Recommendation model for student loan consideration

Name: Klangwaree Chaiwut

Address: 333 Moo1, Thasud, Muang, Chiang Rai 57100

Organization : Mae Fah Luang University. School of Information Technology. Computer Engineering

Email : ask.library@mfu.ac.th

keyword: Student loan

LCSH: Fuzzy logic

Classification :.LCCS: QA278.5

LCSH: Cluster analysis

LCSH: Factor analysis

LCSH: Student loans

Abstract: The objective of this dissertation is to improve the performance of student loan approval process in order to provide study opportunities to the needy students. In this dissertation, first of all, this study has to reveal a fault of traditional consideration process by identifying a mixing of both accepted and rejected applicants in the same group of cluster. This process designs an efficient mechanism to evaluate the fault in student loan approval process using the clustering technique with the concrete number of cluster. In this stage of the dissertation, K-mean was used to analyze the historical data sets of student loan candidates provided by anonymous universities, collected between 2014 and 2015. There are 1,146 samples within two years of application. The result shows that, with k=2, K-mean performs its best partitioning to categorize the results by validating with three well-known clustering measurement indexes. It also performs the best when comparing the overall average distance, at 4.41, with the highest density around the centroid of the cluster. This proposed model can identify the error in using the traditional selection criteria for the student loan fund granted and leads to a redesigning of the student loan approval criteria. In the second stage of this dissertation, an importance of student loan consideration criteria redesigning is revealed, regarding the fault in evaluating the applicants from the first stage. The historical data of student loan candidates elicited from their application form in the 2016 academic year was collected and analyzed by using Factor Analysis. There are 507 samples with 17 formal attributes designed by the government. The factor analysis reduced the dimensions of the variance in the samples by identifying the discriminative factors for student loan consideration. The experimental result shows that only nine factors were identified as discriminative factors, which are (1) Part-time job taken by the student, (2) Other scholarships that the student had been receiving, (3) Father’s salary, (4) Family ownership of the land, (5) House rental expense, (6) Number of siblings in the family, (7) Number of siblings currently studying, (8) Amount of money that the student gets from other scholarships, and (9) Parental Marital Status. The clustering technique was used to evaluate the group of important factors reduced from the factor analysis. The final stage of dissertation applies the Artificial Intelligence and the Multi Attribute Decision Making on the newly redesigned criteria, to support the student loan approval and improve the quality of the decision making for selecting the student loan candidates to have a scholarship. In this process, a Fuzzy TOPSIS was used to construct the final criteria by optimizing these nine factors based on the experts’ weighting scores for suggesting the SLF committee to choose the right needy students who suffering from the financial issues. The data was collected from the 2017 academic years of application forms for student loan fund which are 10 samples. Three experts who have more than five years of experiences in student loan consideration were invited to the study. The optimization process was done by assignment rating of the samples (alternatives) and the criteria by these experts. Then, a fuzzy number was applied for each criteria and alternative, follows by the creation of weight normalized fuzzy decision matrix for each criteria and alternative. At this point, the TOPSIS was used to calculate the positive and negative ideal, before ranking of the selection results as the final criteria. By using this optimized student loan criteria, the student loan fund committees could select the right students who really need the scholarship for enhancing their future.

Mae Fah Luang University. Learning Resources and Educational Media Center

Address: CHIANG RAI

Email: library@mfu.ac.th

Name: Worasak Rueangsirarak

Role: Advisor

Email : ask.library@mfu.ac.th

Name: Roungsan Chaisricharoen

Role: Co-Advisor

Email : ask.library@mfu.ac.th

Created: 2019

Issued: 2563

Modified: 2020-03-02

Issued: 2020-03-02

วิทยานิพนธ์/Thesis

application/pdf

CallNumber: Thesis QA278.5 K633r 2019

eng

DegreeName: Dotor of Philosophy

Level: Doctoral Degree

Descipline: Computer Engineering

Grantor: Mae Fah Luang University

©copyrights Mae Fah Luang University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	129593.pdf	3.12 MB	15	2025-07-11 18:43:55

ใช้เวลา

0.039961 วินาที