This work offers a robust decision-support framework for higher education institutions, accreditation bodies, and employers by systematically linking graduates’ knowledge, skills, and behavioral attributes to technical demand by systematically linking graduates’ knowledge, skills, and behavioral attributes to technical demand.
Abstract
Assessing graduate students’ academic performance is a multifaceted task that requires consideration of both quantitative metrics and qualitative evaluations. Conventional assessment approaches rely on numerical scores and strict grading schemes, often neglecting critical soft factors such as student behavior, classroom participation, learning consistency, and extracurricular engagement. To address the ambiguity and imprecision inherent in human-centric evaluation, this work proposes an Explainable Artificial Intelligence (XAI) approach to produce a transparent, interpretable estimate of graduates’ technical demand. In the first phase, a fuzzy logic inference system was developed that considers academic scores, project outcomes, research productivity, the desired graduate outcome index, and qualitative feedback to produce a comprehensive assessment of graduate performance. The fuzzy model was developed using expert insights, supported by empirical analysis of anonymized academic data from graduates of the 2023-25 cohorts. The fuzzy system was developed using both Python and MATLAB and later deployed as a web-based application to provide real-time decision support for mapping educational outcomes to graduate employability. In the second phase, validated data from 1,875 students were used to build an explainable machine learning model using the no-code KNIME analytics platform. A Random Forest classifier, augmented with SHAP-based explainability, was employed to provide both global and local interpretations of student performance. The proposed model achieved an accuracy of 99.57% with Cohen’s kappa of 0.933, demonstrating strong predictive reliability and interpretability. This work offers a robust decision-support framework for higher education institutions, accreditation bodies, and employers by systematically linking graduates’ knowledge, skills, and behavioral attributes to technical demand.
Scholarship programs are essential for increasing access to higher education, yet selection processes are frequently hindered by subjective reasoning and procedural inefficiencies. This study presents an advanced Decision Support System (DSS) for academic achievement scholarships at Universitas Islam Majapahit using a modified Fuzzy Tahani method characterized by a dual-platform architecture, socio-economically tailored membership functions, and complete operational transparency. The model evaluates four primary criteria—Grade Point Average (GPA), parental income, number of dependents, and non-academic achievements—to provide objective prioritization through fuzzification and defuzzification processes. Testing conducted on 18 applicants across Microsoft Excel and Android platforms demonstrated 100% computational accuracy and a Spearman’s rank correlation of 1.0, confirming identical ranking consistency. Furthermore, the system exhibited high efficiency with processing times of 35 seconds on Excel and 2 seconds on Android, while sensitivity analysis and a 94% expert validation rate underscored its procedural robustness. The results indicate that students possessing high academic merit alongside significant financial need were consistently prioritized, addressing previous systemic deficiencies in fairness and transparency. Ultimately, this modified Fuzzy Tahani framework offers a reliable and validated multi-criteria decision-making system that other higher education institutions can adapt to ensure more equitable scholarship allocations.
Feriyanto, Subanji, Sukoriyanto et al.· Journal of Physics, Conferen...· 0 citations
Uncertainty in instructors' judgments and qualitative evaluations is difficult to capture in traditional student performance evaluations centered on descriptive statistics and averages. While traditional fuzzy systems are capable of handling this kind of ambiguity, they become less efficient and harder to understand as the number of evaluation variables rises since their rule bases are exponentially larger. Exam results may be used to evaluate students' performance, and this research suggests a genetic-fuzzy architecture with rule-base reduction that can be understood. The model was trained in MATLAB using the following parameters: 50 nodes, 25 generations, 0.8 crossover probability, 0.2 mutation rate, and roulette-wheel selection; Mamdani inference; centroid defuzzification; and a genetic algorithm. The non-optimized fuzzy method's MAE is 0.0208 and the genetic-fuzzy method's MAE is 0.0101, according to recalculation utilizing the 25 records shown in Table 1. These values are unique to the descriptive versus creative 25-record comparison and should not be used interchangeably with the separate function membership creative sensitivity results from Table 4. The original basic set of 125 rules was pared down to 46 rules. The leakage-free train-validation-test methodology is defined in this version, and final generalization estimations must be given from it.
Zainab Hammoodi Noori· Journal of Al-Farabi for Eng...· 0 citations
Educational assessment systems conventionally rely on crisp numerical scores that inadequately capture the gradual and uncertain nature of student learning, and while fuzzy inference systems (FIS) have been applied to address this limitation, prior studies rarely compare alternative membership function designs under identical experimental conditions, leaving open the practical question of which membership function type best supports fuzzy-based educational assessment. This study addresses that gap by presenting a comparative evaluation of Gaussian, Triangular, and Trapezoidal membership functions, which were selected because they are among the most widely used membership functions in fuzzy educational assessment and differ in their smoothness, computational simplicity, and plateau characteristics within a Mamdani Fuzzy Inference System (FIS) for assessing student learning achievement. Three fuzzy models were developed using examination score, assignment completion, and class participation as input variables, while learning achievement served as the output variable. A dataset comprising 360 secondary school students was used for model development and validation. Teacher evaluation scores were employed as benchmark values for assessing model performance. The three models were implemented using identical rule bases, inference mechanisms, and defuzzification procedures, differing only in the type of membership function employed. Model performance was evaluated using the Pearson correlation coefficient, root mean square error (RMSE), and mean absolute error (MAE). The results revealed that the Gaussian model achieved the highest correlation with teacher evaluations (r = 0.9623), indicating superior agreement with teacher assessment patterns, whereas the Triangular model produced the lowest RMSE (7.194) and MAE (4.867), demonstrating greater numerical prediction accuracy.
Abubakar Audu, Adeku Ibrahim Musa, Kehinde Rotimi· FUDMA Journal of Sciences· 0 citations
The informational leadership of college instructors for English teaching is becoming increasingly important for modern schooling, but its evaluation is somewhat non-rigorous due to the unclear indicator system and the unrestrained expert appraisal. To this end, a new neutrosophic modeling method combining fuzzy expression and MCDM is proposed to establish a rigorous indicator system and determine the key factors. It consists of three parts: establishing and verifying the indicator system; coding experts' comments into the neutrosophic-fuzzy format to express their support, rejection, and indeterminacy of the indicators (and multiple hesitant values); and applying a correlation-based MCDM approach for ranking against the ideal. One complete case study, including stepwise computations, a comparative study with the conventional fuzzy MCDM, and a sensitivity analysis, is presented. The results produce a stable ranking with acceptable perturbations and provide transparent core indicators and driving factors for education decision-makers.
Na Zhang· International Journal of Agr...· 0 citations
A hierarchical student evaluation enhanced decision-making model which is based on a zero-order fuzzy classifier as the construction unit and achieves the collaborative optimization of evaluation efficiency and interpretability is proposed.
Aiqin Wang, Wenliang Li· Journal of Advanced Computat...· 0 citations
Traditional youth badminton talent selection often fails to capture the uncertainty and subjectivity of expert evaluations. This research proposes a hybrid intuitionistic fuzzy (IF) multi-attribute decision-making framework in which the IF analytic hierarchy process is applied to criterion weights, and the IF technique for order preference by similarity to ideal solution (IF-TOPSIS) is applied to athlete ranking. The framework was evaluated on 40 elite junior badminton athletes. Validation using Wilcoxon signed-rank, Friedman, Pearson correlation, and Kendall's concordance tests demonstrated that the proposed IF multi-attribute decision-making model achieved strong agreement with expert rankings. The model accuracy was 0.91, the ranking stability was 97.4%, and the closeness coefficient variance was 0.0079, indicating high performance under uncertainty. The comparative analysis demonstrated that the proposed method achieved better ranking consistency and uncertainty handling than the analytic hierarchy process, TOPSIS, and fuzzy TOPSIS.
Shi-Rong Wu, Hui-Jing Xiao, Wenqi Hu· International Journal of Cog...· 0 citations