These activities include knowledge
ability or expertise [10] [1][15]. In early 1980‘s, remarkable
acquisition, knowledge inference and knowledge
developments in various fields of AI had brought the need of
representation [17], [9-10].
expert systems. Initially knowledge based systems were
primarily regarded as expert systems. Knowledge based A) Knowledge Acquisition
systems and expert systems include two major components; Knowledge representation is one of the most basic and
one is knowledge base, and other is an inference engine [28]. actively researched areas of artificial intelligence [32-33]. And
However, the two systems can be distinguished based on how yet, there are many important is-sues underlying knowledge
and for what the system is being utilized [30]. The expert representation which have not been adequately addressed. One
system had been usually build to substitute or assist a human such issue is that of the representation of knowledge which is
expert in efficiently resolving a complex task with less lexically imprecise and/or uncertain [18]. The process of
complexity to save time [30]. On the other hand, Knowledge human knowledge transfer to knowledge base of expert system
based system provides a structured architecture to represent is termed knowledge acquisition. It involves extracting
knowledge explicitly [29]. One of the most promising areas of problem-solving expertise from knowledge sources, which are
applied Artificial intelligence is expert system which is usually domain experts [32-33]. According to [9-10]
program that solves problem normally requiring the knowledge acquisition is the process of extracting, structuring
knowledge and skills of human expertise thus combination of and organizing knowledge knowledge from human expert
facts and experience. According to [3] expert system is a domain to knowledge base of the expert system [8]. It can also
computer program that uses knowledge and inference be extracted from books, documents, sensors, or computer

DOI: 10.4316/JACSM.201902006 39
Computer Science Section

files. Knowledge acquisition of is a major and critical phase in


the development of an expert system. Knowledge acquisition
is considered by many to be the most difficult and precarious
stage in the knowledge engineering process [9], [16]. The
methods of knowledge acquisition can be divided into manual
and automated approaches. The techniques for manual
approaches include questionnaire, observation, and interview
e.t.c. Automated approach uses an induction system with case
histories and examples as input to derive the knowledge base
and it is also known as machine learning approach [16].

B) Knowledge Reasoning Fig. 1 A Simple example of a Semantic Network


Knowledge reasoning is also called Knowledge inference
[1]. Knowledge reasoning involves application of logical rules Semantic networks have been used for various purposes.
to the knowledge available in the knowledge base of the expert They are usually used to describe hierarchical information (see
system to deduce new information or knowledge [14]. The Fig. 1).
inference engine draws conclusions from the replicated human Therefore, semantic network is basically a graph where the
expertise in the knowledge base of the expert system, which is nodes are labeled by atomic formulas, and arcs represent
the hallmark of the expert system [5]. relations between them. The nodes of this graph then represent
entities and classes of entities. These classes then may be
C) Knowledge Representation
hierarchically ordered to represent the knowledge [10]. This
Knowledge representation implies a systematic means of
leads to two basic relations between the node, that is: subclass,
encoding what an expert knows about a knowledge domain in
entity of. Semantic networks techniques were first developed
an appropriate medium [3],[13]. It defines the manner in which
as a method of representing human knowledge. In fact every
the human expertise should be represented in the computer so
semantic network can be represented by the language of first
that it can be conveniently accessed for problem solving [13].
order logic. Even better, semantic nets can be directly
In expert system development, a good solution depends on
translated to logic programs [12]. Most of the principles of
a good representation. For expert systems applications, the
semantic networks are also employed in rule-based expert
initial choice of a representation technique is particularly
systems. Though, the basic feature of semantic networks that
important[2]. This is because the possible representation
are employed in many very large expert systems. It indeed,
techniques are diverse and the forcing criterion for the choice
leads to enhanced generalization, information reduction and
is normally not clear at the outset of the project [1]. The
also can dramatically increase performance. On the other hand
consequence of inadequate selection can be a major problem
it can lead to reduced precision [21]
in the later stages of an Expert System development, if it is
An Expert System stores knowledge as a semantic net
discovered that critical information cannot be encoded within
incorporates an inference engine devoted to operations like
the chosen representation technique [28]. A knowledge
inheritance. Such an inference engine will consist of two parts.
representation technique should support a number of different
One part will deal with rules by forward chaining, backward
activities. Various techniques may be appropriate for
chaining or some other method. The other part will handle net
representing different types of things and for supporting
operations matching relevant links in the net to deduce facts.
different kinds of activities [28]. This can force us in some
The main advantage of a semantic network is that related
cases to either find a representation system that supports
objects are located next to each other, so searching for
multiple uses of knowledge or else represent the same
information is usually straightforward. Also, semantic
knowledge several times in various representational systems.
networks give a clear description of knowledge because they
It would be desirable to have a single representation technique
give a picture of the structure of the information.
in which knowledge that has multiple uses will support those
A shortcoming of this technique is that there are no formal
uses [4].
ways of dealing with them. In addition, the structure of a real-
problem may get very complex. Thus, the search may be done
III. MODELING TECHNIQUES FOR KNOWLEDGE
in an inefficient way [12].
REPRESENTATION

