JOIV : International Journal on Informatics Visualization
Not a member yet
786 research outputs found
Sort by
A Review of Livestock Smart Farming for Sustainable Food Security
Get PDFMaintaining food security via sustainable farming methods is a significant problem as the global population grows. This study aims to examine the impact of smart farming methods on enhancing farm animal output to satisfy rising demand while fostering sustainability. Smart livestock farming incorporates automation, Internet of Things (IoT) sensors, and machine learning algorithms to improve production, efficiency, and resource utilization. With an emphasis on essential factors including automated feeding, environmental monitoring, and health tracking, this study takes a methodical approach to reviewing IoT-based livestock farming. The efficiency of several sensor technologies, including motion, temperature, humidity, and biometric sensors, is examined in gathering data and making decisions in real time. The potential of machine learning methods like pattern identification, anomaly detection, and predictive analytics to maximize the production and health of farm animals is assessed. According to the results, IoT-driven livestock farming improves illness diagnosis, minimizes resource waste, and optimizes feeding practices, increasing production efficiency. These developments minimize the impact on the environment while promoting steady food production. Additionally, less human interference results from automation in livestock production, which lowers costs and improves decision-making. This study demonstrates how smart agricultural technology may be used to address issues related to food security. Further research is needed to increase real-time data processing, hone machine learning models, and investigate affordable options for broadly adopting these ideas into practice. Livestock management may be transformed, guaranteeing a robust and sustainable agricultural environment
Push-Pull-Mooring Theory and The Moderating Effect of Inertia on Switching Intention to Mobile Learning
No full textInertia is a hindering factor in transition, which is essential to investigate as a push-pull-mooring factor influencing the switching intention to use mobile learning. Mobile learning research in Indonesia is still new, and only a few studies analyze the moderating effect of inertia on switching intention to mobile learning. The research aims to examine students' intentions to adjust to mobile learning at universities in Indonesia and analyze the moderating effect of inertia in weakening the correlation between pull and push factors and shifting intention. This study employed a quantitative method involving a sample of 163 students. To explain inertia, this study adopted habits, switching costs, student innovation, network externalities, and technological self-efficacy as independent variables leading to inertia. This research reveals that inertia moderates learning convenience, learning autonomy, and task technology fit. Meanwhile, inertia is influenced by habit, switching costs, student innovativeness, and technological self-efficacy. This research also confirms that service quality, perceived enjoyment, and task technology fit significantly impact switching intention to employ the use of mobile learning. This research reveals that inertia moderates learning convenience, learning autonomy, and task technology fit. Meanwhile, inertia is influenced by habit, switching costs, student innovativeness, and technological self-efficacy. This research also confirms that service quality, perceived enjoyment, and task technology fit have a significant effect on switching intention to use mobile learning. University management and practitioners must increase students’ awareness of the benefits of mobile learning in higher education institutions. Further research should test additional variables such as gender and student satisfaction with mobile learning
Enhancing Coffee Marketing Strategies through Multi-Criteria Decision-Making
Get PDFCoffee is a globally preferred beverage, and Indonesia, as a major supplier, provides a wide variety of high-quality coffee varieties with unique characteristics from each region. East Manggarai, East Nusa Tenggara, Indonesia, produces Colol coffee, a high-quality variety with unexplored market potential. The marketing of Colol coffee faces significant challenges, including limited accessibility, lack of market information, and inadequate logistics infrastructure. A comprehensive marketing strategy necessitates the consideration of numerous criteria, which generate a range of alternative decisions to identify the marketing area. This study proposes a framework to optimize the marketing strategy of Colol coffee using the MCDM (Multi-Criteria Decision-making) approach, which integrates AHP, SMARTER, and TOPSIS methods. This framework is applied to rank marketing areas in 38 provinces in Indonesia based on five criteria, namely, accessibility, market potential, logistics, environmental conditions, and safety. The results show that the MCDM method can increase the effectiveness of marketing strategies. The top three alternative coffee marketing regions are Papua, East Kalimantan, and South Papua, with eigenvalues of 0.0569, 0.0424, and 0.0421. With incomplete data, in some marketing areas, it is a challenge to integrate multiple MCDM methods to have a better ranking that represents the real world of marketing strategy. This study supports the enhancement of the digital economy in the agricultural sector. It provides a meaningful understanding of the application of MCDM in marketing agricultural products, with far-reaching implications for marketing strategies in similar sectors
Comparative Study of CNN Architectures for Real-Time Audio-Based Car Accident Detection on Edge Devices
Get PDFTraffic accidents often result in fatalities for both drivers and bystanders. Traditionally, accident information relies heavily on community reports, which can delay the provision of victim assistance. To address this issue, a system capable of detecting accidents responsively in various weather conditions and traffic densities is necessary. One approach involved using audio analysis techniques to evaluate collision sounds. Thus, this study proposed an audio classification system for detecting car accidents using Convolutional Neural Networks (CNNs). The system’s performance was evaluated on personal computers and edge devices, such as the Raspberry Pi 4 and NVIDIA Jetson Nano, to compare inference times and power consumption. To enhance the dataset, segmentation and augmentation techniques were applied before converting the audio data into a 2D Mel-spectrogram. The dataset was then trained and assessed with four CNN architectures: custom sequential, custom with shared input layer, transfer learning EfficientNetB0, and transfer learning MobileNetV2. Both original and Lite models were deployed on experimental devices. Results showed that the custom CNN model had faster inference times across devices in both original and lite forms, though it had a 4% increase in the false positive rate. The Lite MobileNetV2 model recorded the fastest inference time on edge devices at 86 ms. Jetson Nano exhibited faster inference times compared to Raspberry Pi 4. However, Raspberry Pi 4 showed a minor increase in power consumption of 0.6 watts during inference. In future work, this system can be tested in real-time environments using embedded systems to evaluate its robustness against noise and varying environmental conditions
