Implementasi Sistem Rekomendasi Berbasis Konten untuk Rekomendasi Resep MPASI

Nurdini, Faraz (2025) Implementasi Sistem Rekomendasi Berbasis Konten untuk Rekomendasi Resep MPASI. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026201007-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (2MB) | Request a copy

Abstract

Periode pertumbuhan dan perkembangan (periode keemasan dan kritis) pada bayi usia 0-24 bulan membutuhkan perhatian khusus, terutama dalam pemilihan makanan pendamping Air Susu Ibu (MPASI). Pada usia 6-12 bulan, bayi mengalami masa transisi mengonsumsi makanan lunak dengan nutrisi gizi seimbang, dengan catatan bahwa bayi memiliki kerentanan dalam imunitas sehingga memerlukan perhatian khusus sesuai dengan pantangan alergen bayi. Salah satu tantangan utama dalam penyusunan MPASI adalah bagaimana cara mendeteksi tujuh alergen dan turunannya. Dengan itu, dirancang sistem deteksi alergen pada resep MPASI dengan mengimplementasikan content-based recommender system menggunakan metode feature based recipe similarity. Penelitian ini mengembangkan model yang dapat diimplementasikan dalam aplikasi berbasis web dengan open source framework Streamlit dan metode pengembanagan Rapid Application Development (RAD). Proses pengembangan model dan aplikasi meliputi pengumpulan data, praproses untuk menyaring informasi relevan, perhitungan similarity menggunakan feature based recipe similarity, serta filtering berdasarkan umur. Dengan ini, model dapat memberikan rekomendasi MPASI untuk bayi 6-12 bulan, terutama bagi yang memiliki pantangan alergen. Hasil pengujian dilakukan secara terpisah untuk model dan aplikasi, dimana hasil uji model dapat menyaring resep sesuai kombinasi alergen seperti alergi telur pada umur 6 bulan, alergi ikan pada umur 8 bulan, serta alergi keju dan susu beserta turunannya pada umur 9 bulan. Dari keempat hasil uji coba dengan kombinasi yang berbeda, similarity score yang didapatkan senilai 0.0. Akan tetapi, model belum dapat mengenal hubungan variabel serperti ‘keju’ sebagai bagian dari ‘susu’ secara otomatis. Kekurangan ini diperbaiki melalui pengujian aplikasi dengan tiga skenario; tanpa alergen, seluruh kombinasi alergen, dan beberapa kombinasi alergen. Hasil ketiga uji coba aplikasi menunjukkan bahwa aplikasi mampu memberikan 10 rekomendasi resep dengan similarity score senilai 0.0, memastikan MPASI yang direkomendasikan aman untuk konsumsi dengan pantangan alergen tertentu. Keseluruhan penelitian ini diharapkan dapat memudahkan orang tua dalam memilih MPASI agar dapat memenuhi kebutuhan gizi bayi secara optimal.
==================================================================================================================================
The growth and development (golden and critical period) of babies aged 0-24 months requires special attention, especially in selecting certain choices of complementary foods (MPASI). Due to their age of 6-12 months, they begin to experience a transition of eating soft-textured foods with balanced nutrition, noting that infants have vulnerabilities in immunity that requires special attention with allergen restriction. One of the main challenges is how to detect seven alergens and their derivatives in recipes. Therefore, an allergen detection system for complementary foods are needed and designed by implementing a content-based recommendation system using a feature based recipe similarity method. This research created a model that can be implemented into website application with open source framework Streamlit and the Rapid Application Development (RAD). The workflow begins with collecting complementary food’s data, through preprocessing to filter out irrelevant information, then computing the recipe similarity method and age filtering. Therefore, the model can provide recommendations for babies aged 6-12 months, especially for those who have allergen restrictions. The test were conducted separately for the model and application, where the model test results can filter recipes according to allergen combinations, such as egg allergy for babies aged 6 months, fish allergy for babies aged 8 months, and allergy of cheese and milk and its derivatives for babies aged 9 months. From the test we can conclude that the similarity score obtained is 0.0. However the model couldn’t recognize the relation between ‘cheese’ variable as part of ‘milk’ automatically. This drawback are corrected through the application testing with three test scenarios; without allergens, all allergen combinations, and some allergen combinations. The result of three tests showed that the application are able to provide 10 recipe recommendations with similarity score of 0.0, ensuring the recommended recipe are safe for consumption with certain allergen restrictions. The whole research is hopefully able to facilitate parents in choosing complementary food to fulfill the baby’s nutritional needs.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sistem Rekomendasi, Alergi Makanan, Content-based, MPASI, Recommender System, Food Allergy
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Faraz Nurdini
Date Deposited:	17 Jan 2025 07:23
Last Modified:	17 Jan 2025 07:23
URI:	http://repository.its.ac.id/id/eprint/116382

Actions (login required)

View Item