A) Semantic networks technique B) Rule-based technique


Quillian is generally acknowledged as the inventor of Rule-based systems (also called production systems) are
semantic networks. A semantic network consists of nodes composed of production rules (rule-base), a database (working
which are connected together by arcs or links[12]. Nodes memory) and a control system (rule interpreter) [10]. The
represent objects, concepts, events or hypotheses and arcs domain knowledge in a rule-based system is represented as a
(links) represent relationships between these nodes. set of rules of the type [19]. The traditional, expert system

40
Journal of Applied Computer Science & Mathematics, Issue 2/2019, vol.13, No. 28, Suceava

engineering is based on the production systems approach (rule- the occurrence of exceptions [10]. The reasoning process starts
based systems) which emphasizes building a single monolithic by identifying a frame as applicable to the current situation.
knowledge-base. Rule-based approach is defined as one, Matching the set of frames against the facts available selects
which contains information obtained from a human expert, and an appropriate frame. The use of frames is relevant to non-
represents that information in the form of rules, such as IF– monotonic logic. Non-monotonic reasoning expresses
THEN. The rule can then be used to perform operations on reasoning with default attributes. Frames are based open many
data to inference in order to reach appropriate conclusion [11]. sources, but mainly and have been the first attempt to mimic
These inferences are essentially a computer program that human reasoning and knowledge hierarchy representation.
provides a methodology for reasoning about information in the Frames found tremendous application in early expert systems.
rule base or knowledge base, and for formulating conclusions Frames are grouping of slots that represent semantically close
[15]. Rules represent a very human friendly knowledge knowledge. Despite their wide-spread application, their
representation. One of the often cited rule-based systems is background is mostly technically based, and more simplify the
their modularity, simplicity and good performance [23]. development for humans then offer a solid base for sound
inference [20]. The principle of frames has been further
According to [30] [11] rule-based approach proved very enhanced and refined in Object Oriented Programming
efficient in simple domains, also due to very fast Rete paradigm and Multi-Agent System.
matching algorithm, but it’s use in complex systems is at The advantage of frames technique is good for representing
least very questionable. The major advantages of rule-based descriptive information. Also, procedural and declarative
systems include easy to formulate and emulate human knowledge can be represented by a frame. It can also represent
cognitive process and decision making ability, and also easy to hierarchies of information [10].
represent knowledge in a structured homogeneous and Its shortcoming is that, there is no formal theory of frames,
modular way. However, there are several limitations i.e., there are no formal ways of dealing with them yet [14].
associated with those systems: control structures contained in
the order of rules cause loss of flexibility, large number of D) Case-based technique
rules in the knowledge base causes the system to become Case-based technique is a major approach in automated
unwieldy and complicates its maintenance especially in the knowledge reasoning and machine learning. In case-based
case of subtle updates and the difficulty in assigning technique, a reasoner solves a new problem by noticing its
confidence rating to each rule which has great influence on the similarity with one or several previously solved problems and
Expert System comprehensibility [14]. by adapting their known solutions instead of working out a
The disadvantages of rule based approaches motived solution from scratch[20]. Case-based technique is a problem
researchers to finding solution to the large number of rules in solving method different from other AI approaches in many
the rule-based systems knowledge-base in order to enhance the aspects, Instead of using only general domain dependent
system understandability. The simplicity of this approach has heuristic knowledge like in the case of expert systems, Case-
encouraged its common usage in modeling a rule based expert based technique is able to use the specific knowledge of
system [14]. concrete, previously experienced, problem situations [24].
According to [21] rule-based technique are generally not Very important characteristic of the techniques is that, it is
suitable for modelling complex world relationships, and incremental learning since a new experience is memorized and
creation world models, but instead can used to represent available for future problem solving each time a problem is
procedural or shallow knowledge. Reasoning in a partially solved. Case-based reasoning is a powerful and frequently
observable domain requires a measure of certainty in the used way of human problem solving. Results from cognitive
proposition. There have been many more or less successful psychology have shown its psychological plausibility [24].
approaches to represent the uncertainty in a rule-based The ttechnique is a process that uses similar problems to solve
framework [21]. the current problem. It consists of two steps:
i. Find those cases in memory that solved problems
C) Frame based technique similar to the current problem, and
Frames were originally developed by Minsky[10]. It is a ii. Adapt previous solutions to fit the current problem
techniques where objects are described by a number of features The case library forms an extra important component in case-
or attributes, many of which stay constant from one instance based reasoning. The inference cycle using Case Based
to the next. A frame is a piece of structured data about typical Reasonin consists of:
characteristics of an object, act or event [20]. The knowledge i. Retrieving solutions
is more descriptive than procedural. Similar to rules, frames ii. Adapting solutions, and
must be able to deal with uncertainty and missing values. iii. Testing solutions
Frames may have default values and slot-filling procedures The critical step is to find and retrieve a relevant case from
associated with the slots, to cope with missing values. Frames the case library. Cases are stored using indexes. The stored
enable reasoning about object features such as inheritance and case contains a solution, which is then adapted by modifying

