ACCEPTANCE MODEL PREDICTION'S FOR E-ORIENTATION SYSTEMS CASE OF STUDY : PLATFORM "orientation-chabab.com"

Abstract

The orientation is the construction or development process of an educational or career plan. This process is adopting the information and the communication technologies through various platforms to help students making their own career decision. The purpose of this study is to generate an acceptance model prediction's of the e-orientation Moroccan platform "orientation-chabab.com" that can be used during the conceptual design of the future e-orientation platforms. The Technology Acceptance Model (TAM) is used as a theatrical model for early user acceptance of the e-orientation systems by evaluating an extended Technology Acceptance Model (TAM). Our experiment was conducted with the WEKA machine learning software by using five algorithms namely: NaïveBayes, J48, SMO, SimpleLogistic and OneR.According to the comparison of the accuracy rates of our simulation, the Sequential Minimal Optimization classifier gives us the best performance outcomes.

Full text

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4227
ACCEPTANCE MODEL PREDICTION’S FOR
E-ORIENTATION SYSTEMS
CASE OF STUDY : PLATFORM “orientation-chabab.com”
1RACHIDA IHYA, 1ABDELWAHED NAMIR, 1SANAA ELFILALI, 3FATIMA ZAHRA GUERSS,
4HAJAR HADDANI, 1MOHAMMED AIT DAOUD
1Laboratory of Information Technologies and Modeling, Department of Mathematics and Computer
Science, Faculty of Sciences Ben M'Sik, University Hassan II of Casablanca, Morocco.
2Computer Laboratory of Mohammedia, Computer Sciences Department, Faculty of Sciences and
Technicals Mohammedia, University Hassan II of Casablanca, Morocco
3Laboratory of Search Optimization, Computer Sciences Department, faculty of science, University CHOUAIB
DOUKKALI, MOROCCO
E-mail: 1rachida.ihya@gmail.com, 1abd.namir@gmail.com ,1el_filali_s@yahoo.fr, 2fatiguerss@gmail.com,
3haddani2009@gmail.com, 1aitdaoud.mohammed@gmail.com
ABSTRACT
The orientation is the construction or development process of an educational or career plan. This process is
adopting the information and the communication technologies through various platforms to help students
making their own career decision. The purpose of this study is to generate an acceptance model
prediction's of the e-orientation Moroccan platform “orientation-chabab.com” that can be used during the
conceptual design of the future e-orientation platforms. The Technology Acceptance Model (TAM) is used
as a theatrical model for early user acceptance of the e-orientation systems by evaluating an extended
Technology Acceptance Model (TAM). Our experiment was conducted with the WEKA machine learning
software by using five algorithms namely: NaïveBayes, J48, SMO, SimpleLogistic and OneR.According to
the comparison of the accuracy rates of our simulation, the Sequential Minimal Optimization classifier
gives us the best performance outcomes.
Keywords: E-orientation, Technology Acceptance Model, Extended TAM, Machine Learning, Algorithm.
1. INTRODUCTION
Choosing a suitable career may be difficult for
students because they have to consider several
criteria if they want to be on the success path.
Today, the orientation is the construction and
development process of an academic and career
plan as it is related to the knowledge of the learner
[1]. The e-orientation process is adopting the
Information Technologies (IT) to automate the
orientation task throughout various platforms [2]
which are accessible to everyone and where the
students can choose their educational and
professional orientation.
Today there is a lack of current research on the
acceptance of Moroccan electronic guidance
systems. Thus, most of the examples focus on
“Meta-model of e-orientation platforms” [3] and
“Modernization of a domain e-orientation Meta-
model” [4]. However, research that would focus on
the acceptance prediction's model for e-orientation
system has not been previously conducted.
The research on acceptance of an e-orientation
systems use takes a variety of theoretical
perspectives. Of all the theories, the Technology
Acceptance Model (TAM) is considered the most
influential and commonly employed theory for
describing an individual’s acceptance of
information systems [5].
The TAM comprises several variables explaining
behavioral intentions and the use of technology
directly or indirectly (i.e., perceived usefulness,
perceived ease of use, attitudes toward technology)
[6]. Researchers have investigated and replicated
these constructs and agreed that they are valid in
predicting the individual's acceptance of various
corporate information technologies [7–11]. TAM
has gained considerable prominence, particularly
due to its transferability to e-orientation context by
extending it with mediator variables, such as
perceived risk and perceived information quality
[12–14].
The goal of our research is to generate an
acceptance model of e-orientation systems that can
be used during the conceptual design of e-
orientation platforms.

