Available via license: CC BY 4.0
Content may be subject to copyright.
education
sciences
Review
Academic Advising Systems: A Systematic Literature
Review of Empirical Evidence
Omiros Iatrellis 1, *, Achilles Kameas 2and Panos Fitsilis 3,4
1Department of Computer Science and Engineering, University of Applied Sciences of Thessaly,
Larissa 41110, Greece
2School of Science & Technology, Hellenic Open University, Patras 26335, Greece; kameas@eap.gr
3Department of Business Administration, University of Applied Sciences of Thessaly, Larissa 41110, Greece;
fitsilis@teilar.gr
4Innopolis University, Kazan 420500, Russia
*Correspondence: iatrellis@teilar.gr; Tel.: +30-2410-584-594
Received: 12 September 2017; Accepted: 13 December 2017; Published: 19 December 2017
Abstract:
This paper aims to provide the reader with a comprehensive background for understanding
current knowledge on Academic Advising Systems (AAS) and its impact on learning. It constitutes
an overview of empirical evidence behind key objectives of the potential adoption of AAS in generic
educational strategic planning. The researchers examined the literature on experimental case studies
conducted in the domain during the past ten years (2008–2017). Search terms identified 98 mature
pieces of research work, but inclusion criteria limited the key studies to 43. The authors analyzed
the research questions, methodology, and findings of these published papers and categorized them
accordingly. The results have highlighted three distinct major directions of the AAS empirical research.
This paper discusses the emerged added value of AAS research and highlights the significance of
further implications. Finally, the authors set their thoughts on possible uncharted key questions to
investigate both from pedagogical and technical considerations.
Keywords: academic advising systems; education
1. Introduction
With the advent of flexible curriculum systems in many Higher Educational Institutions (HEI) and
an ever wider variety range of courses and programs being offered, there is a present need to ensure
that students make the best use of available information to make more informed decisions regarding
their academic plan [
1
]. Besides required courses, which are compulsory for each student to be taken,
HEI also offer elective courses, but generally students lack information about the objectives and the
content of the course and often fail to take the appropriate ones for their academic plan. Moreover,
because the number and variety (in backgrounds, in knowledge, in goals) of students is also expanding
rapidly, it is increasingly important to tailor learning processes to students, since the same learning
pathway is unlikely to best serve all students [2].
HEI usually employ guidance counselors, people who are tasked with helping students making
their choices. Three principle models of advising include developmental advising, prescriptive
advising, and intrusive advising, each of which is informed by the goals of the advisor-student
interaction [
3
]. In all approaches, this support is based on the establishment of educational collaboration
between students and their academic advisors. Advisors assist students by guiding them through the
university’s educational requirements, helping them schedule the most suitable modules, introducing
them to pertinent resources, promoting leadership and campus involvement, assisting in career
development, helping them with the timely completion of their studies, and helping them find ways
to make their educational experience personally relevant [
4
]. However, in practice the counselors are
Educ. Sci. 2017,7, 90; doi:10.3390/educsci7040090 www.mdpi.com/journal/education
Educ. Sci. 2017,7, 90 2 of 17
often overloaded with too many students and not enough time, and some students are not satisfied
with the quality of academic advising that the counselors provide [
5
]. Good advising yields a good
outcome in terms of understanding, planning, and applying strategies for academic success, while bad
advising will be frustrating and may have a damaging effect on students’ progress [6].
In order to support the educational process and to relief the HEI actors, a software system is needed
that will handle the advisory process in an efficient and effective way. An AAS can serve as a strategic
partner responsible for the process of supporting, motivating the student’s study plan, and assisting in
the achievement of their educational goal [
7
]. However, unlike most existing recommendation systems,
AAS require dealing with a large decision space, which grows combinatorially with the number of
courses, programs, and the students’ different backgrounds, knowledge, and goals but is also subjected
to many constraints (e.g., course prerequisites, maximum credit hour load, course priorities, etc.). In [
8
],
the authors describe the conceptual framework of a web-based AAS. In their study, more than 60% of
users, consisting of 361 students and 155 faculty members, agreed that the tasks of academic-related
matter such as course registration, course selection, academic progress, course information, course
scheduling, plan of study, and academic calendar should be included in the AAS. They also discuss
the importance of making AAS that are more than data repositories and including more intelligence so
the systems are able to provide reliable advice when helping the advisor and the student alike.
2. Motivation and Rationale of the Study
The motivation for this review is derived from the fact that empirical evidence is required for
theoretical frameworks to gain acceptance in the scientific community [
9
]. A search in relevant literature
did not reveal any reviews of empirical evidence of the added value of research in the specific domain.
Consequently, there was a need to supply the audience with an accredited overview. This paper aims
to fill that gap.
The value of any single study is derived from how it fits with and expands previous work, as well
as from the study’s intrinsic properties. Thus, putting together all the unbiased, credible results from
previous research would be a step towards understanding the whole picture and constructing a map of
our knowledge of the domain. In a sense, the rationale of our study was to manage the overwhelming
amount of publications through the critical exploration, evaluation, and synthesis of the previous
empirical results that merit reflection.
This paper’s goal is to carry out a systematic review of empirical evidence in order to
contribute towards
•a complete documentation of the applied research approaches so far;
•a feasibility study that captures the strengths and weaknesses of research in the domain, and
•
the identification of possible threats, and thus to motivate the research community to redefine or
refine related questions or hypotheses for further research (opportunities).
3. The Research Questions
The following research questions need to be addressed and are distinguished into primary
(generalized: set to fulfill the goals of the review) and secondary (sub-objectives/specific: refine the
primary—explanatory):
•
RQ1 (Primary)—Research Objectives: What are the basic research directions of AAS up to
now (in terms of measurable metrics) and which approaches do researchers follow to achieve
these goals?
•
RQ1.1 (Secondary)—What are the most significant results from past AAS research that
constitute empirical evidence with regard to their impact on the learning process?
•
RQ1.2 (Secondary)—Interpretation of the results: What do these results indicate regarding
the added value of this technology?
Educ. Sci. 2017,7, 90 3 of 17
•
RQ2 (Primary)—Future challenges: Which other emerging research approaches should be
explored in the AAS research area?
4. Research Methodology
Our process for finding and including or excluding papers is summarized in Figure 1. To increase
our coverage, we searched for relevant papers by conducting both a systematic search of available
research paper databases and by “snowballing”, starting with a set of core papers believed to be
in-scope, and expanding our set of consideration based on papers referenced by these papers.
Educ. Sci. 2017, 7, 90 3 of 16
• RQ1.2 (Secondary)—Interpretation of the results: What do these results indicate regarding
the added value of this technology?
• RQ2 (Primary)—Future challenges: Which other emerging research approaches should be
explored in the AAS research area?
4. Research Methodology
Our process for finding and including or excluding papers is summarized in Figure 1. To
increase our coverage, we searched for relevant papers by conducting both a systematic search of
available research paper databases and by “snowballing”, starting with a set of core papers believed
to be in-scope, and expanding our set of consideration based on papers referenced by these papers.
Figure 1. Survey methodology.