41
Computer Science Section

the parameters of the old problem to suit the new situation inheritance and collaboration. States and events are formalized
resulting in a proposed solution. The solution is tested and if with qualitative calculus. The terminological knowledge was
found successful, added to the case library. very useful in the development of the assertional component.
The main advantage of case based technique; it makes the It assisteed in understanding the problem domain, and in the
knowledge acquisition to be easier in case-based reasoning implementation stage, it assisted in building good inheritance
because of the granularity of the knowledge. Knowledge is hierarchies [25].
presented in precedent or resultant cases [18] One of the main limitations of object oriented
While the shortcoming of the technique is that, it makes the systems is that the system may be slower in execution. Also,
process of system reasoning so complex and slow. the message-passing mechanism in the object oriented
approach does not include the provisions of how a message has
E) Fuzzy based technique to be handled by the receiving object from a sender object’s
Fuzzy logic is an extension of classical logical systems. It background and perspective. In addition, as human
provides an effective conceptual framework for dealing with communication includes context and intention apart from
the problem of knowledge representation in an environment of content, merely sending a message content becomes a very
uncertainty and imprecision [31]. Meaning representation in restricted form of communication [14].
fuzzy logic is based on test-score semantics [10]. Fuzzy
experts are developed using the method of fuzzy logic, which G) Database technique
deals with uncertainty. This technique, which uses the A database is a collection of data organized to efficiently
mathematical theory of fuzzy sets, simulates the process of serve many applications by centralizing the data and
normal human reasoning by allowing the computer to behave minimizing redundant data. A DBMS is the software that
less precisely and logically than conventional computers. This permits an organization to centralize data, manage them
approach is used because decision-making is not always a efficiently, and provide access to the stored data by application
matter of black and white, true or false; it often involves gray programs. However, some large databases make knowledge
areas and the term may be. Accordingly, creative decision- discovery computationally expensive because some domains
making processes can be characterized unstructured, playful, or background knowledge, hidden in the database may guide
contentious, and rambling [18] and restrict the search for important knowledge. Therefore,
The main advantage of fuzzy based technique is that, it modern database methodologies need to process large
uses the theory of fuzzy logic is based upon the notion of volumes, multiple hierarchies, and different data formats to
relative graded membership and so are the functions of discover in-depth expert knowledge from large databases, such
mentation and cognitive processes. The utility of fuzzy sets as data mining and searching approach. [7]. In [26] a database
lies in their ability to model uncertain or ambiguous data, so expert system for coronary heart disease diagnosis in Jordan
often encountered in real life [18]. was developed using database techniques. The system was
The main shortcoming of the fuzzy based technique is that, developed based on the medical records in Jordan and
it usually implemented using soft computing programing implemented using visual studio. The system shows that, the
language. system has high ability to diagnose the patients based on the
records with a high accuracy percentage.
F) Object-oriented technique The main advantage database technique is that, it simplifies
Object-oriented methodology combines into one object the development process of the expert system.
data together with the specific procedures that operate on this The main disadvantage database technique is that large
data, where the object combines data and program code. databases make knowledge discovery computationally
Instead of passing data to procedures, programs send a expensive.
message for an object to perform a procedure that is already
embedded in it [10]. Then, the same message may be sent H) Scripts technique
many different objects, but each will implement that message Scripts are frame-like structure which represents a
differently. An object’s data are encapsulated from other parts common sequence of events such as "going into a shop,
of the system, so each object is an independent software selecting some goods, and paying for them. Prototype scripts
building block that can be used in many different systems consist of slots which can be filled in with details about a
without changing the program code [18]. specific event when a script is instantiated. Further, scenes
An object-oriented approach has been applied to the within scripts may lead into other scripts as well. There are
different stages involved in developing a knowledge base preconditions which should be met before the sequence takes
about insulin metabolism in [25] At an early stage the place [10]. Roles are slots for the people who appear in an
separation of terminological and assertional knowledge was instance of the script, and props are slots for the objects which
made. An object-oriented knowledge acquisition process was appear in an instance of the script. Results are the conditions
applied to the assertional knowledge. A frame description is which should be true after the completed script. Point of view
proposed which includes features like states and events, is a slot for a person which indicates that the sequence of

