Jurnal Sosioteknologi

Interpretasi Gambar Digital pada Warisan Budaya: Eksplorasi Bentuk Visual Melalui Deep Learning Studi Kasus Patung Megalitik Taman Nasional Lore Lindu

Sigied Himawan Yudhanto, Ovide Decroly Wisnu Ardhi

The megalithic statues in Lore Lindu National Park are one of priceless cultural heritage sites in Central Sulawesi, Indonesia. Advances in digital imaging technology and artificial intelligence have enabled new approaches that can improve our understanding and documentation of these megalithic artefacts. This study explores how deep learning can be applied to classify and interpret images of three statues in Lembah Bada: the Palindo Statue, the Tugu Statue of Bewa Village, and the Langke Bulawa Statue. We trained our deep learning classification model (YOLOv8-CLS) on each dataset and conducted interpretability analysis using EigenCAM to identify regions of interest in each class. The results show good to excellent precision, recall, F1 scores, and accuracy for all preprocessing methods, with heatmaps consistently highlighting culturally significant features such as facial contours and body posture. These findings indicate that deep learning models achieve high accuracy and align with human visual perception by emphasising visual forms embedded in the cultural memory of megalithic statue forms.

English English

Rayhan Muhammad Yasin, Iis Kurnia Nurhayari

Penelitian ini bertujuan untuk menganalisis strategi komunikasi digital yang dilakukan oleh Yayasan Odesa terhadap pengelolaan konten dan Pembangunan keterlibatan audiens di media sosial. Metode yang digunakan Adalah penelitian kualitatif dengan pendekatan studi kasus, melibatkan wawancara secara mendalam, memanfaatkan media sosial, Tingkat keterlibatan audiens sangat rendah, jenis konten yang masih sangat kurang interaktif, ketidaksesuaian dengan preferensi audiens, kurangnya konsustensi terhadap unggahan. Penelitian ini menjadi rekomendasi peningkatan kualitas konten, penggunaan fitur-fitur media sosial, serta konsistensi unggahan dalam meningkatkan efektivitas komunikasi digital.

English English

Iis Kurnia Nurhayari, Rayhan Muhammad Yasin

Students' Critical Thinking: Using the Resource Based Learning (RBL) Model Through Differentiated Learning

M Hafizul Furqan

The Resource Based Learning (RBL) learning model is a learning model that is oriented towards student independence in accessing and processing various learning resources according to their interests and learning styles. This study aims to determine the effect of implementing the Resource Based Learning learning model through differentiated learning on students' critical thinking skills in Geography subjects for class X of SMA Negeri 16 Banda Aceh. This study uses a quantitative approach with an experimental research type, and the research design used is Quasi Experimental Design with the form of One Group Pretest-Posttest Design. The sampling technique was carried out using the Purposive Sampling method. The instrument used was an essay test question that measures students' critical thinking skills. Data analysis techniques include normality, homogeneity, and hypothesis testing. Based on the calculation results, the t count value = 12.5 and t table = 1.729 were obtained. Because the t count value> t table, then Hₐ is accepted. The test results show that Hₐ is accepted, so it can be concluded that there is a positive effect of implementing the Resource Based Learning (RBL) learning model through differentiated learning on critical thinking skills in students of SMA Negeri 16 Banda Aceh

Balinese Small Entrepreneurs’ Self-Concept Behind the Adoption Levels of Social Media Innovation for Promotion

Indra Novianto Adibayu Pamungkas

This study aims to understand the Level of social media adoption for promotion and to identify the self-concept behind each stage among small business owners of the digital immigrant generation in Bali. The background of this research is the challenges small entrepreneurs face in adapting to the digital environment, where promotional issues are a significant obstacle. The digital divide and Indonesia's strong collective culture also influence the rate of innovation adoption. A qualitative approach was employed, involving interviews with 26 small business owners in Bali who are part of the digital immigrant generation, specifically Generation X (born between 1967 and 1980). Data were processed with NVivo 12 Pro software for analysis and visualization. The results show the highest adoption rate is at the Late Majority level (47%), followed by Early Adopters (19%), and Innovators and Laggards (14% each). There is a correlation between self-concept and adoption level. Early Adopters are characterized by initiative, hard work, and planning. Laggards are influenced by external factors, such as input from close family, and feel anxious about the rapid pace of technological development. The study concludes that innovation adoption is influenced by self-concept and collective culture. The Majority of entrepreneurs in the Late Majority stage show potential for adaptation. However, their process is often hindered by self-negotiation and dependence on social proof from those closest to them.