Two main methods were initially employed for this search. The first was to use our University’s
academic search engine, which covers several academic databases such as Xplore, ScienceDirect,
SpringerLink, and Wiley. The second was to consult Google Scholar (http://scholar.google.com/),
which covers more publications but is less structured and provides also more low quality material.
Search terms used included “academic advising systems, academic planning” in combination with
“universities, higher educational institutions”, and all words were used in different combinations.
The search process spanned from March 2017 to August 2017. The parameters of the search were
bound within the last ten years (2008–2017), in which emergence and adoption of AAS has grown
based on the number of relevant research papers.
Due to the orientation of this work towards the practical implementation and exploitation of
AAS, at the end of the data collection stage, the authors explicitly determined the article
inclusion/exclusion criteria (Table 1).
Table 1. Inclusion/exclusion criteria.
Include Exclude
• Articles published in Journals with Impact Factor
• Full-length articles published in International
Conference/Workshop Proceedings
• Approaches applied to formal education (e.g.,
universities, colleges, distance education, etc.)
• Date from 2008 to 2017
• Articles that do not present significant
empirical evidence (e.g., theoretical and
conceptual articles, essays, tool
demonstration, etc.)
• Book chapters
• Approaches applied to non-formal education
(e.g., Virtual Learning Environments, MOOC,
etc.)
This first round of analysis led to the identification of only 19 unique papers. Due to such a
limited number of publications, the literature base was expanded by applying a snowballing
Figure 1. Survey methodology.
Two main methods were initially employed for this search. The first was to use our University’s academic
search engine, which covers several academic databases such as Xplore, ScienceDirect, SpringerLink,
and Wiley. The second was to consult Google Scholar (http://scholar.google.com/), which covers
more publications but is less structured and provides also more low quality material. Search terms
used included “academic advising systems, academic planning” in combination with “universities,
higher educational institutions”, and all words were used in different combinations. The search process
spanned from March 2017 to August 2017. The parameters of the search were bound within the last
ten years (2008–2017), in which emergence and adoption of AAS has grown based on the number of
relevant research papers.
Due to the orientation of this work towards the practical implementation and exploitation of AAS,
at the end of the data collection stage, the authors explicitly determined the article inclusion/exclusion
criteria (Table 1).
Table 1. Inclusion/exclusion criteria.
Include Exclude
•Articles published in Journals with
Impact Factor
•Full-length articles published in International
Conference/Workshop Proceedings
•Approaches applied to formal education
(e.g., universities, colleges, distance education, etc.)
•Date from 2008 to 2017
•
Articles that do not present significant empirical
evidence (e.g., theoretical and conceptual
articles, essays, tool demonstration, etc.)
•Book chapters
•Approaches applied to non-formal education
(e.g., Virtual Learning Environments, MOOC, etc.)
Educ. Sci. 2017,7, 90 4 of 17
This first round of analysis led to the identification of only 19 unique papers. Due to such a limited
number of publications, the literature base was expanded by applying a snowballing technique [
10
] to
the initial set of papers. The reference list of each of the papers was screened for relevant contributions
and 43 papers were selected for detailed analysis at the end of the process. Saturation was used
as the stop criterion: we stopped the search when new papers no longer provided new challenges.
The selected papers were later analyzed by classifying their content and contribution in relation to
the AAS.
Next, the authors proceeded on with an article classification according to the adopted research
strategy (category), recommendation technique (method), research objectives (goals), and results.
Finally, we used non-statistical methods to evaluate and interpret findings of the collected studies,
and to conduct the synthesis of this review.
5. Results
In this section, the authors present their findings based on the analysis of the published
case studies.
According to the followed research strategy, most of the published case studies are exploratory
or experimental studies. Some of them are evaluation studies, while others are empirical studies or
surveys. Furthermore, the research topics differ from study to study, but most of them focus on science,
technology, engineering, and mathematics (STEM).
One important parameter is the research method adopted by authors to propose personalized
academic recommendations. In the field of AAS, the most popular filtering algorithms are
knowledge-based, followed by Computational intelligence-based, Collaborative filtering-based,
Hybrid, and content-based. Table 2displays the classification of the key studies according to the
recommendation method they adopt.
Table 2. Classification of case studies according to the research method.
Research Method Authors & Year (Paper Ref.)
Content-based Mostafa et al., 2014 [7], Lin et al., 2015 [11], Poeppelmann, 2011 [12]
Collaborative
filtering-based Chang et al., 2016 [1], Unelsrød, 2011 [5], Dhabi & Advisor, 2012 [13], Li, 2015 [14]
Knowledge-based
Xu et al., 2016 [2], Werghi, Naoufel, & Kamoun, 2009 [15], Roushan et al., 2014 [16],
Ai-nory, 2012 [17], Mohamed, 2015 [18], Koutrika, Bercovitz, & Garcia-Molina, 2009 [19],
Kristiansen, Sørensen, & Stidsen, 2011 [20], Engin et al., 2014 [21],
Henderson & Goodridge, 2015 [22], Mohamed, 2016 [23], Feghali, Zbib, & Hallah, 2011 [24],
Prof & Shakeel, 2012 [25], Hashemi & Blondin, 2010 [26], Laghari & Khuwaja, 2012 [27],
Ahmar, 2011 [28], Aslam & Khan, 2011 [29], Naini, Sadasivam, & Tanik, 2008 [30],
Albalooshi & Shatnawi, 2010 [31], Zhou & Yu, 2008 [32], (Al-ghamdi et al., 2012 [33],
Nambiar & Dutta, 2010 [34], Nguyen, Hoang, Tran, Nguyen, & Nguyen, 2008 [35]
Hybrid
Deorah, Sridharan, & Goel, 2010 [36], Daramola, Emebo, Afolabi, & Ayo, 2014 [6],
Sobecki & Tomczak, 2010 [
37
], Ragab, Mashat, & Khedra, 2012 [
38
], Lee & Cho, 2011 [
39
],
Fong, Si, & Biuk-aghai, 2009 [40]
Computational
intelligence-based
Abdulwahhab, Salem, & Makhmari, 2015 [
41
],
Meller, T., Wang, E., Lin, F., & Yang, 2009 [42]
,
Williams, 2013 [43], Adak, Yumusak, & Campus, 2016 [44], Goodarzi & Rafe, 2012 [45],
Fong & Biuk-aghai, 2009 [46]
The article classification according to the research objectives (goals) is illustrated in Table 3.
As seen in this table, the majority of studies investigate issues related to Selecting Courses followed by
Long-Term Academic planning and Choosing Programs/Majors.
Educ. Sci. 2017,7, 90 5 of 17
Table 3. Classification of case studies according to the research objectives.
Research Objective Authors & Year (Paper Ref.)
Choosing
Programs/Majors
Mostafa et al., 2014 [7], Meller, T., Wang, E., Lin, F., & Yang, 2009 [42], Deorah et al., 2010 [36],
Aslam & Khan, 2011 [29], Zhou & Yu, 2008 [32], Ragab et al., 2012 [38], Fong et al., 2009 [40],
Fong & Biuk-aghai, 2009 [46], Engin et al., 2014 [21]
Selecting Courses
Dhabi & Advisor, 2012 [
13
], Deline, G., Lin et al., 2015 [
11
], Li, 2015 [
14
], Daramola et al., 2014 [
6
],
Sobecki & Tomczak, 2010 [37], Adak et al., 2016 [44], Unelsrød, 2011 [5],
Chang et al., 2016 [1], Koutrika, Bercovitz, & Garcia-Molina, 2009 [19], Xu et al., 2016 [2],
Huang, Chung-Yi, Chen, & Chen, 2013 [47], Koutrika, Bercovitz, Ikeda, et al., 2009 [48],
Engin et al., 2014 [21], Albalooshi & Shatnawi, 2010 [31], Feghali et al., 2011 [24],
Laghari & Khuwaja, 2012 [27], Ahmar, 2011 [28], Naini et al., 2008 [30],
Albalooshi & Shatnawi, 2010 [31], Nguyen et al., 2008 [35], Poeppelmann, 2011 [12],
Al-ghamdi et al., 2012 [33], Nambiar & Dutta, 2010 [34], Abdulwahhab et al., 2015 [41],
Hashemi & Blondin, 2010 [26], Henderson & Goodridge, 2015 [22]
Long-Term
Academic planning
Werghi et al., 2009 [15], Roushan et al., 2014 [16], Williams, 2013 [43],
Kristiansen et al., 2011 [20], Albalooshi & Shatnawi, 2010 [31], Mohamed, 2016 [23],
Feghali et al., 2011 [24], Prof & Shakeel, 2012 [25], Ai-nory, 2012 [17], Mohamed, 2015 [18]
6. Key Studies Analysis
In this section, the authors present the findings of the review process and answer the initial set
research questions RQ1 and RQ1.1. The rest of the research questions (mostly the results of the case
studies and their comparative evaluation, as well as current and future trends, possible gaps, and new
research directions) are discussed in the next section.
RQ1: What are the basic research directions of AAS up to now (in terms of measurable metrics), and which
approaches do researchers follow to achieve these goals?
Firstly, we used technology roadmapping to explore the relationships between technological
resources, objectives, and the changing environment [
49
]. Figures 2–4show the frequency of the
categories listed on Tables 2and 3over the last ten years.
Looking at Figures 2–4, we can observe some trends. In Figure 2, it seems that publications
related to AAS peaked in 2010–2012, but is now beginning to rise again. The same could be said for
“Knowledge-based” research method in Figure 3and for “Selecting courses” research objective in
Figure 4. However, as the counts for the remainder of the categories captured in these charts are very
low overall, other trends may not be significant.
Educ. Sci. 2017, 7, 90 5 of 16
Unelsrød, 2011 [5], Chang et al., 2016 [1], Koutrika, Bercovitz, & Garcia-Molina,
2009 [19], Xu et al., 2016 [2], Huang, Chung-Yi, Chen, & Chen, 2013 [47], Koutrika,
Bercovitz, Ikeda, et al., 2009 [48], Engin et al., 2014 [21], Albalooshi & Shatnawi,
2010 [31], Feghali et al., 2011 [24], Laghari & Khuwaja, 2012 [27], Ahmar, 2011
[28], Naini et al., 2008 [30], Albalooshi & Shatnawi, 2010 [31], Nguyen et al., 2008
[35], Poeppelmann, 2011 [12], Al-ghamdi et al., 2012 [33], Nambiar & Dutta, 2010
[34], Abdulwahhab et al., 2015 [41], Hashemi & Blondin, 2010 [26], Henderson &
Goodridge, 2015 [22]
Long-Term Academic
planning
Werghi et al., 2009 [15], Roushan et al., 2014 [16], Williams, 2013 [43], Kristiansen
et al., 2011 [20], Albalooshi & Shatnawi, 2010 [31], Mohamed, 2016 [23], Feghali et
al., 2011 [24], Prof & Shakeel, 2012 [25], Ai-nory, 2012 [17], Mohamed, 2015 [18]
6. Key Studies Analysis
In this section, the authors present the findings of the review process and answer the initial set
research questions RQ1 and RQ1.1. The rest of the research questions (mostly the results of the case
studies and their comparative evaluation, as well as current and future trends, possible gaps, and
new research directions) are discussed in the next section.
RQ1: What are the basic research directions of AAS up to now (in terms of measurable metrics), and
which approaches do researchers follow to achieve these goals?
Firstly, we used technology roadmapping to explore the relationships between technological
resources, objectives, and the changing environment [49]. Figures 2–4 show the frequency of the
categories listed on Tables 2 and 3 over the last ten years.
Looking at Figures 2–4, we can observe some trends. In Figure 2, it seems that publications
related to AAS peaked in 2010–2012, but is now beginning to rise again. The same could be said for
“Knowledge-based” research method in Figure 3 and for “Selecting courses” research objective in
Figure 4. However, as the counts for the remainder of the categories captured in these charts are very
low overall, other trends may not be significant.
Figure 2. Classification of research papers by year of publication.
Figure 2. Classification of research papers by year of publication.
Educ. Sci. 2017,7, 90 6 of 17
Educ. Sci. 2017, 7, 90 6 of 16
Figure 3. Roadmaps for Academic Advising Systems (AAS) research method evolution.
Figure 4. Roadmaps for AAS research objective evolution.
6.1. Choosing Programs/Majors
Major selection is a very important step in the academic student life. Authors in [7,36],
implemented a case based reasoning (CBR) system that recommends to the candidate the most
suitable major, after comparing the historical cases by the student case. Paper [42] presents two
novel nearest neighbor-like classification algorithms for program recommendation, which provide a
program planning service to academic advisors and students of post-secondary institutions, while
paper [32] proposes an auto-decision system, which helps distance education students choose their
majors. Furthermore, the basic idea in paper [29] is to design a model for testing and measuring the
student capabilities like intelligence, understanding, comprehension, and mathematical concepts,
plus his/her past academic record and his/her intelligence level, and applying the results to a
rule-based decision support system to determine the compatibility of those capabilities with the
available faculties/majors.
A HEI is always interested in predicting the number of students into programs, because it is
important for the school to track the interest of students in program studies so as to ensure that
adequate classroom seats are available, appropriate instructors are available, and an appropriate
schedule is produced for the students. Paper [38] presents an approach using hybrid
recommendation based on data mining techniques and knowledge discovery rules, while papers
Figure 3. Roadmaps for Academic Advising Systems (AAS) research method evolution.
Educ. Sci. 2017, 7, 90 6 of 16
Figure 3. Roadmaps for Academic Advising Systems (AAS) research method evolution.
Figure 4. Roadmaps for AAS research objective evolution.
6.1. Choosing Programs/Majors
Major selection is a very important step in the academic student life. Authors in [7,36],
implemented a case based reasoning (CBR) system that recommends to the candidate the most
suitable major, after comparing the historical cases by the student case. Paper [42] presents two
novel nearest neighbor-like classification algorithms for program recommendation, which provide a
program planning service to academic advisors and students of post-secondary institutions, while
paper [32] proposes an auto-decision system, which helps distance education students choose their
majors. Furthermore, the basic idea in paper [29] is to design a model for testing and measuring the
student capabilities like intelligence, understanding, comprehension, and mathematical concepts,
plus his/her past academic record and his/her intelligence level, and applying the results to a
rule-based decision support system to determine the compatibility of those capabilities with the
available faculties/majors.