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4228
In this study we will evaluate the acceptability of
an e-orientation Moroccan platform: orientation-
chabab.com. The choice of these platform has been
based on a previous study that evaluated Multiple e-
orientation Systems [4].
The survey instruments for this study was
developed using validated items from the
theoretical constructs of the extended TAM model
for e-orientation. The items for measuring High
level of validity is ensured through extensive
revision by experts and supported by previous
literature review that we will discuss in more detail
in the section 3. The participants were asked to be
familiar with e-orientation Moroccan platform:
“orientation-chabab.com” and complete the
questionnaire at a convenient time for them.
Since our data is tremendously increasing, it
becomes difficult for us to establish a relationship
between multiple features. This makes it difficult
for us to manually analyze the data for strategic
decision making. Machine learning is a method of
data analysis that automates analytical model
building. It is a branch of artificial intelligence
based on the idea that systems can learn from data,
identify patterns and make decisions with minimal
human intervention.[15,16].
The database included 256 samples. Using
various Machine learning classifier algorithms, the
best results were obtained by a SMO with accuracy
rates of “98.8281%”.
The rest of the paper is structured as follows.
Section 2 exhibits the state of arts. Section 3
presents the TAM. Section 4 extends the TAM in
the e-orientation systems based on literature review.
Section 5 outlines our research methodology.
Section 6 describes our exploratory results and
analysis discussion. Finally, section 7 concludes the
main results and gives an outline of possible future
research directions
2. STATE OF ARTS
To put our research into the context, we
summarize the most relevant works about some
pedagogical ontologies of Orientation Domain. For
example, in [4] the authors described the use of
ontology in the field of orientation and defines a
model of the set of domain knowledge. In this
work, the authors established an ontological model
and existing guidance platforms and it used a more
abstract model namely meta-model to modernize
the field guidance. This modernization allows us to
facilitate the understanding of the orientation field,
specify a core platform and simulate its operation.
In another work [3], the authors introduce a
comparison and description of the existing e-
orientation platforms, which is based on the
WSDL1. The purpose of this work is to have a
descriptive file enriched by features to propose a
meta-model of e-orientation platforms to facilitate
the guidance of students.
We notice that all the previous research has not
conducted the acceptance prediction's model for e-
orientation system. All these works have served as a
basis for the development of our approach which is
the choice of our platform based in the comparison
of the existing e-orientation platforms according to
the following features: • Create an account. • Look
for similar profiles. • Add Parent profiles. • Manage
portfolio. • Seek guidance. We choose the platform
orientation-chabab.com to predict its acceptability
by the users.
3. TECHNOLOGY ACCEPTANCE MODEL
TAM, proposed by Davis in 1985 [17], explains
and predicts the usage of information technologies
based on the Theory of Reasoned Action (TRA) of
[18]. The TAM includes perceived usefulness and
perceived ease of use as the main influencing
variables of an individual's acceptance of
information technologies [19]. Figure 1 illustrates
the TAM model [20].
Figure 1: Technology Acceptance Model (TAM) [20].
Perceived usefulness (PU) is “the degree to which a
person believes that a particular technology would
enhance his or her performance”. Perceived ease of
use (PEOU) is “the degree to which a person
believes that using a particular technology would be
effortless”. Behavioral intention (BI) refers to
possible actions of individuals in the future, which
can be based on forecasting people behavior [21].
The using of external variables depends on the type
of research and reflects the flexibility of TAM[22].
According to “Li, Yuanquan, Jiayin Qi, and
Huaying Shu” [22], attitude toward using
1 Web Services Description Language

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4229
technology is the connection between belief
variables (PEOU, PU) and BI. BI is the trend of the
user’s cognition about likes or dislikes to use the
information systems (IS). Usage Behavior (UB) is
the final IS use behavior. Thus, the TAM has been
verified for several information technologies by
researchers and practitioners [11,23,24].
Because many information technologies have
emerged since the mid-1990s, many researchers
have expanded Davis' TAM, pointing out its limits.
When information technology has a hedonic
characteristic such as on the web and in games,
studies extending the TAM [25,26] have added
playfulness or enjoyment variables to the TAM.
Moon and Kim [25] have extended the TAM in the
WWW domain by adding perceptual playfulness
variables. Van der Heijden [27] has also proposed a
user acceptance model of hedonic information
systems with a perceived enjoyment variable. For
compulsory use of information systems or for
information technologies for collaboration (e.g.,
groupware, instant messaging), social influence
variables such as subjective norms have been added
to extend the TAM with perceived usefulness and
perceived ease of use variables of the TAM [28–
30]. Over the last few decades, Davis' TAM has
been proposed with additional and expanded factors
on technology acceptance according to
technological characteristics, target users, and
context [19].
As shown in Figure 2, our proposed model extends
the TAM by adding to the variables that exist in
TAM two other variables that are: the perceived
risk and quality of information.
Figure 2: Extending TAM
Since the 1960s, perceived risk theory has been
used to explain consumers' behavior. Considerable
research has examined the impact of risk on
traditional consumer decision making [31]. Peter
and Ryan [32] defined perceived risk as a kind of
subjective expected loss, and [33] also defined
perceived risk as the possible loss when pursuing a
desired result. Cunningham [34] noted that
perceived risk consisted of the size of the potential
loss (i.e. that which is at stake) if the results of the
act were not favorable and the individual's
subjective feelings of certainty that the results will
not be favorable.
Information quality reflects the quality of service
or product and it is related to intentions of
consumers to purchase. Therefore it is plausible to
assume that perceived information quality
influences consumer purchase intentions [35–39].
4. EXTENDING TAM IN E-ORIENTATION
SYSTEMS
Recently e-orientation platforms has been a very
important tool in the students’ life, so they can
choose the best path of their academic and
professional development [3]. Among the most
used platforms in Morocco we named “orientation-
chabab.com”, which is a guide for high school
Moroccan graduates for access to private and public
universities and colleges. However, the
establishment of this e-orientation platform has
never been exposed to a study that shows its
acceptance by users.
To develop a successful orientation system, the
designer must familiarize him or herself with the
specifics of that environment, as well as the typical
and learned behavioral patterns that occur within it.
Orientation systems need to be accessible and
understandable for as many people as possible.
The acceptance and the usage of the platform
“orientation-chabab.com” have been examined
using Extending TAM. To understand the behavior
of the individual towards the orientation systems, it
is essential to research for the factors which explain
the users’ acceptance of e-orientation systems.
We divided our extending TAM into three
categories: The explanatory variables, the mediator
variables and the variable to predict (see Figure 3).

