Mochammad Ilham Aziz, Atika Nur Fadilla, Anis Sholikhah, Saifulloh Azhar, Muhammad Oktoda Noorrohman

Analisis Algoritma Apriori untuk Mendukung Kesalahan Suku Kata pada Hasil Tes Literasi Siswa Sekolah Dasar

Introduction

Analisis algoritma apriori untuk mendukung kesalahan suku kata pada hasil tes literasi siswa sekolah dasar. Analisis algoritma Apriori menemukan pola kesalahan suku kata pada tes literasi siswa SD. Identifikasi hubungan kesalahan fonologis untuk strategi pembelajaran & intervensi tepat.

Abstract

Penelitian ini bertujuan untuk menganalisis pola kesalahan pengucapan suku kata pada hasil tes literasi siswa sekolah dasar dengan menerapkan algoritma Apriori sebagai salah satu teknik data mining dalam menemukan pola keterkaitan antar kesalahan. Melalui pendekatan kuantitatif deskriptif, data diperoleh dari hasil tes membaca siswa yang menunjukkan berbagai jenis kesalahan pada suku kata tertentu. Data kemudian diolah menggunakan algoritma Apriori untuk menghasilkan aturan asosiasi dengan memperhitungkan nilai support, confidence, dan lift guna mengidentifikasi hubungan antar kesalahan pengucapan yang paling signifikan. Hasil penelitian menunjukkan bahwa terdapat hubungan kuat antara kesalahan pengucapan beberapa suku kata, seperti Hon, De, dan Dak, dengan kesalahan pada suku kata Yam, yang memiliki nilai confidence di atas 50%. Temuan ini mengindikasikan adanya pola sistematis dalam kesalahan fonologis siswa yang dapat digunakan sebagai indikator kesulitan membaca dan mengenali bunyi bahasa. Kesimpulannya, penerapan algoritma Apriori efektif dalam mengungkap pola tersembunyi pada data kesalahan literasi siswa dan dapat menjadi alat diagnostik pendukung bagi guru untuk merancang strategi pembelajaran fonetik serta intervensi literasi yang lebih tepat sasaran.

Review

This study presents a timely and relevant investigation into the application of the Apriori algorithm for analyzing syllable pronunciation errors in elementary school students' literacy tests. By employing a data mining approach, the research aims to uncover hidden patterns and associations within error data, offering a novel perspective on diagnostic tools for early literacy challenges. The focus on identifying systematic phonological error patterns using a quantitative descriptive methodology is commendable, proposing a valuable contribution to both the fields of educational assessment and data science application in learning analytics. The methodology, utilizing the Apriori algorithm with consideration for support, confidence, and lift metrics, is well-suited for identifying significant association rules among different types of syllable errors. The abstract highlights a key strength in its specific findings, demonstrating a strong relationship between errors in pronouncing syllables such as "Hon," "De," and "Dak" and errors in "Yam," with a confidence level exceeding 50%. This empirical evidence powerfully supports the claim of systematic phonological difficulties, which can serve as critical indicators for reading acquisition challenges. The effectiveness of Apriori in revealing these intricate connections underscores its potential as a robust analytical tool in educational research. The practical implications of this research are significant, positioning the Apriori algorithm as an effective diagnostic aid for educators. The ability to identify specific error patterns can empower teachers to design more targeted phonetic learning strategies and tailored literacy interventions, moving beyond generic approaches. While the abstract effectively outlines the utility of this method, future work could benefit from detailing the specific context of the elementary schools, the demographic of the students, and the size of the dataset to fully appreciate the generalizability of these findings. Nevertheless, this study successfully demonstrates a promising avenue for leveraging data mining techniques to enhance literacy education, providing actionable insights for improving student learning outcomes.

Full Text

Comments

You need to be logged in to post a comment.