42
Journal of Applied Computer Science & Mathematics, Issue 2/2019, vol.13, No. 28, Suceava

events in the script is from that person's point of view. Scenes [3] C. Enrico, C. The Guide to Health Informatics, London, 3rd
are ordered and represent the chain of events which take place. Edition, 2003.
The main advantage of a script is that it gives a natural way [4] E. C. Okafor and C. C. Osuagwu (2007), Issues in Structuring the
of representing and organizing descriptive information [18]. Knowledge-base of Expert Systems, The Electronic Journal of
Knowledge Management,Vol. 5 (3), pp. 313 – 322.
The drawback of using scripts is that they can only be [5] E. Turban and E. Jay. Decision Support Systems and Expert
utilized when dealing with common sequences of events [18]. Systems”. 6th Edition, Prentice Hall, Upper Saddle River, NJ, 2002.
[6] G. Hwang. Knowledge elicitation and integration from multiple
I) Hybrid technique experts (1194). Journal Information Science and Engineering, Vol.
Sometimes more than one single form of knowledge 10, pp. 99-109
representation is required for developing an expert system. [7] K. C. Laudon, and J. P. Laudon. Essential of management
Some expert systems have utilized the advantages of disparate information systems” (5th ed), Englewood cliffs, NJ: Prentice Hall,
knowledge representation techniques by combining them in 2002.
one system [27]. This is known as hybrid representation. The [8] K. L. McGraw & K. Harbison-Briggs, Knowledge acquisition:
Principles and guidelines,NJ: Prentice-Hall, 1989.
main objective of utilizing hybrid knowledge representation is [9] L. J. Muhammad, E. J. Garba, N. D. Oye and G. M. Wajiga
to permit the representation of both shallow and deep (2018), On the Problems of Knowledge Acquisition and
knowledge. One of the most common hybrid representations Representation of Expert System for Diagnosis of Coronary Artery
involves rule-based and frame or network representation. Disease (CAD), International Journal of u- and e- Service, Science
Hybrid representation is not recommended for a situation and Technology, Vol. 11, No. 3, pp. 49-58
where the knowledge readily fits a single paradigm. [10] L. J. Muhammad (2019). Fuzzy Rule Driven Data Mining
Framework to Knowledge Acquisition for Expert System, Modibbo
IV. DISCUSSION Adama University of Technology, Yola, Unpublished PhD, thesis.
[11] M. B. Wiga, A. S. Noor and A. Igi. Rule Extraction for Fuzzy
The various modeling techniques have so far discussed in Expert System to Diagnose Coronary Artery Disease, (Published
the study, each of the technique has its advantages and Conference Proceedings style) in Proc of International Conference
on Information Technology, Yogyakarta, Indonesia 2016, 1(1),136-
disadvantages which have elaborated. Therefore, selecting 141
appropriate knowledge representation modeling techniques [12] M. R. Quillian. Semantic memory, In Semantic information
