Claim Missing Document
Check
Articles

Found 22 Documents
Search

Analyzing the #KaburDuluAja Phenomenon as an Indicator of Brain Drain in Indonesia: A Hybrid Text Mining Approach Meilinda Putri, Ulfa; Margono, Hendro
Jurnal Ilmiah Sumber Daya Manusia Vol 9 No 2 (2026): JENIUS (Jurnal Ilmiah Sumber Daya Manusia)
Publisher : Universitas Pamulang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32493/JJSDM.v9i2.59252

Abstract

In recent years, an increasing number of Indonesian youth have expressed a desire to seek better opportunities and life prospects abroad. One of the hashtags widely used to articulate this aspiration is #KaburDuluAja. This hashtag has emerged as an expression of emotions, anxiety, and even frustration among certain segments of society, particularly the younger generation, in response to the current social, economic, and political conditions in Indonesia. This study aims to comprehensively examine the phenomenon of public discourse on social media related to the #KaburDuluAja hashtag as a reflection of the potential for brain drain in Indonesia. To achieve this objective, a hybrid approach is employed, integrating Clustering, Association Rule Mining, and Sentiment Classification methods. The research adopts a quantitative approach supported by text mining techniques and social media data analysis. Furthermore, this study is both descriptive and predictive in nature. The research does not focus on a specific geographical location; rather, it examines public conversations on social media platforms, particularly Twitter (currently known as X). The data utilized in this study were collected through a web crawling process. The findings indicate that public conversations related to the #KaburDuluAja phenomenon can be categorized into five main clusters. The evaluation of several classification models reveals that the Naïve Bayes algorithm achieves the highest accuracy in predicting sentiment, reaching 99.85%. The K-Nearest Neighbor (KNN) model achieves an accuracy of 75.79%, while the Decision Tree and Random Forest models demonstrate relatively lower performance, with accuracy levels around 60%. The #KaburDuluAja phenomenon is not merely a form of humor or casual expression on social media; rather, it represents a genuine reflection of social unrest, particularly concerning employment opportunities, welfare conditions, and the future prospects of Indonesia’s younger generation.
Implementation of Archives Digitization as a Measure to Preserve and Enhance Access to Information Lestari, Nur Laili Indah Fuji; Anggraini, Handari Niken; Margono, Hendro
Bubungan Tinggi: Jurnal Pengabdian Masyarakat Vol 8, No 1 (2026): FEBRUARY 2026
Publisher : Universitas Lambung Mangkurat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20527/btjpm.v8i1.17663

Abstract

Archive digitization is a strategic step in modern information management, especially in response to the challenges of preserving physical documents and the increasing public demand for fast, efficient, and easily accessible information. Dinas Perpustakaan dan Arsip Daerah (DPAD) Daerah Istimewa Yogyakarta (DIY) is one of the institutions responsible for managing regional archives and has implemented digitization as part of the archive preservation process. This activity aims to examine the implementation of community service activities through field study practice at DPAD DIY, focusing on archiving institutional challenges and the application of digital technology in archive management. The methods used in this activity refer to the service-learning model and include observation and interviews, preparation, implementation, and evaluation. The evaluation reviews the process and results, highlighting effective aspects and identifying any obstacles or shortcomings. The instruments used in the evaluation process include observation and interview results, field notes, documentation, and literature review. The results show that the activity contributed directly to archive digitization, including 4-5 audiovisual recordings per day, more than 1,000 photo archives in 167 digital files within six working days, and one box of textual archives consisting of 37 files converted into digital format. These outputs support better archive preservation and improve access to information. However, several obstacles were also found, such as limited, outdated equipment, insufficient human resources compared to the volume of archives, and inadequate workspace conditions. Nevertheless, this activity made a real contribution to the digitization process and produced recommendations for improvements to support the sustainability of the archive media transfer program at DPAD DIY.