Sistem Chatbot Informasi PPDB Berbasis TF-IDF dan Cosine Similarity Menggunakan Adaptive Threshold pada Web Sekolah

Raditya Mayesha Beldyq; Rika  Apriani

doi:10.58602/chain.v4i4.349

Raditya Mayesha Beldyq ^{(Corresponding Author)} Universitas Bina Insani
Rika Apriani Universitas Bina Insani

DOI: https://doi.org/10.58602/chain.v4i4.349

Keywords: Chatbot, Cosine Similarity, Natural Language Processing, Term Frequency-Inverse Document Frequency, Text Preprocessing

Abstract

Layanan informasi Penerimaan Peserta Didik Baru (PPDB) di SMP Sejahtera 2 Cileungsi masih dilakukan secara manual sehingga menyebabkan keterlambatan respons dan penumpukan pertanyaan dari calon pendaftar. Penelitian ini bertujuan merancang dan mengembangkan sistem chatbot berbasis web yang mengimplementasikan pipeline Natural Language Processing (NLP) secara penuh, mencakup Text Preprocessing, pembobotan Term Frequency–Inverse Document Frequency (TF-IDF), dan pencocokan intent menggunakan Cosine Similarity. Tahapan Text Preprocessing meliputi case folding, cleansing, tokenization, normalisasi sinonim, stopword removal, stemming, dan ekstraksi bigram. Knowledge base disusun dari 32 intent, 655 pola pertanyaan, dan 46 jawaban yang diperoleh melalui wawancara dan observasi di SMP Sejahtera 2. Sistem dikembangkan menggunakan metode Rapid Application Development (RAD) dengan arsitektur serverless berbasis Firebase Firestore dan Vanilla JavaScript sehingga seluruh komputasi TF-IDF dan Cosine Similarity dijalankan pada sisi klien tanpa backend khusus. Adaptive threshold diterapkan secara dinamis berdasarkan keberadaan kata kunci domain PPDB untuk meningkatkan relevansi jawaban. Evaluasi akurasi model dilakukan terhadap 35 sampel pertanyaan uji yang dibagi dalam tiga kategori: in-pattern, out-of-pattern, dan out-of-domain, menghasilkan akurasi keseluruhan sebesar 82,9%. Pengujian fungsional menggunakan Black Box Testing terhadap 8 skenario seluruhnya dinyatakan valid. Evaluasi latensi menunjukkan rata-rata waktu respons chatbot sebesar 1,05 detik. User Acceptance Test (UAT) terhadap 25 responden memperoleh rata-rata kepuasan 87,4% dengan kategori Sangat Baik. Hasil penelitian membuktikan bahwa implementasi pipeline NLP berbasis TF-IDF dan Cosine Similarity mampu meningkatkan kualitas layanan informasi PPDB secara otomatis, cepat, dan dapat diakses kapan saja.

References

S.-C. Necula, F. Dumitriu, and V. Greavu-Șerban, “A Systematic Literature Review on Using Natural Language Processing in Software Requirements Engineering,” Electronics, vol. 13, no. 11, p. 2055, May 2024, doi: 10.3390/electronics13112055.

E. Grassini, M. Buzzi, B. Leporini, and A. Vozna, “A systematic review of chatbots in inclusive healthcare: insights from the last 5 years,” Univers. Access Inf. Soc., vol. 24, no. 1, pp. 195–203, Mar. 2025, doi: 10.1007/s10209-024-01118-x.

S. Shah, H. Ghomeshi, E. Vakaj, E. Cooper, and S. Fouad, “A review of natural language processing in contact centre automation,” Pattern Anal. Appl., vol. 26, no. 3, pp. 823–846, Aug. 2023, doi: 10.1007/s10044-023-01182-8.

W. S. Nsaif, H. M. Salih, H. H. Saleh, and B. T. Al-Nuaim, “Conversational Agents: An Exploration into Chatbot Evolution, Architecture, and Important Techniques,” Eurasia Proc. Sci. Technol. Eng. Math., vol. 27, pp. 246–262, Aug. 2024, doi: 10.55549/epstem.1518795.

Kementerian Pendidikan dan Kebudayaan Republik Indonesia, “Peraturan Menteri Pendidikan dan Kebudayaan Nomor 1 Tahun 2021 tentang Penerimaan Peserta Didik Baru,” Kementerian Pendidikan dan Kebudayaan, 2021. Accessed: Apr. 28, 2026. [Online]. Available: https://peraturan.bpk.go.id/Details/163568/permendikbud-no-1-tahun-2021

R. Suci, Ismarmiaty, and R. Rismayati, “SISTEM INFORMASI PENERIMAAN SISWA BARU BERBASIS WEBSITE SDN 3 GERUNG UTARA,” JATISKOM J. Apl. Teknol. Inf. Dan Sains Komput., vol. 1, no. 2, pp. 134–146, Dec. 2024, doi: 10.20414/jatiskom.v1i2.12689.

