Astonjadro
Vol. 15 No. 4 (2026): ASTONJADRO

Sentiment Analysis of the Free Nutritious Meals Program using Linear Learner and Adaboost Algorithms

Muhammad Annash Ash Shidiq (INDONESIA)
Sriyanto Sriyanto (INDONESIA)



Article Info

Publish Date
23 Oct 2026

Abstract

The government initiative through the Free Nutritious Meals Program is designed to intervene in the nutritional quality of the community, especially for students in schools. Given that this policy has sparked widespread discussion on social media platforms, this study aims to map public opinion trends through sentiment analysis. The main focus of this study is to apply a Logistic Regression-based Linear Learner algorithm integrated with the AdaBoost ensemble learning technique to classify text data. Primary data was collected from the Threads platform, which was then processed through a series of systematic stages including text preprocessing, feature extraction, and separation between training and test data for positive and negative sentiment categories. In this experiment, Logistic Regression acts as the base classifier, while AdaBoost is used as a strategy to optimize classification accuracy. Model performance evaluation was conducted comprehensively using accuracy, precision, recall, and F1-score parameters. The research findings confirm that adding the AdaBoost method to the Logistic Regression framework significantly improves classification quality compared to when the algorithm is used alone. The developed model was proven to produce higher metrics and a more stable distribution of values. Therefore, the synergy between Logistic Regression and AdaBoost is a reliable approach to evaluate public perception of public policy implementation.

Copyrights © 2026






Journal Info

Abbrev

ASTONJADRO

Publisher

Subject

Civil Engineering, Building, Construction & Architecture

Description

Astonjadro: Jurnal Rekayasa Sipil adalah jurnal ilmiah dibidang teknik sipil p-ISSN : 2302-4240 dan e-ISSN : 2655-2086. Jurnal Rekayasa Sipil Astonjadro diterbitkan untuk mengapresiasi dan memberikan informasi ilmiah bagi peneliti, dosen serta para profesional. termasuk dalamnya pengembangan model ...