A HEI is always interested in predicting the number of students into programs, because it is
important for the school to track the interest of students in program studies so as to ensure that
adequate classroom seats are available, appropriate instructors are available, and an appropriate
schedule is produced for the students. Paper [38] presents an approach using hybrid
recommendation based on data mining techniques and knowledge discovery rules, while papers
Figure 4. Roadmaps for AAS research objective evolution.
6.1. Choosing Programs/Majors
Major selection is a very important step in the academic student life. Authors in [
7
,
36
],
implemented a case based reasoning (CBR) system that recommends to the candidate the most
suitable major, after comparing the historical cases by the student case. Paper [
42
] presents two novel
nearest neighbor-like classification algorithms for program recommendation, which provide a program
planning service to academic advisors and students of post-secondary institutions, while paper [
32
]
proposes an auto-decision system, which helps distance education students choose their majors.
Furthermore, the basic idea in paper [
29
] is to design a model for testing and measuring the student
capabilities like intelligence, understanding, comprehension, and mathematical concepts, plus his/her
past academic record and his/her intelligence level, and applying the results to a rule-based decision
support system to determine the compatibility of those capabilities with the available faculties/majors.
A HEI is always interested in predicting the number of students into programs, because it is important
for the school to track the interest of students in program studies so as to ensure that adequate classroom
seats are available, appropriate instructors are available, and an appropriate schedule is produced for the
students. Paper [
38
] presents an approach using hybrid recommendation based on data mining techniques
and knowledge discovery rules, while papers [
40
,
46
] propose a hybrid model of neural network and
decision tree classification for tackling college admissions prediction problems for each available program.
Paper [
21
] suggests an expert system implemented and tested using Oracle Policy Automation
(OPA) software for scholarship recommendation and eligibility checking.
Educ. Sci. 2017,7, 90 7 of 17
6.2. Selecting Courses
As seen from Table 3, selecting courses is a primary research objective. One crucial issue in this
category that authors attempt to address is how to propose careful recommendations for different
courses according to students’ goals and preferences. More specifically, the authors seek to prepare
course lists that satisfy several student and university constraints, some of which depend on individual
student cases [
2
,
5
,
6
,
22
,
24
,
27
,
28
,
50
]. In some case studies, authors choose to apply computational,
intelligence-based algorithms to reach a degree of automatic advising by combining genetic algorithms
with decision trees for developing the short-term curricular schedule, as well as by combining
perception marks with the registered courses [
41
] or by assisting in data mining and intelligent adaptive
fuzzy logic for implementing an elective course suggestion system [
44
]. In paper [
37
], the authors
present recommendations of student courses using ant colony optimization and concluded that their
solution is promising, since it overcomes most of the disadvantages of classical approaches based
on collaborative filtering in terms of performance. AAS proposed in [
24
] employ recommendations
based on decision support tools, while other authors [
33
,
34
], use expert systems combined with
semantic infostructures [
22
,
28
,
47
], or database management systems [
6
]. The course recommendation
problem gets more complex when considering a large number of courses that have different weights,
values, and priorities addressed by the Course-Petri net, a specialization of Petri net that is used on
paper [
26
] as the foundation for development of an advising system. Student diversity is another
aspect that complicates the problem [
13
,
14
]. For example, authors in [
13
] introduce an XML user-based
collaborative system called Automatic Academic Advisor, which advises a student to take courses that
were taken successfully by students with the same interests and academic performance. In another
case study, K-means algorithm has been used to determine the similarity of the students [14].
Another orientation is anticipating students’ future course selection as a means of short/mid-term
demand forecasting [12,31]. Paper [1] proposes a two-stage user-based collaborative filtering process
using an artificial immune system for the prediction of student grades, along with a filter for professor
ratings in the course recommendation for college students.
6.3. Long-Term Academic Planning
Long-term academic planning is a challenging and time-consuming task involved in AAS helping
a student prepare a complete study plan towards graduation. More importantly, because the number
and variety (in backgrounds, in knowledge, in goals) of students is expanding rapidly, it is more and
more important to tailor course sequences to students, since the same learning pathway is unlikely to
best serve all students [16,17,23,25]. For example, the system described in [15] implicitly implements,
via the decision tree, many academic rules and allows a systematic and exhaustive browse of the
different student plan instances, and it permits a methodological assessment and measurement of the
appropriateness of a given student academic plan. Work in [
18
] follows mathematical optimization
to generate a set of optimal or near optimal alternative plans that are of a similar quality and yet
structurally different. Human intelligence is then employed to analyze these alternatives and to either
approve one of them or refine the problem settings for generating further solutions.
To reduce the time-to-graduation, it is therefore of paramount importance for the student to
elect courses in a foresighted way by taking into account the possible subsequent course sequences
(including which courses are mandatory and which ones are not, and the course prerequisites) and
when the various courses are offered. The problem of planning the elective courses is modelled
in [
20
] using integer programming and three different solution approaches are suggested, including
a Branch-and-Price framework using partial Dantzig–Wolfe decomposition.
RQ1.1: What are the most significant results from past AAS research that constitute empirical evidence
with regard to their impact on the learning process?
Educ. Sci. 2017,7, 90 8 of 17
According to the research objectives explored by the authors, Table 4a displays a categorization of
the algorithmic-oriented findings from the collected studies. Table 4b displays a categorization of the
pedagogy-oriented findings.
Table 4.
(
a
) Cassification of the results of AAS case studies (algorithmic). (
b
) Classification of the results
of AAS case studies (pedagogical).
(a)
Objectives Results
Content-based
•Cosine Similarity function can be used to effectively measure the similarity
between the new student case and the other old cases by checking the nearest
features and values [7]
•Ontology-driven Software Development is appropriate for the development of
intelligent academic advising system with an ontology-based architecture and can
leverage familiar and proven software engineering tools and techniques [50]
•A case-based reasoning combined with rule-based reasoning AAS is able to
acquire new knowledge as usage of the system increases, while its rules can also
be modified with minimal effort. This is unlike when an ontology is used for
knowledge representation, which, although quite effective, require an advance
investment in quality ontology development before efficient course advising can
be obtained [6]
Collaborative
filtering-based
•The Authors in [1] use a demographic property of the student population and
their department to segregate the data and to address the sparsity problem that
limits the usefulness of collaborative filtering
•AAS can predict a student’s academic performance and interest for a course
based on a collection of profiles of students who have similar interests and
a similar academic performance in prior courses. In paper [13], students in
a Computer Science major are categorized in biclusters based on their similarity
on course features: comprehension skills, memorization skills, programming
skills, math skills, inferential thinking skills, problem solving skills, application of
strategies skills
•
K-means algorithm can be used to determine the similarity of the students. It was
identified that when K = 7, the clusters generated from the K means algorithm are
more informative and effective [14]
•Using an existing and relatively large dataset can be helpful in testing out the
collaborative filtering signals, as well as in choosing the appropriate
correlation algorithm [5]
•
Using friend relationships to weight the correlation used in the collaborative filtering
process did not turn out to improve the accuracy of the recommendations [5]
Educ. Sci. 2017,7, 90 9 of 17
Table 4. Cont.