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4230
Figure 3: theoretical model of acceptance of E-orientation systems
Explanatory variables: Presents the external
variable that affects the decision making of an e-
orientation: Individual and Social factors.
Individual factors: It consists on individual
variables characteristics (gender, age, level
education, formation educational, experience and
resources) [40–42].
Social factors: The concept of social influence is
based on the subjective norm proposed in the TAM,
and describes the influence of people who are
important to the subject making decisions. In our
study context, we are talking about the social
factors that affect the acceptance and use of an e-
orientation (influence of : professional categories
and study’s level of parents, career professional of
relatives, support of relatives, effect of relative’s
and networks financial dependence) [40].
As Mediator Variables, we positioned four
factors:
Perceived usefulness (PU) The influence of user
perception on the usability of the e-orientation
Platform (Behavior Intention) [5].
Perceived ease of use (PEOU) The influence of
perceived ease of using E-orientation platform on
users’ intentions to use the e-orientation platform
(Behavior Intention) [5].
Perceived risk: Perceptions of risk in using the e-
orientation platform (in Perceived Risk) affect the
intention to use this platform (Behavior Intention).
The research of Wang [43] has demonstrated in
“Understanding the effects of trust and risk on
individual behavior toward social media platforms:
A meta-analysis of the empirical evidence’ that risk
are theorized and approved to have effects on
individuals' behaviors toward SMPs.
Perceived information quality: Perceptions of
users’ information quality in the e-orientation
platform (Behavior Intention).
As a Variable to predict we have the user’s
prediction of decision to accept or not using the e-
orientation platform.
The variables cited in our theoretical model
(Figure3) are supported by previous literature
review by experts as seen in (table1).

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4231
Table 1: Literature related to factors affecting intention
to use e-orientation systems.
Factors determine
e-orientation
platform intention
to use
Supported literature References
Individual factors Hung et al.2006
[44]
Social factors Hung et al.2006
Van Dijk et al
(2008).
[44,45]
Perceived
usefulness
Hung et al.2006,
Davis, Fred
D(1989)
[44,46]
Perceived ease of
use
Hung et al. 2006;
CDavis, Fred
D(1989)
[44,46]
Perceived risk (Featherman &
Pavlou, 2002; Gefen
et al., 2002; Sitkin
& Weingart, 1995)
[33,47]
Perceived
information
quality
Parasuraman et al.
1988; Lee et al.
2002; Kumar et al.
2007; Prybutok et
al. 2008; Nicolaou,
A. I., & McKnight,
D. H. (2006).
[48,49]
[13,50]
5. METHODOLOGY
4.1 Data collection
The study was conducted in MOROCCO and
our field of study is predicting the utilization
acceptance of the platform “orientation-
chabab.com”. We first distribute the questionnaire
for the interviewers and then asked them to use the
platform of Moroccan e-orientation during a period
of one month. The questionnaire was accompanied
by a covering letter explaining the research
objectives. By the end of the examination period,
the interviewers return to us the answered
questionnaire.
The extending TAM model has been used for
identifying suitable items. The questionnaire was
divided into two parts, one is the demographic
information and other is the structured
questionnaire as seen in Table2.
The structured questionnaire part includes the
different variables presented in the extending TAM.
There are 27 questions and each item is measured
on a 5-point Likert scale [51]. The targeted
individuals are between 18 years and 60 years or
older invited to take part in this survey and we
asked them to use the platform of e-orientation
during a month and to make an evaluation by
answering our questionnaire survey.
Table 2: Demographic Information.
Measure Item
Gender

Male
 Female
Age

18-20
 21-24
 25-45
 46-60
 >60
socio-professional
categories

Student
 Farmer
 Merchant, artisan,
Entrepreneur
 Senior, Professor,
Intellectual, Supervisor
 Intermediate Occupation
 Employee
 Worker
 Unemployed
 Inactive
 Other
Education level

