A novel hybrid recommender system approach for student academic advising named COHRS, supported by case-based reasoning and ontology

Abstract

The recent development of the WorldWideWeb, information, and communications technology have transformed the world and moved us into the data era resulting in an overload of data analysis. Students at high school use, most of the time, the internet as a tool to search for universities/colleges, university?s majors, and career paths that match their interests. However, selecting higher education choices such as a university major is a massive decision for students leading them, to surf the internet for long periods in search of needed information. Therefore, the purpose of this study is to assist high school students through a hybrid recommender system (RS) that provides personalized recommendations related to their interests. To reach this purpose we proposed a novel hybrid RS approach named (COHRS) that incorporates the Knowledge base (KB) and Collaborative Filtering (CF) recommender techniques. This hybrid RS approach is supported by the Case based Reasoning (CBR) system and Ontology. Hundreds of queries were processed by our hybrid RS approach. The experiments show the high accuracy of COHRS based on two criteria namely the accuracy of retrieving the most similar cases and the accuracy of generating personalized recommendations. The evaluation results show the percentage of accuracy of COHRS based on many experiments as follows: 98 percent accuracy for retrieving the most similar cases and 95 percent accuracy for generating personalized recommendations.

A Systematic Review on the Application of Ontologies to Improve Career Guidance

“…It was established that only 11 studies met the inclusion criteria. Of these 11 studies, 6 of them, that is (Alimam et al, 2014), (Obeid et al, 2018), (Obeid et al, 2022), (Dascalu et al, 2022), (Ibrahim et al, 2019), and (Abdellah et al, 2019) proposed recommendation systems to help students during the career pursuit. However, each of these studies had its shortcomings as indicated in Table 1, hence the need to address these shortcomings/loopholes by proposing an Ontology-Based Model (OBM) that would provide a one-stop center for all career guidance needs, by providing information to students from the time they choose a combination/major to the time they gain meaningful employment by providing them with the required skills that meet the needs of the 21 st -century world of work.…”

Could the Use of AI in Higher Education Hinder Students With Disabilities? A Scoping Review

“…In profiling and prediction applications, three types of decisions were identified: 1) identification of students at risk of dropping out of a course, study programme, or school [20], [42], [46], [47], [49], [51]- [63], 2) prediction of student performance prior to their admission [21], [64]- [67], 3) identification of students' learning profile [43], [50], [68]. The first two types of decisions have a more significant impact on access to studies than the last one.…”

“…This type of decision has a rather low impact as it simply provides a list of courses or materials that could be of interest to students. In comparison, five articles utilised AI to recommend higher education institutions or programmes to students [21], [64]- [67]. These recommendations can potentially impact students' decision to apply to a specific programme or university, impacting their future job and salaries.…”

“…Assessment is defined broadly as "the process of gathering information and intervening in that information using some criteria in order to form a judgment" [20, p.4]. For instance, an AI could assess a student's skills, interests, and preferences to recommend suitable universities [21].…”

A Recommendation System for Selecting the Appropriate Undergraduate Program at Higher Education Institutions Using Graduate Student Data

“…Then, the outcome of the CF was input into the KB recommendation system to recommend personalized suggestions based on a student's demographic and academic history. The study [22] utilized data on academic results, personality, and intelligence to select the appropriate major using a hierarchal classification approach; the first classifier was responsible for predicting the main streams, and another classifier (for each stream) predicted the subcategory of the major. Each classifier in this hierarchal model was trained using two classification algorithms, Random Forest and Multi-Layer Perceptron, in addition to 10-fold cross-validation to confirm the classification accuracy, which ranged from 89.29% to 96.10% using the RF and confirmed that the hierarchal model outperformed the flat one.…”