E. Mursidah, L. Ambarwati, and F. A. Karima, “IMPLEMENTASI CHATBOT LAYANAN INFORMASI PENDAFTARAN MAHASISWA BARU PROGRAM PASCASARJANA DEPARTEMEN TEKNIK INFORMATIKA ITS,” Netw. Eng. Res. Oper., vol. 7, no. 1, p. 43, Apr. 2022, doi: 10.21107/nero.v7i1.276.

A. Jalili, H. Tabrizchi, A. Mosavi, and A. R. Varkonyi-Koczy, “Enhancing Language Model Performance with a Novel Text Preprocessing Method,” Acta Phys. Pol. A, vol. 146, no. 4, pp. 542–552, Oct. 2024, doi: 10.12693/APhysPolA.146.542.

A. A. Nafea et al., “A Brief Review on Preprocessing Text in Arabic Language Dataset: Techniques and Challenges,” Babylon. J. Artif. Intell., vol. 2024, pp. 46–53, May 2024, doi: 10.58496/BJAI/2024/007.

M. Das, Selvakumar Kamalanathan, and Pja Alphonse, “A Comparative Study on TF-IDF feature Weighting Method and its Analysis using Unstructured Dataset,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 5, Aug. 2023, doi: https://doi.org/10.48550/arXiv.2308.04037.

L. Zhang, “Features extraction based on Naive Bayes algorithm and TF-IDF for news classification,” PLOS One, vol. 20, no. 7, p. e0327347, Jul. 2025, doi: https://doi.org/10.1371/journal.pone.0327347.

J. Halim and D. Lasut, “Document Plagiarism Detection Application Using Web-Based TF-IDF and Cosine Similarity Methods: English,” Bit-Tech, vol. 7, no. 2, pp. 202–213, Dec. 2024, doi: 10.32877/bt.v7i2.1697.

M. B. Arrisalah, M. M. A. Haromainy, and A. Junaidi, “Design of Thesis Topic Recommendation System Using TF-IDF and Cosine Similarity,” J. Inf. Syst. Technol., vol. 8, no. 2, Apr. 2026.

M. R. A. Prasetya and A. M. Priyatno, “Dice Similarity and TF-IDF for New Student Admissions Chatbot,” RIGGS J. Artif. Intell. Digit. Bus., vol. 1, no. 1, pp. 13–18, Jul. 2022, doi: 10.31004/riggs.v1i1.5.

R. H. Saputra and H. Sucipto, “Implementation of a WhatsApp-based Canteen Menu Chatbot using TF– IDF and cosine similarity: A case study of employees at Matahari Department Store, Sunrise Mall and Mojokerto,” vol. 9, no. 1, 2026.

E. Aprianto, D. Mahdiana, and A. Wibowo, “Optimizing Bag of Words and Word2Vec with Vocabulary Pruning and TF-IDF Weighted Embeddings for Accurate Chatbot Responses in Indonesian Treasury Services,” J. Tek. Inform. Jutif, vol. 7, no. 1, pp. 587–605, Feb. 2026, doi: 10.52436/1.jutif.2026.7.1.5370.

R. M. Holis, P. E. P. Utomo, and B. F. Hutabarat, “Semantic FAQ Chatbot Using SBERT (Sentence-BERT) and Cosine Similarity for Academic Services,” Brill. Res. Artif. Intell., vol. 5, no. 2, pp. 915–922, Oct. 2025, doi: 10.47709/brilliance.v5i2.7027.

Mohammed Salah Ibrahim, Anas Diab Sallibi, Ahmed Adnan Hadi, and Ahmed Adil Nafea, “A Chatbot for Frequently Asked Questions using TF-IDF and Query Expansion Techniques,” Passer J. Basic Appl. Sci., no. 1, Jun. 2025, doi: 10.24271/psr.2025.487515.1805.

D. Muazarah, B. Fachri, and A. Khaliq, “Web-Based New Student Admission System using the Rapid Application Development (RAD) Method at Penraujan Private Junior High School,” vol. 16, no. 02, 2026.

W. J. T. Subroto, Lady Silk Moonlight, and Hadi Prayitno, “Implementation of the Rapid Application Development (RAD) Method for the Information System of Sultan Babullah Ternate Airport Facilities,” J. Syst. Inf. Technol., pp. 27–34, 2025.