(a)
Objectives Results
Knowledge-based
•Contrary to other contributions, which were intensively based on rule-based
approaches, paper [15] proposes a Decision Support System that formulates the
student planning and advising tasks as a search problem
•In paper [20], the problem of planning the elective courses is modelled using
integer programming and three different solution approaches are suggested,
including a Branch-and-Price framework using partial Dantzig–Wolfe
decomposition. The suggested algorithms achieved better results than the
currently applied meta-heuristic
•
By implementing a prescriptive advising model and a developmental advising model,
system testing revealed that 93% of academic advising test cases show an agreement
between the system advising in course selection and human advising [28]
•Authors in [29] collected all the required criterions, abilities, and capabilities for
each faculty/major and converted the knowledge into facts and rules in CLIPS
syntax, which is suitable for forward reasoning and can be used easily
Hybrid
•Authors in [37] concluded that Ant Colony Optimization-based method is
promising, since it enables students to overcome of the disadvantages of classical
approaches and gives higher values of performance measures.
•Authors in [42] conclude that the proposed nearest neighbour-like algorithms
outperform the two well-known classification algorithms Naïve Bayes and
J48 algorithm in terms of student classification success rate when there is
uncertainty present in the data
•
Classification algorithms can be used to identify what programs a student may fit
into; however, to involve multiple schools would require undesirable delays in
a web application [42]
•Paper [36] acquires implicit and latent interests of students in addition to the
explicitly stated ones, with the help of questionnaire responses that also
determine the degree of dilemma that the candidate faces and the time taken by
him/her to respond to each question
Computational
intelligence-based
•The major advantage of fuzzy AAS is that knowledge gradually turns into
wisdom and can be used as a decision making tool in critical situations that
replace the conventional FAQ [45]
•In paper [46] the experiments showed that the hybrid decision tree and neural
network approach improved accuracy in classification task
Educ. Sci. 2017,7, 90 10 of 17
Table 4. Cont.
(b)
Objectives Results
Choosing
Programs/Majors/HEI
•
Authors in [
46
] claim that scores are an important factor for qualifying students to
universities, especially for Mathematics and English courses. In their experiments,
they found that when students’ score of Mathematics and English are higher than
80%, almost all of them will be admitted to the universities.
•
Many students choose a university faculty/major because it has a good social
reputation or their friends have chosen it [
29
]. In the same paper, using abilities
tests, intelligence tests, and their past academic record, authors can measure some
student capabilities and abilities and determine which faculty/major is suitable
for them
Selecting Courses
•Authors in [1] claim that more feedback information from students is needed for
effective course selection and introduces a quality-control mechanism based on
filtering courses with poor instructor ratings
•
AAS can be used to overcome ineffective student academic advising in Distance
Education aimed at predicting a student’s academic performance and interest for
a course, based on a collection of profiles of students who have similar interests
and academic performance on prior courses [13].
•
Authors in [
5
,
14
] suggested that students’ major is a possible factor that is
related to the students’ performance. Experiment claim that male students aged
between 24 and 27 in the software major, mostly of European, Maori, and Asian
backgrounds, as well as middle aged (around 45) students, are more likely to be
high performing students. On the other hand, young male students with network
majors are more likely to be low performing students
•
Complex cases for AAS have to do with students that have changed from one
program to another (many of them more than once) and have failed and dropped
courses that are spread among different departments. Cases in which a student
has failed multiply in different departments and are more intricate to handle even
for the human course adviser [6]
•
Authors in [
44
] observed that the success of students in their previous required
courses and the student’s skills are found to be determinants of student’s success
in elective courses
•
Authors in [
2
] claim that an AAS framework has important implications
on how the curriculum planner should design the curriculum and allocate
teaching resources
•
AAS approaches for rolling prediction of future course enrolments should aim
at minimizing maintenance effort, especially in terms of adaptability to curricula
changes and graduation requirements [12]
Educ. Sci. 2017,7, 90 11 of 17
Table 4. Cont.
(b)
Objectives Results
Scheduling Courses/
Academic planning
•
Authors in [
43
] propose a multi-criteria fuzzy system model in academic advising
area to advise probation students to register an appropriate number of credit
hours so as to minimize their risk of losing their enrolment due to an incorrect
decision. The factors investigated are: (1) the Cumulative Grade Point Average,
(2) the Number of Covered Credit Hours, (3) Number of times that the student is
on Probation, and (4) the maximum grade that student has received.
•
The path associated with the students’ academic plan can be used to derive
a metric that measures the similarity of the students’ course history. This will
be quite useful for mining student profiles and analyzing and predicting
student performance [15]
•
In small institutions, an issue that needs to be considered in an AAS is the number
of students in each class to guarantee a minimum number of enrolled students [
17
]
•
Solving the Elective Course Planning Problem optimal can be used to reduce
students’ switch to another high school, which is highly undesirable [20]
•
During academic planning, an “explanation component” is important to explain to
students the rationale behind courses assignments to various semesters, or changes
to study plans after modifying any input setting [23]
•
More than 90% of respondents rated the online AAS increased their awareness of
the curriculum [24]
•
Multiple course selections from different faculties and departments enable
students to maximize their opportunities in registering courses of their own
interest and completing their degree requirements in the best possible way [25]
Long-Term
Course Planning
•Students with similar background will achieve similar expected Grade Point
Average (GPA) by following the same course sequence recommendation policy [
2
]
•
By analyzing the differences between the generated academic plans, students should
be able to better understand their choices and thus make appropriate decisions [18]
7. Discussion and Future Research
We examined the frequency of AAS publications and categories derived from Tables 2and 3
in Figures 2–4. AAS publications have peaked in 2011 and declined in 2013 but now are beginning
to rise again. “Knowledged-based” approaches are on the rise, peaking in 2010–2012. “Selecting
courses” are also on the upwising, rising in 2015. Approaches for “Choosing Programs/Majors” show
a small decline, while the remainder of the categories does not show obvious trends. From the former
analysis, it becomes apparent that, recently, the educational research community has started applying
sophisticated algorithmic methods on AAS. As seen in Table 4, the landscape of the AAS research
combines diverse and often conflicting aspects and results; however, the authors have highlighted
three distinct major axes of the AAS empirical research including:
Educ. Sci. 2017,7, 90 12 of 17
•
Pedagogy-oriented issues (e.g., student modeling, prediction of performance, assessment and
feedback, reflection and awareness): Several studies focus on pedagogically meaningful analysis
of students’ data in order to shed light on the whole picture. Academic advising, as a teaching
and learning process, requires a pedagogy that incorporates the preparation, facilitation,
documentation, and assessment of advising interactions.
•
Learning analytics (e.g., content analysis, discourse analytics, prediction and information
visualization): A number of studies combine institutional data, statistical analysis, and predictive
modeling to create intelligence upon which learners, instructors, or administrators can change
academic behavior.
•
Educational Data Mining (e.g., data mining, machine learning, and statistics): Several studies focus on
techniques, tools, and research designed to automatically extract meaning from large repositories of
data generated by or related to people’s learning activities in educational institutions.
These three axes are not completely autonomous, since significant overlaps may occur. However,
this statement could only constitute a limitation that does not reduce the added value of the findings.
RQ1.2: What do these results indicate regarding the added value of this technology?
One of the most important goals of the systematic review was to reveal the added value of the field
explored. From the above findings, it follows that analysis of factors that influence academic decisions,
student academic profile, and preferences in order to “recommend” the appropriate learning resources
has always been a request. In many cases, a typical software solution to AAS includes a rule-based
expert system. However, the dynamic nature of program requirements might turn maintenance of
the system into a crippling task. Implementing a user-based collaborative AAS is an appropriate
choice, although a major problem limiting the usefulness of such a system that should be addressed is
the sparsity problem, which refers to a situation in which data are insufficient to identify similarities
in students’ interests. AAS research results indicate that hybrid and content-based AAS are also
well suited to the academic advising domain, because these approaches allow the system to adapt to
the changes.
Researchers set academic advising within limits in which previously it was almost impossible
to infer recommendations automatically, due to the high levels of complexity that such a process
demands. In today’s advanced learning contexts, the AAS research community determines simple
and/or sophisticated approaches for recommending learning resources and explores their capabilities
by tracking actual changes and the progress of the learning process. The goal is to identify the most
significant factors in order to develop better systems. These systems will allow HEI and students to
evaluate and adjust their learning strategies and improve their performance in terms of learning outcomes.
Moreover, the learning analytics dimension and the opportunity of applying educational data
mining approaches are also explored with encouraging results. Consequently, the research community
could gain insight into the learning mechanisms that previously were a “black box”.
RQ2: Which other emerging research approaches should be explored in the AAS research area?
Complementary, the literature overview has revealed a number of unexplored issues in this
rapidly growing domain, including (but not limited to) the following:
•
Many modern educational models (for example Accelerated Study in Associate Programs
(ASAP) [
51
] and Guided Pathways (GP) [
52
]) share an emphasis on acceleration, programs that
offer fewer choices and more support, greater transparency of paths to completion for students,
and more mandatory and intrusive advising from day one through completion [
53
]. In their book
“Redesigning America’s Community Colleges”, Bailey, Jaggars, and Jenkins described the idea of
Guided pathways, a framework around which to structure program maps, meta-majors, e-advising,
and early alert systems [
54
]. AAS are critical to enabling the kind of monitoring and support demanded
by these models and must be understood as tools that are part of the broader reform. HEI need to
Educ. Sci. 2017,7, 90 13 of 17
carefully consider and plan how to change advising structures and daily practices so that existing
advisors can leverage the potential of emerging AAS research trends to improve student outcomes.
•
Technology acceptance is also a well addressed issue in educational research. Regarding AAS
acceptance, authors in [
55
] proposed a model that considers mainly two parameters: usability
and efficiency. However, more parameters should be explored in order to create a reliable AAS
acceptance model, e.g., effectiveness, maintainability, and portability. Researchers from the AAS
domain could also examine respective models that are suitable for the purposes of AAS.
•
The review process yielded very few articles related to scholarship recommendation and eligibility
checking, exploring life, and career goals. Assisting students in the clarification of their life/career
goals means helping students explore and define plans for the realization of these goals and evaluating
the progress of their efforts. It would be interesting to take advantage of the plethora of results from
AAS research by introducing innovative educational recommender systems in these areas.
•
One primary way of assisting student’s career development is by helping them understand their
own intrinsic interests and abilities through self-exploration and career exploration [
56
]. In this
context, existing literature highlights a need to examine AAS in a more holistic manner, one that
considers the connected nature of student’s interests, skills, and personality type. For example,
one of the most frequently used classification systems guiding personality type exploration that
can be utilized by AAS researchers is Holland’s [
57
] theory of vocational personality types and
work environments.
8. Conclusions
Previous works on AAS research provided significant insight into the conceptual basis of this
rapidly growing domain. However, these studies did not conduct an analysis of actual research results.
The current paper presents a systematic review of empirical evidence of AAS research. We searched
the literature and gathered representative, mature, and highly-cited articles of real case studies with
actual data from AAS domain. The analysis of selected case studies and their results shed light on
the approaches followed by the respective research communities and revealed the potential of this
emerging field of educational research. Along with the arising opportunities, we discovered a number
of gaps that require the researchers’ attention. Table 5illustrates our findings regarding the strengths,
weaknesses, opportunities, and threats (SWOT) of AAS research.
Table 5. Strengths, weaknesses, opportunities, threats (SWOT) of AAS research.
Strengths Weaknesses
•Large volumes of available educational data
increased accuracy of experimental results.
•Use of pre-existing powerful and valid
algorithmic methods.
•Interpretable multiple recommendations to
support learners/teachers.
•More precise user models for guiding
adaptation and personalization of systems.
•Reveal critical moments and approaches
of learning.
•
Lack of holistic AAS approach encompassing all
multi-facet factors like student competencies,
interests, and personality type.
•Misinterpretation of results due to human
judgment factors, wrong input data, etc.
•Heterogeneous data sources: not yet a unified
data descriptive vocabulary—data
representation issues.
•
Mostly quantitative research results. Qualitative
methods have not yet provided
significant results.
•Information overload—complex systems.
•Uncertainty: In many cases, only skilled
advisors could trust the recommendations and
interpret the results correctly.
Educ. Sci. 2017,7, 90 14 of 17
Table 5. Cont.
Opportunities Threats
•Modeling programs of studies and learning
pathways for data standardization, data
interoperability, and compatibility among
different tools and applications support multiple
course selections from different faculties and
departments generalized platform development.
•
Intellectual and affective learning opportunities
based on sophisticated decisions.
•Self-reflection/self-awareness in intelligent,
autonomous and massive systems.
•Feed machine readable results from the AAS
procedures to other data-driven systems for
diving decision making.
•Acceptance Model: e.g., perceived usefulness,
goal expectancy, perceived playfulness,
trust, etc.
•Over-analysis: the depth of analysis becomes
profound and the results lack generality.
•Possibility of misclassification of outcomes.
•Trust: contradictory findings
during implementations.
Every HEI needs to have effective academic advising mechanisms in order to increase student
development, which in turn can benefit enrollment, retention, and graduation rates. Many HEI offer
AAS designed to help students and their academic advisors recommend learning resources and review
degree requirements and the student’s progress towards the intended degree. However, existing
advising support software tools can augment the student-advisor relationship, but cannot and should
not replace in-person advising [
6
,
8
,
16
,
18
,
58
]. This is the point at which learning science, psychology,
pedagogy, and computer science intersect. The issue of understanding the deeper learning processes
and recommending them remains in the middle of this cross-path.
This work can be beneficial for several types of readers. For researchers interested in AAS,
this paper helps one to build upon existing work, avoiding the proverbial ‘reinvention of the
wheel’, helping to understand trends, and guiding efforts towards new directions. For practitioners,
this paper helps demonstrate the ways in which AAS approaches can be integrated into existing
system development paradigms, offering ideas on how academic advising can be adopted in practice,
including pointers to work containing further details. We believe that this active research area will
continue contributing with valuable pieces of work towards the development of more efficient and
effective advising services to both learners and HEI.
Author Contributions: Three authors contributed equally to the study.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Chang, P.; Lin, C.; Chen, M. A hybrid course recommendation system by integrating collaborative filtering
and artificial immune systems. Algorithms 2016,9, 47. [CrossRef]
2.
Xu, J.; Xing, T.; Member, S.; Van Der Schaar, M. Personalized course sequence recommendations.
IEEE Trans. Signal Process. 2016,64, 5340–5352. [CrossRef]
3.
Noaman, A.Y.; Ahmed, F.F. A new framework for e academic advising. Procedia Comput. Sci.
2015
,65,
358–367. [CrossRef]
4.
Pizzolato, J.E. Complex partnerships: Self-authorship and provocative academic-advising practices.
NACADA J. 2006,26, 32–45. [CrossRef]
5.
Unelsrød, H.F. Design and Evaluation of a Recommender System for Course Selection. Master’s Thesis,
Institutt for Datateknikk og Informasjonsvitenskap, Trondheim, Norway, June 2011.
Educ. Sci. 2017,7, 90 15 of 17
6.
Daramola, O.; Emebo, O.; Afolabi, I.; Ayo, C. Implementation of an intelligent course advisory expert system.
Int. J. Adv. Res. Artif. Intell. 2014,3, 6–12. [CrossRef]
7.
Mostafa, L.; Oately, G.; Khalifa, N.; Rabie, W. A case based reasoning system for academic advising in
egyptian educational institutions. In Proceedings of the 2nd International Conference on Research in Science,
Engineering and Technology (ICRSET’2014), Dubai, UAE, 21–22 March 2014.
8.
Ishak, I.B.; Lehat, M.L.B. A conceptual framework of web-based academic advisory information system.
In Proceedings of the 2012 IEEE Symposium on Humanities, Science and Engineering Research,
Kuala Lumpur, Malaysia, 24–27 June 2012; pp. 957–961.
9.
Follette, W.C.; Houts, A.C. Models of scientific progress and the role of theory in taxonomy development:
A case study of the DSM. J. Consult. Clin. Psychol. 1996,64, 1120–1132. [CrossRef] [PubMed]
10.
Goodman, L. The annals of mathematical statistics. JSTOR
1961
,32, 1. Available online: http://www.jstor.
org/stable/i312806 (accessed on 14 December 2017).
11.
Lin, F.; Leung, S.; Wen, D.; Zhang, F.; Rory, K. e-Advisor: A Multi-Agent System for Academic Advising.
Available online: http://io.acad.athabascau.ca/~oscar/pub/ABSHL2007.pdf (accessed on
14 December 2017
).
12.
Poeppelmann, D. A refined case-based reasoning approach to academic capacity planning. In Proceedings of
the 6th Conference on Professional Knowledge Management: From Knowledge to Action, Bonn, Germany,
21–23 February 2011; Available online: https://www.semanticscholar.org/paper/A-Refined-Case-Based-
Reasoning-Approach-to-Academi-P%C3%B6ppelmann/4c8e926dc484c684e826248ece160e4a0475c624?
tab=abstract (accessed on 14 December 2017).
13.
Dhabi, A.; Advisor, A. Automatic Academic Advisor. In Proceedings of the 2012 8th International Conference on
Collaborative Computing: Networking, Applications and Worksharing (CollaborateCom),
Pittsburgh, PA, USA
,
14–17 October 2012; pp. 262–268.
14.
Ganeshan, K.; Li, X. An intelligent student advising system using collaborative filtering. In Proceedings of
the 2015 IEEE Frontiers in Education Conference (FIE), El Paso, TX, USA, 21–24 October 2015.
15.
Werghi, N.; Kamoun, F.K. A decision-tree-based system for student academic advising and planning in
information systems programmes. Int. J. Bus. Inf. Syst. 2009,5, 1–18. [CrossRef]
16.
Roushan, T.; Chaki, D.; Hasdak, O.; Rasel, A.A.; Rahman, M.A.; Arif, H. University Course Advising:
Overcoming the challenges using decision support system. In Proceedings of the 2013 16th International
Conference on Computer and Information Technology (ICCIT), Khulna, Bangladesh, 8–10 March 2014;
pp. 8–10.
17.
Ai-Nory, M.T. Simple decision support tool for University Academic Advising. In Proceedings of the 2012
International Symposium on Information Technology in Medicine and Education (ITME), Hokkaido, Japan,
3–5 August 2012; Volume 1, pp. 53–57.
18.
Mohamed, A. A decision support model for long-term course planning. Decis. Support Syst.
2015
,74, 33–45.
[CrossRef]
19.
Koutrika, G.; Bercovitz, B.; Garcia-Molina, H. FlexRecs: Expressing and combining flexible recommendations.
In Proceedings of the 2009 ACM SIGMOD International Conference on Management of Data,
Providence, RI, USA
,
29 June–2 July 2009.
20.
Kristiansen, S.; Sørensen, M.; Stidsen, T.R. Elective course planning. Eur. J. Oper. Res.
2011
,215, 713–720.
[CrossRef]
21.
Engin, G.; Aksoyer, B.; Avdagic, M.; Bozanlı, D.; Hanay, U.; Maden, D.; Ertek, G. Rule-based expert systems
for supporting University Students. Procedia Comput. Sci. 2014,31, 22–31. [CrossRef]
22.
Henderson, L.K.; Goodridge, W. AdviseMe: An intelligent web-based application for academic advising.
Int. J. Adv. Comput. Sci. Appl. 2015,6, 233–243.
23.
Mohamed, A. Interactive decision support for academic advising. Qual. Assur. Educ.
2016
,14, 251–267.
[CrossRef]
24.
Feghali, T.; Zbib, I.; Hallah, S. A web-based decision support tool for academic advising. Educ. Technol. Soc.
2011,14, 82–94.
25.
Prof, A.; Shakeel, M. Student advising & planning software. Int. J. New Trends Educ. Implic.
2012
,3, 158–175.
26.
Hashemi, R.R.; Blondin, J. SASSY: A petri net based student-driven advising support system. In Proceedings of
the ITNG2010—7th International Conference on Information Technology: New Generations,
Las Vegas, NV, USA
,
12–14 April 2010; pp. 150–155.
Educ. Sci. 2017,7, 90 16 of 17
27.
Laghari, M.S.; Khuwaja, G.A. Electrical engineering department advising for course planning. In Proceedings
of the IEEE Global Engineering Education Conference (EDUCON), Marrakech, Morocco, 17–20 April 2012.
28.
Al Ahmar, M.A. A prototype student advising expert system supported with an object-oriented database.
Int. J. Adv. Comput. Sci. Appl. 2011, 100–105. [CrossRef]
29.
Aslam, M.Z.; Khan, A.R. A proposed decision support system/expert system for guiding fresh students in
selecting a faculty in Gomal University. Ind. Eng. Lett. 2011,1, 33–41.
30.
Naini, V.R.; Sadasivam, R.S.; Tanik, M.M. A web-based interactive student advising system using Java
frameworks. In Proceedings of the 2008 IEEE Southeastcon, Huntsville, AL, USA, 3–6 April 2008; pp. 172–177.
31.
Albalooshi, F.; Shatnawi, S. Online academic advising support. In Technological Developments in Networking,
Education and Automation; Springer: Dordrecht, The Netherlands, 2010; pp. 1–5.
32.
Zhou, Q.; Yu, F. Knowledge-based major choosing decision making for remote students. In Proceedings of the
2008 International Conference on Computer Science and Software Engineering (CSSE 2008), Wuhan, China,
12–14 December 2008; Volume 5, pp. 474–478.
33.
Al-Ghamdi, A.; Al-Ghuribi, S.; Fadel, A.; AL-Ruhaili, F.A.A.T. An expert system for advising postgraduate
students. IJCSIT Int. J. Comput. Sci. Inf. Technol. 2012,3, 4529–4532.
34.
Nambiar, A.N.; Dutta, A.K. Expert system for student advising using JESS. In Proceedings of the 2010
International Conference on Educational and Information Technology (ICEIT 2010), Chongqing, China,
17–19 September 2010; Volume 1, pp. 312–315.
35.
Nguyen, T.B.; Hoang, T.A.D.; Tran, H.; Nguyen, D.N.; Nguyen, H.S. An integrated approach for an academic
advising system in adaptive credit-based learning environment. VNU J. Sci. Nat. Sci. Technol.
2008
,24,
110–121.
36.
Deorah, S.; Sridharan, S.; Goel, S. SAES- expert system for advising academic major. In Proceedings of the
2010 IEEE 2nd International Advance Computing Conference (IACC), Patiala, India, 19–20 February 2010;
pp. 331–336.
37.
Sobecki, J.; Tomczak, J.M. Student courses recommendation using ant colony optimization. In Intelligent Information
and Database Systems; Nguyen, N.T., Le, M.T., Swiatek, J., Eds.; Springer: Berlin, Germany, 2010; Volume 5991,
pp. 124–133. Available online: https://pdfs.semanticscholar.org/5d8e/d2f7f865f94ce0f84815d7105d1eb45b4b4e.
pdf (accessed on 14 December 2017).
38.
Ragab, A.H.M.; Mashat, A.F.S.; Khedra, A.M. HRSPCA: Hybrid recommender system for predicting college
admission. In Proceedings of the International Conference on Intelligent System Design and Application
(ISDA), Kochi, India, 27–29 November 2012; pp. 107–113.
39. Lee, Y.; Cho, J. An Intelligent Course Recommendation System. SmartCR 2011,1, 69–84. [CrossRef]
40.
Fong, S.; Si, Y.; Biuk-Aghai, R.P. Applying a hybrid model of neural network and decision tree classifier
for Predicting University Admission. In Proceedings of the 7th International Conference on Information,
Communications and Signal Processing (ICICS 2009), Macau, China, 8–10 December 2009.
41.
Abdulwahhab, R.S.; Salem, H.; Makhmari, A. An educational web application for academic advising.
In Proceedings of the 2015 IEEE 8th GCC Conference and Exhibition (GCCCE), Muscat, Oman,
1–4 February 2015
;
pp. 1–4.
42.
Meller, T.; Wang, E.; Lin, F.; Yang, C. New classification algorithms for developing online program
recommendation systems. In Proceedings of the International Conference on Mobile, Hybrid, and On-Line
Learning (ELML’09), Cancun, Mexico, 1–7 February 2009.
43.
Williams, P. Fuzzy academic advising system for on probation students in colleges of applied
sciences—Sultanate of Oman. In Proceedings of the International Conference on Computing, Electrical and
Electronic Engineering (ICCEEE), Khartoum, Sudan, 26–28 August 2013; pp. 372–377.
44.
Adak, M.F.; Yumusak, N.; Campus, E. An elective course suggestion system developed in computer
engineering department using fuzzy logic. In Proceedings of the 2016 International Conference on Industrial
Informatics and Computer Systems (CIICS), Sharjah, UAE, 13–15 March 2016.
45.
Goodarzi, M.H.; Rafe, V. Educational advisor system implemented by web-based fuzzy expert systems.
Asian J. Inf. Technol. 2012,11, 77–82. [CrossRef]
46.
Fong, S.; Biuk-aghai, R.P. An Automated University admission recommender system for secondary school
students. In Proceedings of the 6th International Conference on Information Technology and Applications,
Sydney, Australia, 28–29 October 2017; pp. 978–981.
Educ. Sci. 2017,7, 90 17 of 17
47.
Huang, C.Y.; Chen, R.C.; Chen, L.S. Course-recommendation system based on ontology. In Proceedings
of the 2013 International Conference on Machine Learning and Cybernetics (ICMLC), Tianjin, China,
14–17 July 2013; Volume 3, pp. 14–17.
48.
Koutrika, G.; Bercovitz, B.; Ikeda, R.; Kaliszan, F.; Liou, H.; Garcia-Molina, H. Flexible recommendations for
course planning. In Proceedings of the IEEE 25th International Conference on Data Engineering (CDE’09),
Shanghai, China, 29 March–2 April 2009.
49.
Phaal, R.; Farrukh, C.; Probert, D. Technology roadmapping—A planning framework for evolution and
revolution. Technol. Forecast. Soc. Chang. 2004,71, 5–26. [CrossRef]
50.
Deline, G.; Lin, F.; Wen, D.; Gasevic, D. Ontology-driven development of intelligent educational systems.
In Proceedings of the IEEE Pacific Rim Conference on Communications, Computers and Signal Processing
(PacRim 2007), Victoria, BC, Canada, 22–24 August 2007; pp. 34–37.
51.
Kolenovic, Z.; Linderman, D.; Karp, M.M. Improving student outcomes via comprehensive supports:
Three-year outcomes from CUNY’s Accelerated Study in Associate Programs (ASAP). Community Coll. Rev.
2013,41, 271–291. [CrossRef]
52.
Jenkins, D.; Cho, S.-W. Get with the program ... and Finish it: Building guided pathways to accelerate student
completion. New Dir. Community Coll. 2013,2013, 27–35. [CrossRef]
53.
Tennat, G. SIX SIGMA: SPC and TQM in Manufacturing and Services; Gower Publishing Ltd.: Aldershot, UK,
2001; p. 6.
54.
Jenkins, D.; Bailey, T.; Jaggars, S. Redesigning America’s Community Colleges: A Clearer Path to Student Success;
Community College Research Center: New York, NY, USA, 2015.
55.
Ding, X. Web-Based Academic Advising System; Florida Atlantic University: Boca Raton, FL, USA, 2002; p. 95.
56.
Turner, S.; Conkel, J.; Starkey, M.; Landgraf, R.; Lapan, R.; Siewert, J.; Reich, A.; Trotter, M.; Neumaier, E.;
Huang, J.P. Gender differences in Holland vocational personality types: Implications for school counselors.
Prof. Sch. Couns. 2008,11, 317–326. [CrossRef]
57.
Holland, J.L. Making Vocational Choices: A Theory of Vocational Personalities and Work Environments;
Psychological Assessment Resources: Lutz, FL, USA, 1997.
58.
Mattei, N.; Dodson, T.; Guerin, J.T.; Goldsmith, J.; Mazur, J.M. Lessons learned from development of
a software tool to support academic advising. In Proceedings of the 2014 Zone 1 Conference of the American
Society for Engineering Education (ASEE Zone 1), Bridgeport, CT, USA, 3–5 April 2014; Volume 38238.
©
2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