College
 High school
 Baccalaureate
 Baccalaureate+2
 University degree,
Mastery(Bac+3or4)
 Master, DEA, DSS
 PhD
 No diploma
Marital Status

Single
 married
 divorced
 widowed
Lodgment
Situation

At my parents
 I live alone
 I live with other students
 I live in a couple
Commune Size

Big City
 Small Town
 Campaign
 Town
The distribution of the questionnaire was
administered online by mail or by SMS and
distributed among groups, forums and social
networks, and paper form by realizing direct
interviews. Their returns are recorded on an Excel
file in Google drive. The data collection was then
carried out in April 2018 which lasted 6 months.
We received 256 Returns.
In this study 140 respondents were male
(54.69%) as shown in figure 4. Those who sent
more returns are those who live in big cities (72%)

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4232
as seen in figure 5. About 237 of the respondents
are between the ages of 18-45, and only (7, 42%) of
the respondents are 60 years or older. Most
respondents had higher education (96, 48%).
Figure 4: The returns of women and men
Figure 5: The size of the municipality of responders to
the survey
Once we have collected our data, we start to
examine it and work out what we can do with it.
The objective we have is one of prediction: given
the data we have, predict what the next person will
make the decision form the e-orientation platform.
Prediction models were developed through
rigorous comparative study of important and
relevant machine learning classifier algorithms
techniques namely: NaiveBayes, SMO, J48,
SimpleLogistic and OneR. Performance
comparison was also carried out for measuring
unbiased estimate of the prediction models using
full-training set method. We conducted experiment
in the WEKA environment.
Machine learning (ML) is the scientific study of
algorithms and statistical models that computer
systems use to effectively perform a specific task
without using explicit instructions, relying on
patterns and inference instead. It is seen as a subset
of artificial intelligence. Machine learning
algorithms build a mathematical model based on
sample data, known as "training data", in order to
make predictions or decisions without being
explicitly programmed to perform the task [52–54].
4.2 Material
The WEKA (Waikato Environment for
Knowledge Analysis) is a popular suite of machine
learning software written in Java, developed at the
University of Waikato, New Zealand. It was first
introduced by [55] as a workbench designed to aid
in the application of machine learning technology
to real world data sets.
In our data analysis, we have downloaded and
installed the software “WEKA” version 3.9 which
is available from WEKA University of Waikato
website2.
We have exported in CSV format our data file
in the “WEKA” tool which will in turn show us the
27 attributes that will allow us to implement a
model to predict the acceptation of the e-orientation
systems (see table 3).
Table 3: List of attributes from our data.
4.3 Classification Algorithms
Our research study uses different well-known
classifiers, such as NaïveBayes, SMO, J48,
SimpleLogistic and OneR for validating the output
of "decision making for using the platform
“orientation-chabab.com”.
2 http://www.cs.waikato.ac.nz/ml/weka

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4233
Decision tree algorithm J48: J48 classifier is a
simple C4.5 decision tree for classification. It
creates a binary tree. The decision tree approach is
most useful in classification problem. With this
technique, a tree is constructed to model the
classification process. Once the tree is built, it is
applied to each tuple in the database and results in
classification for that tuple. The basic idea is to
divide the data into range based on the attribute
values for that item that are found in the training
sample. J48 allows classification via either decision
trees or rules generated from them [56].
Sequential Minimal Optimization classifier
(SMO): ) is an algorithm for efficiently solving the
optimization problem which arises during the
training of support vector machines. It was invented
by John Platt in 1998 at Microsoft Research. SMO
is widely used for training support vector machines
and is implemented by the popular libsvm tool. The
publication of the SMO algorithm in 1998 has
generated a lot of excitement in the SVM
community, as previously available methods for
SVM training were much more complex and
required expensive third-party QP solvers. SMO is
an iterative algorithm for solving the optimization
problem described above. SMO breaks this problem
into a series of smallest possible sub-problems,
which are then solved analytically. Because of the
linear equality constraint involving the Lagrange
multiplier, the smallest possible problem involves
two such multipliers [57].
Naive Bayes classifier: The Naive Bayes algorithm
is a simple probabilistic classifier that calculates a
set of probabilities by counting the frequency and
combinations of values in a given data set. The
algorithm uses Bayes theorem and assumes all
attributes to be independent given the value of the
class variable. This conditional independence
assumption rarely holds true in real world
applications, hence the characterization as Naive
yet the algorithm tends to perform well and learn
rapidly in various supervised classification
problems [6]. Naïve Bayesian classifier is based on
Bayes’ theorem and the theorem of total probability
[56].
Classification One Rule algorithm (OneR), short
for "One Rule", is a simple, yet accurate,
classification algorithm that generates one rule for
each predictor in the data, and then selects the rule
with the smallest total error as its "one rule". To
create a rule for a predictor, we have to construct a
frequency table for each predictor against the target.
OneR Algorithm for each predictor, For each value
of that predictor, make rule as follows [57] :
 Count how often each value of
target(class)appears
 Find the most frequent class
 Make the rule assign that class to this
value of the predictors
 Calculate the total error of the rules of
each predictor.
 Choose the predictor with the smallest
total error.
 Find the best predictor which possess the
smallest total error using OneR algorithm
Classification JRIP algorithm (JRip): JRip
(RIPPER) is one of the basic and most popular
algorithms. Classes are examined in growing size
and an initial set of rules for the class is generate
using incremental reduced error JRip (RIPPER)
proceeds by treating all the examples of a particular
decision in the training data as a class, and finding a
set of rules that cover all the members of that class.
Thereafter it proceeds to the next class and does the
same, repeating this until all classes have been
covered [36].
The choice of the learning algorithm that we
should use, is a critical step. Once the preliminary
testing is judged to be satisfactory, the classifier is
available for routine use. The classifier’s evaluation
is most often based on prediction accuracy (the
percentage of correct prediction divided by the total
number of predictions) [16].
4.4 Classifier Accuracy Measures
There are some parameters on the basis of which
we can evaluate the performance of the classifiers
such as TP rate, FP rate, Precision and Recall F-
Measure areas which are explained below.
The Accuracy of a classifier on a given test set is
the percentage of test set tuples that are correctly
classified by the classifier.
The Confusion Matrix is a useful tool for
analyzing how well your classifier can recognize
tuples of different classes. A confusion matrix for
two classes is shown in Table 3.
Given m classes, a confusion matrix is a table of at
least size m by m. An entry, CMi,j in the first m
rows and m columns indicates the number of tuples
of class i that were labeled by the classifier as class
j.

