View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Computer Science
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Computer Science
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Analisis Frame dan Klasifikasi Judul Berita Media Indonesia terkait Genosida Palestina menggunakan Latent Dirichlet Allocation dan Deep Learning

      Thumbnail
      View/Open
      Cover (1.375Mb)
      Fulltext (4.463Mb)
      Lampiran (882.5Kb)
      Date
      2025
      Author
      Danuningrat, Salma Nadhira
      Ridha, Ahmad
      Wahjuni, Sri
      Metadata
      Show full item record
      Abstract
      Identifikasi frame yang terkandung dalam pemberitaan konflik Palestina-Israel oleh media Indonesia memiliki peran penting dalam membentuk persepsi publik dan menyaring bias informasi. Penelitian ini bertujuan untuk menganalisis frame dan mengklasifikasikan judul berita dari enam media online Indonesia dengan pendekatan Natural Language Processing (NLP). Hasil pemodelan topik dengan Latent Dirichlet Allocation (LDA) digunakan untuk melabeli data, yang selanjutnya diklasifikasi dengan model deep learning LSTM, Bi-LSTM, GRU, Bi-GRU, dan IndoBERT. Hasil penelitian menunjukkan bahwa dari delapan label klasifikasi yang dikembangkan dari enam topik hasil LDA, IndoBERT memiliki kinerja terbaik dengan nilai F1 rata-rata makro tertimbang sebesar 81,20% sementara model RNN mengalami overfitting pada data berlabel otomatis hasil LDA. Penelitian ini menegaskan potensi metode berbasis Transformer dalam memahami konteks pada data serta relevansi pendekatan NLP dalam merepresentasi isu dalam pemberitaan media.
       
      The identification of frames in the coverage of the Palestine-Israel conflict by Indonesian media plays an important role in shaping public perception. This study aims to analyze frames and classify news titles from six Indonesian online media outlets using Natural Language Processing (NLP). Data labeled from the results of topic modeling with Latent Dirichlet Allocation (LDA) are then classified using LSTM, Bi-LSTM, GRU, Bi-GRU, and IndoBERT. The results of this study show that of the eight classification labels developed from the six topics provided by the LDA results, IndoBERT performs the best with a weighted macro-averaged F1-score of 81.20% while the RNN model experiences overfitting on automatically labeled data from LDA. This study confirms the potential of Transformer-based methods in understanding context in data and the relevance of NLP in representing issues in media reporting.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/168931
      Collections
      • UT - Computer Science [88]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository