On-task behavior problems are a common obstacle in the academic activities of school-aged children. Children often experience difficulty maintaining attention, are easily distracted, and exhibit frequent shifts of attention during learning. This has the potential to lead to decreased academic achievement and increased impulsive behavior. This study aims to determine the effectiveness of the shaping behavior modification technique in increasing the duration of on-task behavior. The subject of the study was a 4-year-old boy who exhibited attention problems while working on academic tasks. The research design used a one-group pretest-posttest design with a single-subject approach using an A-B-A data collection model. Data collection was carried out using a duration data sheet to measure the duration of on-task behavior through direct observation. Data analysis was conducted by comparing the average duration of on-task behavior before and after the intervention. The results showed that the shaping technique did not provide a significant increase in the duration of on-task behavior. Although there were fluctuations in the duration of increases and decreases during the intervention phase, these results were not sufficient to prove the effectiveness of shaping in this context. These findings indicate that the shaping approach is less than optimal when applied alone, and that additional intervention methods or other alternatives are needed that are more appropriate to the characteristics and needs of the child. This research provides implications for the importance of selecting appropriate and adaptive behavioral intervention techniques to help children improve focus and engagement in academic activities.
 

This study aims to describe the level of resilience in former drug addicts and analyze its influence on their recovery and social reintegration process. Using a descriptive quantitative approach, data were collected through a questionnaire based on Grotberg’s theory consisting of 48 items. Five participants who had abstained from drugs for approximately two years were involved in this research. The results showed that most participants had high to very high levels of resilience. External factors such as family support and economic conditions, as well as internal factors like self-awareness and future-oriented motivation, significantly contributed to increased resilience. This study emphasizes the crucial role of resilience in helping former drug addicts endure challenging times and prevent relapse.

Indonesia is a multicultural country facing serious challenges related to intolerance, social conflict, and identity polarization. This study aims to examine the strategic role of Civic Education as a means of strengthening national insight in building a tolerant society. Using a qualitative approach through literature review methods, this research analyzes various relevant sources on education, tolerance, and nationalism. The results show that Civic Education contributes to shaping students’ democratic character, inclusiveness, and appreciation of diversity. It not only transfers civic knowledge but also instills social skills and human values. However, challenges in implementation remain, such as cognitively-centered curricula and the negative influence of digital technology. Therefore, a contextual learning approach that emphasizes real-life experiences and active student participation is needed. The findings indicate the importance of synergy between schools, families, and communities in creating a conducive environment for nurturing tolerant attitudes. The implications of this research highlight the need to revitalize Civic Education as a key medium for building a peaceful, inclusive, and civilized nation amid complex social dynamics.

Diabetes adalah ancaman utama bagi kesehatan penduduk dunia yang saat ini merupakan penyebab utama kematian pada penduduk dunia yang berusia kurang dari 60 tahun. Dengan menggunakan Machine Learning diharapkan mampu memprediksi diabetes. Dengan menggunakan dataset Pima Indians Diabetes (PIMA Dataset). Pada penelitian ini dilakukan pengujian dengan menggunakan 2 Algoritma dan 1 Algoritma yang dioptimasi. Pengujian Pertama Menggunakan Support Vector Machine (SVM), Pengujian Kedua menggunakan Deep Neural Network (DNN)  dan Pengujian Ketiga menggunakan DNN yang dikombinakan dengan Particle Swarm Optimize (PSO). Pemilihan data yang digunakan sebagai training dilakukan dengan menggunakan Non-Random Sampling. Dalam Penelitian ini pengujian pertama dengan menggunakan SVM dengan melakukan pengujian tanpa menggunakan kernel dan menggunakan kernel Linear, Sigmoid, Polynomial dan Radial Basis Function (RBF). Untuk Pengujian Kedua dilakukan dengan menggunakan DNN tanpa menggunakan Optimaliasi atau DNN original dengan dilakukan penggujian dengan jumlah hidden layer 2 dan jumlah neuron 8 sampai 10 pada setiap hidden layer. Pengujian Ketiga dilakukan dengan menggunakan DNN yang dioptimalisasi dengan menggunakan PSO. Pada Pengujian Ketiga dilakukan penggujian dengan jumlah hidden layer 2 dan jumlah neuron 8 sampai 10 pada setiap hidden layer. Hasil yang diperoleh menunjukkan bahwa DNN yang dioptimasi dengan PSO mampu memberikan akurasi tertinggi dengan jumlah hidden layer 1 sebanyak 9 node dan jumlah hidden layer 2 sebanyak 8 node dengan jumlah iterasi pada PSO sebanyak 166 iterasi.