Claim Missing Document
Check
Articles

Found 1 Documents
Search

Implementation of Content-Based Filtering for Recommending Traditional Lombok Foods Isti Nurfadiyah; Lalu Mutawalli; Wafiah Murniati
Jurnal IPTEK Bagi Masyarakat Vol 6 No 2 (2026)
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/j-ibm.v6i2.1958

Abstract

This study develops a content-based filtering (CBF) recommendation prototype for traditional Lombok foods using Google Maps reviews. Data from 304 culinary locations and 1,246 reviews were processed using text preprocessing, term frequency–inverse document frequency (TF-IDF), and cosine similarity. A leave-one-out evaluation involved 274 eligible locations and recommendation list sizes of K = 5, 10, and 20. At K = 5, precision, recall, F1-score, and accuracy were 0.2781, 0.1118, 0.1211, and 0.8752, respectively. At K = 20, recall reached 0.3238 and F1-score 0.2154, while accuracy decreased to 0.8498. One tested query reached a cosine similarity score of 0.3775. The Python-based prototype presents five ranked culinary recommendations with similarity scores, TF-IDF terms, and location information. Relevance was assessed using shared culinary categories rather than direct user judgments; therefore, usability and real-world recommendation quality require further evaluation.