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4234
Table 4: Confusion Matrix.
C
1
Predicted
Class
C2
Actual Class
C1
C2
True positives
False positives
False negatives True
negatives
A confusion matrix contains information about
actual and predicted classifications done by a
classification system. Performance of such systems
is commonly evaluated using the data in the matrix.
Some standards and terms [37]:
1. True positive (TP): If the outcome from a
prediction is p and the actual value is also
p, then it is called a true positive.
True positive rate = diagonal element/ sum
of relevant row
2. False positive (FP): However, if the actual
value is n then it is said to be a false
positive.
False positive rate = non-diagonal
element/ sum of relevant row.
Precision and recall: Precision is the fraction of
retrieved instances that are relevant, while recall is
the fraction of relevant instances that are retrieved.
Both precision and recall are therefore based on an
understanding and measure of relevance. Precision
can be seen as a measure of exactness or quality,
whereas recall is a measure of completeness or
quantity. Recall is nothing but the true positive rate
for the class [56].
 Precision = diagonal element/sum of
relevant column.
 F-measures =2*precision*recall/(precision
+ recall)
In this paper, we have used WEKA (Waikato
environment for knowledge analysis) tool for
comparison of NaïveBayes, SMO, J48,
SimpleLogistic and OneR algorithm and calculating
efficiency based on accuracy regarding correct and
incorrect instances generated with confusion
matrix.
6. RESULTS AND DESCUSSION
We have performed classification using
NaïveBayes, SMO, J48, SimpleLogistic and OneR
algorithm on our data of 256 instances in WEKA
tool which provide us with inbuilt algorithms. We
obtained the following results:
Table 5: Classification accuracy test results
Table 5 demonstrates the classification accuracy
results of five classification algorithms. It is evident
from the table 5 that SMO has the highest
classification accuracy (98.8281%) where 253
instances have been classified correctly and 3
instances have been classified incorrectly. The
Second highest classification accuracy for JRip
algorithm is (77.7344%) in which 199 instances
have been classified correctly. Moreover, the J48
and NaiveBayes showed respectively a
classification accuracy of (75.7813 %) and
(71.875%). The OneR results in lowest
classification accuracy which is (53.9063%) among
the five algorithms. So the SMO outperforms the
NaiveBayes, J48, JRip and OneR in terms of
classification accuracy.
Table 6: The performance results of five models
Naïve
Bayes
SMO
J48
Simple
Logistic
OneR
Precision 72,8% 98,8
%
75,6
%
80,3% 54,6
%
Recall 71,9% 98,8
%
75,8
%
80,5% 53,9
%
F-
measure
71,9% 98,8
%
75,5
%
80,2% 52,9
%
TPR 71,9% 98,8
%
75,8
%
80,5% 53,9
%
FPR 8,5% 0,6% 8,5% 7,5% 15,9
%
As can be seen from Table 6, the precision, recall,
Fmeasure of SMO algorithms performed better than
NaïveBayes, J48, SimpleLogistic and OneR.
Furthermore, the points of bagging algorithms are
near the perfect point than the point of the four
remaining algorithm which means this machine
learning algorithm can identify a prediction of an e-
orientation system acceptation with very high
precision, reliability.
A distinguished confusion of SMO (sometimes
called contingency table). SMO is applied on the
data set and the confusion matrix is generated for
class “Decision of E-orientation” having five
possible values: Totally agree, not at all, agree,
mostly agree, and more or less agree.

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4235
Confusion Matrix:
For above confusion matrix, true positives for class
a=’Totally agree’ is 39 while false positives is 1
whereas, for class b=’ Not at all’, true positives is
30 and false positives is null. True positives for
class c=”Agree” is 104 while false positives is 1.
Whilst True positives for class d=” Mostly agree” is
42 while false positives is null. And True positives
for class e=” More or less agree” is 38 while false
positives is null. Diagonal elements of matrix
39+30+104+42+38 =253 represents the correct
instances classified and other elements 1+1+1 = 3
represents the incorrect instances.
Many different metrics are used in machine
learning and to build and evaluate models. In SMO
We employed four performance measures TP rate,
FP rate: precision, recall, F-measure.
In Table.3 SMO shows a high accuracy and true
positives Rate (TP Rate) as well as false positives
Rate (FP Rate). In general, the performance of
SMO is evaluated in term of Precision, TP Rate,
and FP Rate.
The proposed model with SMO classifier can be
used as a predictive tool for researchers,
instructional designers and expert of e-orientation
systems. The results of this study can be used
during the conceptual design of e-orientation
platforms. The proposed model is also useful as a
practical tool to test user’s acceptance, which
would provide early clues to risks of user rejection
of the e-orientation system. The knowledge of risks
at this stage would enable designers and responsible
of e-orientation to take preventive measures to
ensure user’s acceptance of the e-orientation
system.
In this study, a model is proposed based on TAM
model associated with social and individual
external factors to determinate the factors of user’s
acceptance of an e-orientation Moroccan platform:
orientation-chabab.com, by extending it with
variable mediator, as perceived and perceived
quality.
The goal of our research is to generate an
acceptance model of e-orientation systems that can
be used during the conceptual design of e-
orientation platforms.
The survey instruments for this study was
developed using validated items from the
theoretical constructs of the extended TAM model
for e-orientation platform. And we use machine
learning classifier algorithms techniques for
elaborate our predictive model.
7. CONCLUSION
The goal of this study is to generate an
acceptance model prediction’s for the e-orientation
systems that can be used during the conceptual
design of e-orientation platforms. For that we have
evaluated the acceptability of an e-orientation
Moroccan platform: orientation-chabab.com by
using a survey instruments that was developed
using validated items from the theoretical
constructs of the extended TAM model for e-
orientation and we apply machine learning
classifier algorithms techniques in our data for
elaborate our predictive model.
In this research we have performed the
experiments in order to determine the classification
accuracy of five algorithms in terms of which is the
better predictive algorithm of user's decision
making via the e-orientation platform “orientation-
chabab.com”, with the help of an attractive data
mining tool known as WEKA.
Five algorithms namely NaïveBayes, SMO,
J48, SimpleLogistic and OneR were compared on
the basis of different percentage of correctly
classified instances. All these four come under the
classification methods of data mining which makes
a relationship between a dependent (OUTPUT)
variable and independent (INPUT) variable by
mapping the data points. It is clear from the
simulation results that the highest classification
accuracy performance is for the SMO classifier
(98.8281%) for our datasets containing 27 attributes
with each 256.
Furthermore, the Second highest classification
accuracy for JRip algorithm is (77.7344%).
Moreover, the J48 and NaiveBayes showed
respectively a classification accuracy of (75.7813
%) and (71.875%). The OneR results showed less
accuracy as compared to the previous four
mentioned which is (53.9063%). This indicates that
SMO classification algorithm should be favored
over NaïveBayes, J48, SimpleLogistic and OneR
classifiers where classification accuracy
performance is important.
We conclude that the SMO classification
algorithm is the best algorithm for generating an
a b c d e <-- classified as
39 0 1 0 0 | a = Totally agree
0 30 0 0 0 | b = Not at all
1 0 104 0 0 | c = Agree
0 0 1 42 0 | d = Mostly agree
0 0 0 0 38 | e = More or less agree

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4236
acceptance model prediction's of the e-orientation
Moroccan platform “orientation-chabab.com” that
can be used during the conceptual design of the
future e-orientation systems.
In future work, we can include the extension of
the simulation performed in the WEKA
environment by increasing the number of instances
in a given dataset and comparing the classification
accuracy performance of the proposed algorithms.
Moreover, other factor can also be taken for
instance the time requirement to compare the
accuracy of the proposed algorithms which we
believe shall surely bring out certain important
aspects about the different algorithm which can
prove usefulness in the research field.
REFRENCES:
[1] Guerss F zahra, Aitdaoud M, Douzi K, Talbi
M, Namir A. Implementation of a
computerized system for the orientation of
the Moroccan student in the university.
Procedia-Social and Behavioral Sciences
2015;182:381–387.
[2] Guerss F, Aitdaoud M, Douzi K, Guemmat
K, Namir A. Towards a new ontology of the
Moroccan Post-baccalaureate learner profile
for the E-orientation system
“MMSyOrientation.” Dx Doi Org 2015.
[3] Adib J, Moutachaouik H, Marzak A, Hain M.
The Generation of the Components Meta-
Models of an E-Orientation System Using the
WSDL-UML Transformation 2018.
[4] Fatima-zahra G, Khadija D. Modernization of
a domain E-orientation metamodel. 2016
11th International Conference on Intelligent
Systems: Theories and Applications (SITA),
IEEE; 2016, p. 1–6.
[5] Davis FD. Perceived usefulness, perceived
ease of use, and user acceptance of
information technology. MIS Quarterly
1989:319–340.
[6] Zhang X, Prybutok VR, Koh CE. The role of
impulsiveness in a TAM-based online
purchasing behavior. Information Resources
Management Journal (IRMJ) 2006;19:54–68.
[7] Ariff MSM, Yeow S, Zakuan N, Jusoh A,
Bahari AZ. The effects of computer self-
efficacy and technology acceptance model on
behavioral intention in internet banking
systems. Procedia-Social and Behavioral
Sciences 2012;57:448–452.
[8] Chin WW, Todd PA. On the use, usefulness,
and ease of use of structural equation
modeling in MIS research: a note of caution.
MIS Quarterly 1995:237–246.
[9] Doll WJ, Hendrickson A, Deng X. Using
Davis’s perceived usefulness and ease-of-use
instruments for decision making: a
confirmatory and multigroup invariance
analysis. Decision Sciences 1998;29:839–
869.
[10] Mathieson K. Predicting user intentions:
comparing the technology acceptance model
with the theory of planned behavior.
Information Systems Research 1991;2:173–
191.
[11] Segars AH. Pie-Examining Perceived Ease of
Use and Usefulness: A Confirmatory Factor
Analysis. MIS Quarterly 1993:517.
[12] Roy SK, Balaji M, Kesharwani A, Sekhon H.
Predicting Internet banking adoption in India:
A perceived risk perspective. Journal of
Strategic Marketing 2017;25:418–438.
[13] Nicolaou AI, McKnight DH. Perceived
information quality in data exchanges:
Effects on risk, trust, and intention to use.
Information Systems Research 2006;17:332–
351.
[14] Legris P, Ingham J, Collerette P. Why do
people use information technology? A critical
review of the technology acceptance model.
Information & Management 2003;40:191–
204.
[15] Michie D, Spiegelhalter DJ, Taylor C, others.
Machine learning. Neural and Statistical
Classification 1994;13.
[16] Kotsiantis SB. Supervised Machine Learning:
A Review of Classification Techniques.
Informatica (Lithuanian Academy of
Sciences) 2007;31:249–268.
[17] Davis FD. A technology acceptance model
for empirically testing new end-user
information systems : theory and results.
Journal of Multivariate Analysis 1985.
[18] Fishbein M, Ajzen I. Belief, attitude,
intention, and behavior: An introduction to
theory and research 1977.
[19] Yoon C. Extending the TAM for Green IT: A
normative perspective. Computers in Human
Behavior 2018;83:129–139.
[20] Venkatesh V, Davis FD. A model of the
antecedents of perceived ease of use:
Development and test. Decision Sciences
1996;27:451–481.
[21] Molinari LK, Abratt R, Dion P. Satisfaction,
quality and value and effects on repurchase
and positive word-of-mouth behavioral

