DESIGN AND DEVELOPMENT OF AN INTERNSHIP AND THESIS INFORMATION RETRIEVAL USING THE BM25-SBERT METHOD BASED ON THE AHO-CORASICK ALGORITHM

Yusaga, Chasetyo Ivan (2026) DESIGN AND DEVELOPMENT OF AN INTERNSHIP AND THESIS INFORMATION RETRIEVAL USING THE BM25-SBERT METHOD BASED ON THE AHO-CORASICK ALGORITHM. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) Cover.pdf Download (1MB)
	Text (bab 1) Bab 1.pdf Download (218kB)
	Text (bab 2) Bab 2.pdf Restricted to Repository staff only until 20 July 2029. Download (395kB)
	Text (bab 3) Bab 3.pdf Restricted to Repository staff only until 20 July 2029. Download (5MB)
	Text (bab 4) Bab 4.pdf Restricted to Repository staff only until 20 July 2029. Download (3MB)
	Text (bab 5) Bab 5.pdf Download (141kB)
	Text (daftar pustaka) Daftar pustaka.pdf Download (249kB)

Abstract

The process of submitting internship (PKL) and thesis titles in the Faculty of Law is still conducted manually, resulting in several challenges, such as lengthy verification times, the risk of title duplication, and a lack of transparency regarding submission status. In addition, title uniqueness checks are performed manually by administrators, which is time-consuming and prone to errors. To address these issues, this study designs and develops a web-based Internship and Law Thesis Information System (SIMAS-HU) that supports online title submission and provides a title search feature for internship (PKL) and thesis topics. This feature enables students to search for title references and helps prevent similarities or duplication of previously submitted titles. The system incorporates the BM25 algorithm to calculate relevance scores and rank search results based on keyword matching. The Sentence-BERT (SBERT) method is employed to identify semantic similarities between titles, even when different terms are used, thereby improving search accuracy. Furthermore, the Aho-Corasick algorithm is implemented to detect user-input keywords and enhance the scores of the most relevant search results. The integration of these three methods enables faster, more accurate, and more relevant information retrieval. In addition to the search functionality, SIMAS-HU provides facilities for internship (PKL) and thesis title submissions that are verified by academic supervisors, KPS administrators, and the Head of Study Program. The system also records student activity logs and generates approval documents in PDF format with automatically generated QR codes, helping ensure document authenticity and prevent forgery or misuse of approval documents. System testing was conducted using Black Box Testing to evaluate functional suitability, while search performance was assessed using the Normalized Discounted Cumulative Gain (NDCG), Mean Reciprocal Rank (MRR), and Recall metrics.

Item Type:

Thesis (Undergraduate)

Contributors: