Smart Techno (Smart Technology, Informatics and Technopreneurship)

Predicting Crop Water Requirements Using IoT Sensor Data for Deep Learning

Saluky Saluky, Aisya Fatimah — Fri, 17 Oct 2025 13:53:43 +0000

The optimization of irrigation is a crucial factor in enhancing agricultural productivity and resource efficiency. This study proposes a deep learning-based approach to predict plant water requirements using data from IoT sensors. The system collects real-time environmental parameters such as soil moisture, temperature, humidity, and solar radiation, which are then processed using a deep learning model to generate accurate irrigation recommendations. The model is trained and evaluated on historical sensor data to ensure robustness and reliability in varying climatic conditions. The proposed method aims to minimize water wastage while maintaining optimal soil moisture levels, thereby improving crop health and yield. Experimental results demonstrate that the deep learning model outperforms conventional threshold-based irrigation systems in terms of prediction accuracy and water conservation. This research contributes to the advancement of smart farming by integrating IoT and artificial intelligence for precision agriculture.

Digital Marketing Strategy of Threads of Life Ubud in the Context of Local Culture

Nanda Perwira — Mon, 20 Oct 2025 00:00:00 +0000

This study aims to identify and analyze digital marketing strategies based on local culture in creative MSMEs, with a case study focus on Threads of Life in Bali. The background of the research is rooted in the significant role of creative MSMEs in the regional economy, particularly in Bali, which possesses rich cultural heritage as a primary resource. However, in the digital era, business actors face challenges such as limited digital literacy, resource constraints, and the need to preserve cultural authenticity in global marketing. The method used is a Systematic Literature Review (SLR) consisting of planning, literature searches in reputable databases (Scopus, Web of Science, Sinta, Google Scholar), study selection based on inclusion–exclusion criteria, data extraction, and content analysis to identify relevant strategy patterns. The findings reveal that Threads of Life successfully utilizes social media platforms such as Instagram and Facebook, as well as its official website, to develop brand storytelling that emphasizes traditional weaving, sustainability, and community empowerment. The integration of local cultural values such as mutual cooperation (gotong royong) and heritage preservation strengthens brand image and competitiveness in the global market. This study contributes novelty by combining perspectives of digital marketing, cultural preservation, and the creative industry within a single analytical framework. Research recommendations include expanding digital collaborations, developing interactive content, and strengthening sustainability narratives to enhance consumer engagement. These findings are relevant as strategic references for culture-based creative MSMEs seeking to optimize digital marketing without losing their local identity.

Clustering of Regencies and Municipalities Based on the Number of Livestock in East Java Province Using the Fuzzy C-Means Method

Intan Agnesa Salsabilla — Fri, 24 Oct 2025 04:24:11 +0000

This study was conducted to cluster regencies and municipalities in East Java Province based on the population of livestock, aiming to identify regional distribution patterns according to livestock characteristics. The clustering was performed using the Fuzzy C-Means algorithm and validated through the Partition Coefficient Index method. The implementation was carried out in a web-based application using the Laravel framework. The stages of this research included data collection, normalization, Fuzzy C-Means computation, evaluation using the Partition Coefficient Index, and profiling of cluster characteristics. The results of the study, tested with cluster numbers ranging from 2 to 10, indicated that the optimal number of clusters was two for both 2021 and 2022, with Partition Coefficient Index values of 0.7507 for 2021 and 0.7486 for 2022. In 2021, the optimal clustering produced Cluster 1
consisting of 7 regencies and 9 cities, and Cluster 2 consisting of 22 regencies. In 2022, the optimal clustering resulted in Cluster 1 consisting of 21 regencies, and Cluster 2 consisting of 8 regencies and 9 cities.

Digital Payment Integration in Accounting Information Systems to Support MSMEsRevitalization in Bali: A Literature Review

Gita Apsari Dewi, Dewa Gde Yoga Permana, A.A. Gde Agung Nanda Perwira — Sun, 19 Oct 2025 09:24:28 +0000

MSMEs are a cornerstone of Bali’s economy, relying heavily on local cultural strengths and tourism as the main attraction. The decline in tourist numbers due to the pandemic has necessitated adaptive revitalization strategies aligned with technological developments, including the digitalization of payments. Previous literature indicates that QRIS (Quick Response Code Indonesian Standard) can enhance transaction efficiency and MSME revenue; however, few studies have explored integrating these transaction data into Accounting Information Systems (AIS) to support strategic decision-making within the context of Balinese culture.This study employs a systematic literature review (SLR) to identify findings, gaps, and development opportunities from relevant studies published over the last five years. The analysis shows that integrating QRIS with AIS can produce accurate, real-time, and transparent financial data flows, facilitating reporting, cost control, and business planning. Such integration also has the potential to optimize the competitiveness of Bali’s tourism-based MSMEs by considering cultural factors, such as the Tri Hita Karana values and the banjar social structure, which influence technology adoption. The study concludes that successful implementation requires supporting infrastructure, digital literacy, and policies aligned with the local socio-cultural context.

Sentiment Analysis Of Comments On Indonesian Political Speech Videos On Youtube Using FastText

Bella Risma Khailla Savana, Deni Arifianto, Lutfi Ali Muharom — Sun, 19 Oct 2025 09:27:02 +0000

The advancement of digital technology has transformed how society accesses and responds to political information, particularly through platforms like YouTube, which serve as arenas for public discourse. Comments on political speech videos often contain complex sentiments such as irony, slang, and code-mixing, which are difficult to identify using traditional sentiment analysis methods. This study aims to analyze public sentiment toward the Indonesian President’s political speeches on YouTube from 2014 to 2024 using the FastText word embedding approach and to compare its performance with the TF-IDF + Logistic Regression method. The evaluation was conducted on three sentiment classes using automatically labeled data and oversampling experiments to address class imbalance. The results show that FastText achieved an accuracy of 76.82%, slightly higher than TF-IDF + Logistic Regression at 74.11%. Although the difference in accuracy is relatively small, the FastText model demonstrated more stable performance on informal texts and varied contexts. The use of oversampling helped balance predictions across classes without significantly improving accuracy. This study highlights the potential of FastText to enhance the effectiveness of Indonesian-language sentiment analysis, particularly for political comments on social media, while also revealing the limitations of automatic labeling that may affect classification outcomes.