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4237
intentions in a B2B services context. Journal
of Services Marketing 2008;22:363–373.
[22] Li Y, Qi J, Shu H. Review of relationships
among variables in TAM. Tsinghua Science
& Technology 2008;13:273–278.
[23] Adams DA, Nelson RR, Todd PA. Perceived
usefulness, ease of use, and usage of
information technology: A replication. MIS
Quarterly 1992:227–247.
[24] Hendrickson AR, Massey PD, Cronan TP.
On the test-retest reliability of perceived
usefulness and perceived ease of use scales.
MIS Quarterly 1993:227–230.
[25] Moon J-W, Kim Y-G. Extending the TAM
for a World-Wide-Web context. Information
& Management 2001;38:217–230.
[26] Ha I, Yoon Y, Choi M. Determinants of
adoption of mobile games under mobile
broadband wireless access environment.
Information & Management 2007;44:276–
286.
[27] Van der Heijden H. User acceptance of
hedonic information systems. MIS Quarterly
2004:695–704.
[28] Venkatesh V, Davis FD. A theoretical
extension of the technology acceptance
model: Four longitudinal field studies.
Management Science 2000;46:186–204.
[29] Lou H, Luo W, Strong D. Perceived critical
mass effect on groupware acceptance.
European Journal of Information Systems
2000;9:91–103.
[30] Glass R, Li S. Social influence and instant
messaging adoption. Journal of Computer
Information Systems 2010;51:24–30.
[31] Lin W-B. Investigation on the model of
consumers’ perceived risk-integrated
viewpoint. Expert Systems With Applications
2008;34:977–988.
[32] Peter JP, Ryan MJ. An investigation of
perceived risk at the brand level. Journal of
Marketing Research 1976;13:184–188.
[33] Featherman MS, Pavlou PA. Predicting e-
services adoption: a perceived risk facets
perspective. International Journal of Human-
Computer Studies \/ International Journal of
Man-Machine Studies 2003;59:451–474.
[34] Cunningham MS. The major dimensions of
perceived risk. Risk Taking and Information
Handling in Consumer Behavior 1967.
[35] Wang RY, Strong DM. Beyond accuracy:
what data quality means to data consumers.
Journal of Management Information Systems
1996;12:5–33.
[36] Parasuraman A, Grewal D. The impact of
technology on the quality-value-loyalty
chain: A research agenda. Journal of the
Academy of Marketing Science
2000;28:168–174.
[37] Patterson PG, Spreng RA. Modelling the
relationship between perceived value,
satisfaction and repurchase intentions in a
business‐to‐business, services context: an
empirical examination. International Journal
of Service Industry Management
1997;8:414–434.
[38] Tam JLM. Customer Satisfaction, Service
Quality and Perceived Value: An Integrative
Model. Journal of Marketing Management
2004;20:897–917.
[39] Zeithaml VA. Consumer Perceptions of
Price, Quality, and Value: A Means-End
Model and Synthesis of Evidence: Journal of
Marketing 1988;52:2–22.
[40] Venkatesh V, Morris MG. Why don’t men
ever stop to ask for directions? Gender, social
influence, and their role in technology
acceptance and usage behavior. Management
Information Systems Quarterly 2000;24:115–
139.
[41] Radif$^1$ M, Fan$^1$ I-S, McLaughlin$^1$
P. EMPLOYMENT TECHNOLOGY
ACCEPTANCE MODEL (TAM) TO
ADOPT LEARNING MANAGEMENT
SYSTEM (LMS) IN IRAQI
UNIVERSITIES. Inted2016. 10th
international technology, education and
development conference, 2016, p. 7120–
7130.
[42] Kim DJ, Ferrin DL, Rao HR. A trust-based
consumer decision-making model in
electronic commerce: The role of trust,
perceived risk, and their antecedents.
Decision Support Systems 2008;44:544–564.
[43] Wang Y, Min Q, Han S. Understanding the
effects of trust and risk on individual
behavior toward social media platforms: A
meta-analysis of the empirical evidence.
Computers in Human Behavior 2016;56:34–
44.
[44] Hung S-Y, Chang C-M, Yu T-J.
Determinants of user acceptance of the e-
Government services: The case of online tax
filing and payment system. Government
Information Quarterly 2006;23:97–122.
[45] Van Dijk JA, Peters O, Ebbers W. Explaining
the acceptance and use of government
Internet services: A multivariate analysis of
2006 survey data in the Netherlands.

Journal of Theoretical and Applied Information Technology
15th August 2019. Vol.97. No 15
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
4238
Government Information Quarterly
2008;25:379–399.
[46] Davis FD, Bagozzi RP, Warshaw PR. User
acceptance of computer technology: a
comparison of two theoretical models.
Management Science 1989;35:982–1003.
[47] Sitkin SB, Weingart LR. Determinants of
risky decision-making behavior: A test of the
mediating role of risk perceptions and
propensity. Academy of Management Journal
1995;38:1573–1592.
[48] Parasuraman A, Zeithaml VA, Berry LL.
SERVQUAL: A multiple-item scale for
measuring consumer perceptions of service
quality. Journal of Retailing 1988;64:12–37.
[49] Lee YW, Strong DM, Kahn BK, Wang RY.
AIMQ: a methodology for information
quality assessment. Information &
Management 2002;40:133–146.
[50] Prybutok VR, Zhang X, Ryan SD. Evaluating
leadership, IT quality, and net benefits in an
e-government environment. Information &
Management 2008;45:143–152.
[51] Joshi A, Kale S, Chandel S, Pal D. Likert
scale: Explored and explained. British
Journal of Applied Science & Technology
2015;7:396.
[52] Hu H, Li J, Plank A, Wang H, Daggard G. A
comparative study of classification methods
for microarray data analysis. Proceedings of
the fifth Australasian conference on Data
mining and analystics-Volume 61, Australian
Computer Society, Inc.; 2006, p. 33–37.
[53] Shipp MA, Ross KN, Tamayo P, Weng AP,
Kutok JL, Aguiar RCT, et al. Diffuse large
B-cell lymphoma outcome prediction by
gene- expression profiling and supervised
machine learning. Nature Medicine
2002;8:68–74.
[54] Mair C, Kadoda GF, Lefley M, Phalp K,
Schofield C, Shepperd MJ, et al. An
investigation of machine learning based
prediction systems. Journal of Systems and
Software 2000;53:23–29.
[55] Garner SR, others. Weka: The waikato
environment for knowledge analysis.
Proceedings of the New Zealand computer
science research students conference, 1995,
p. 57–64.
[56] Patil TR, Sherekar S, others. Performance
analysis of Naive Bayes and J48
classification algorithm for data
classification. International Journal of
Computer Science and Applications
2013;6:256–261.
[57] Nasa C, Suman S. Evaluation of different
classification techniques for web data.
International Journal of Computer
Applications 2012;52:34–40.