plays a significant role in the development of an expert system. processing”, ed., by M.Minsky, MIT Press, Cambridge, Mass, 1986,
The question of what is the best method of knowledge 216-270.
representation is still an open question in the Artificial [13] M. Minsky. Framework for representing knowledge, in the
Intelligence community. This matter requires its own separate psychology of computer vision, ed., P. Winston, McGraw-Hill NY,
research. According to Mehdi (1993), the following features pp. 211-277, 1975.
should be considered when choosing a knowledge [14] O. O. Oladipupo A fuzzy association rule mining expert-driven
representation method: approach to knowledge acquisition. Ph.D. Thesis, Covenant
University, 2012.
i. Ease of representation for particular domain [15] S. Kaisler (1986). Expert systems: An overview, IEEE Journal
ii. Efficiency of space and time of Oceanic Engineering, Vol. OE-11(4).
iii. Relationship between knowledge base and inference [16] S. Chen, A knowledge acquisition scheme for ruled-based
engine systems (Published Conference Proceedings style) in Proc. IEEE
iv. Ease of human understanding and uncertainty Region Conference on Computer, Communication, Control and
Power Engineering, 1993, Vol. 2, pp. 621-625.
V. CONCLUSION [17] S. Mehdi. Expert Systems Development: Some Problems,
Motives and Issues in an Exploratory Study”, Dissertation submitted
Knowledge for expert system is acquired from human
to Department of Informatics, Lund University, Lund/Sweden, 1993.
experts and other documented sources and then represented or [18] Y. I. Liou. Expert system technology: knowledge acquisition.
organized as rules. These rules are electronically stored as a Handbook of Applied Expert Systems, New York: CRC Press. 2.1 -
body of knowledge or a knowledge base and represented to 2.11, 1999.
user, this usually called knowledge presentation. This paper [19] Y. D. Niranjana and S. Anto (2014). An Evolutionary-Fuzzy
reviewed some various modeling techniques for knowledge Expert System for the Diagnosis of Coronary Artery Disease”
presentation of expert system, identified and discussed the International Journal of Advanced Research in Compute Engineering
advantage and disadvantages of each technique. & Technology Vol. 3 (4).
[20] V. Marik, e. a. (2003). Umela Inteligence 4. ACADEMIA.
REFERENCES [21] M. Petrik , Knowledge representation for expert systems,
Published Conference Proceedings style) in Proc . at International
[1] A. Fayyad, U.,Piatesky-Shapiro, G., Smyth, P., and Uthurusamy, Conference for Undergraduate and Graduate Students of Applied
R. (Eds.), Advances in Knowledge Discovery and Data Mining, Mathematics 2004
Cambridge, 1996. [22] P. N. Stuart Russel. Artificial Intelligence, Modern Approach,
[2] A. Goodall. Expert systems (Computer science). Learned 2003.
Information, University of California, 1985, [23[ R. Girratano. Expert Systems, Priciples and Programming. PWS,
1998.

43
Computer Science Section

[24] R. Lopez de Mantaras (2001). Case-Based Reasoning, Machine [29] F. Hayes-Roth, and N. Jacobstein, (1994). The state of
Learning and Its Applications knowledge-based systems," Communications of the ACM, Vol. 37
[25] M. Ensing, R. Paton, P. Speela, R. Radab (2004). An object- No. 3, pp. 26-39
oriented approach to knowledge representation in a biomedical [30] S. Russell, and P. Norvig, Artificial Intelligence: A Modern
domain Artificial Intelligence in Medicine, 6: 6, pp. 459-482 Approach, 3rd ed., Prentice Hall, 2009.
[26] M. A. Ali (2017). Fuzzy expert system for Coronary Artery [31] L. A. Zadeh (1992) Knowledge Representation in Fuzzy Logic.
Disease diagnosis in Jordan, Heath Technology, Springer In: Yager R.R., Zadeh L.A. (eds) An Introduction to Fuzzy Logic
[27] P. Tanwar, T. V. Prasad, K. Datta (2013) Hybrid Technique for Applications in Intelligent Systems. The Springer International Series
Effective Knowledge Representation. In: Meghanathan N., in Engineering and Computer Science (Knowledge Representation,
Nagamalai D., Chaki N. (eds) Advances in Computing and Learning and Expert Systems), vol 165. Springer, Boston, MA
Information Technology. Advances in Intelligent Systems and [32] H. J. Levesque, and R. Brachman (1987). Expressiveness and
Computing, vol. 178. Springer, Berlin, Heidelberg tractability in knowledge representation and
[28] M. Malhotra and T. R. Gopalakrishnan Nai (2015).. Evolution of reasoning, Computational Intelligence, pp. 378–93
Knowledge Representation and Retrieval Techniques, International [33] R. I. Brachman, and H. J. Levesque, Readings in Knowledge
Journal of Intelligent Systems and Applications, Vol. 07, pp. 18-28 Representation Morgan Kaufmann Publishers, Inc., Los Altos, Calif.,
1985.

L. J. Muhammad is a PhD candidate at Modibbo Adama University of Technology, Yola, Adamawa State, Nigeria. He had
a Diploma in Law from Aminu Kano School of Islamic and Legal Studies, Kano, Certificate in Computer Appreciation from
Bayero University, Kano, International Diploma and Advance Diploma in Computing from Informatics Institute Singapore,
B.Sc. (Hons) in Computing from University of Portsmouth U.K, Masters in Business Administration (MBA), University of
Wales, UK and M.Sc. Computer Science from Bayero University, Kano, Nigeria. He is a Member of IEEE, Nigerian
Institute of Management (NIM) as a Chartered Manager, Nigeria Computer Society (NCS), Universal Association of
Computer and Electronics Engineers (UACEE), International Association of Engineers (IAENG), and Machine Intelligence
Research Labs (MIR Labs). He is presently working with Federal University, Kashere, Gombe State, Nigeria as academic staff
and his area of research interest are data mining, computational intelligence, big data, query optimization and soft computing.

Etemi Joshua Garba (PhD, MEng, BSc) is a Consultant in Data Science, Software Engineering and Multimedia
Technology. He is also a Senior Lecturer (Assistant Professor) in the department of Computer Science at the Federal
University of Technology (Modibbo Adama University of Technology), Yola, Adamawa State, Nigeria. He is a
graduate of B.Sc. Computer Science and has Master Degree in Software Engineering from the St. Petersburg State
University of Information Technology, Mechanics & Optics. St. Petersburg, Russia. He has 14 years of experience in
teaching, research and community development. He is a member of Institute of Electrical and Electronics Engineers
(IEEE); member of editorial board of peer-reviewed Software Engineering Journal and member of editorial board of peer-
reviewed American Journal of Software Engineering and Applications

Oye, N. D., receive his M.Tech OR (Operations Research) degree from the Federal University of Technology
Yola- Nigeria in 2002. He is a lecturer in the department of Mathematics and Computer Science in the same
University (for the past 15yrs). In 2013 he obtains his PhD in Information System in the faculty of Computing and Information
Technology at the Univeristi Teknologi Malaysia, Skudai, Johor, Malaysia. Currently he is an Associate Professor in
ICT/Information System in Modibbo Adama University of Technology Yola, Nigeria, in the School of Physical Sciences in
the Department of Computer Sciences, and he is now the HOD.

G. M. Wajiga is a professor of soft computing at Modibbo Adama University of Technology, Yola, Nigeria. He had his Ph.D.
from Abubakar Tafawa Balewa University, Bauchi, Nigeria. He also attended Ahmadu Bello University Zaria, Nigeria and
Aston University Birmingham, England for B.Sc. and M.Sc. degrees, respectively. He has supervised and examined
dozens of Ph.D. students. Also, he is a visiting scholar to some universities in Nigeria and abroad.

44

View publication stats

You might also like