An Eccentricity for Improvement in Rice Stem Borer Detection Using Sensed Drone Imaging
Get PDFRice stem borers are severe pests that cause significant crop losses. This research aimed to tackle this problem by using a drone equipped with a high-resolution camera to capture detailed images of paddy fields. These images were then processed to estimate the early potential attacks of stem borer pests through color segmentation computing. The detection process relied on analyzing color variations, particularly focusing on symptoms indicative of stem borer presence. The system utilized Hue, Saturation, Value (HSV) color segmentation and advanced image processing algorithms on numerous rice field videos collected from drone flights conducted at altitudes ranging from 5 to 40 meters above the ground. To improve detection accuracy, the study tested the system with and without the eccentricity parameter, which is crucial in eliminating false positives caused by the misidentification of field embankments as stem borers. This research's primary contribution is the implementation of eccentricity, which significantly reduces the false-positive rate. The results demonstrated that the accuracy of the system with the eccentricity parameter included was 75%, compared to a significantly lower accuracy rate of 17.19% when the eccentricity parameter was not used. Overall, this study highlights the effectiveness of using drones for remote sensing and the importance of incorporating eccentricity in image processing algorithms to enhance the precision of early stem borer detection in rice fields. This approach not only improves the reliability of pest detection but also offers a promising method for protecting rice crops from severe pest damage
Recommender System Based on Social Network Analysis of Student Workshop and Event Activities Compared to GPA and Department
Get PDFThis research uses social network connections and academic data to create a recommender system that helps students choose seminars and events that suit their interests. The aim is to address the issue of students' hesitation in selecting activities. This project investigates the use of social network analysis (SNA) to provide individualized suggestions by analyzing student involvement in workshops and events, as well as their grade point average (GPA). The materials contain student data gathered between 2018 and 2023 from Institut Sains dan Teknologi Terpadu Surabaya (ISTTS), emphasizing the student's social media interactions and event participation. Metrics like centrality are employed to identify prominent nodes inside the network, and the approach combines graph-based SNA and cosine similarity for event recommendation. The network of student involvement in events was represented by a dataset comprising 2,293 edges and 602 nodes. The results show that the relevance of recommendations is improved when social network data is integrated with GPA, rather than GPA-based systems alone. The system identified key nodes, such as specific lectures, that significantly impacted student involvement and were rated highly in terms of centrality. Future research implications recommend expanding the dataset to encompass a broader range of events and refining the algorithm by including content-based filtering. The system's application is not limited to educational environments; it may also be tailored for career counselling or professional development
Personalized Tourism in Surabaya: A Bayesian Network Approach
Get PDFThis study investigates the application of Bayesian Networks in developing a personalized tourist destination recommendation system focused on Surabaya, Indonesia. The research incorporates push and pulls factors alongside tourist activities as key input variables to model decision-making processes. Two distinct Directed Acyclic Graph (DAG) structures are evaluated: one proposed based on existing theoretical frameworks and another generated from empirical respondent data. The dataset comprises responses from 1,350 tourists visiting twenty-five popular attractions in Surabaya. The analysis reveals that Bayesian Networks effectively identify correlations between various influencing factors. From the tests carried out, the accuracy obtained from the two DAG structures did not significantly differ. The proposed DAG achieved 35% accuracy for the top-ranked destination recommendations, while the data-driven DAG was 25%. Both achieved 75% accuracy in the top five recommendations. The accuracy increased as the number of output states was reduced. Meanwhile, in the test with binary output, BN was able to accurately classify tourist destinations with an average accuracy of 95% for both DAGs. These findings highlight the potential of Bayesian Networks to enhance tourism decision support systems by providing nuanced insights into tourists' preferences and motivations. For further research, hybridization or feature engineering can be employed to improve model accuracy. In addition, determining more appropriate push factors and tourist activities based on the tourism case studies also needs to be done to obtain better tourist preferences. This research highlights the promising role of Bayesian Networks in improving the personalization and effectiveness of tourist recommendations
A Comprehensive Visualization for Music Education and Artificial Intelligence
Get PDFArtificial intelligence (AI) has revolutionized traditional methods and improved decision-making and automation. AI has also been used to enhance teaching methods, student learning, and research in music education. This study will examine literature on music education and AI. This study aims to investigate significant themes, trends, and achievements in this burgeoning discipline. This study will examine scholarly articles, conference papers, and other relevant literature to explore AI's applications, issues, and future in music education. Machine learning, natural language processing, computer vision, and deep learning are utilized in music education. These techniques are used in music composition, performance evaluation, instructional support, and individualized learning. Adaptive training, real-time feedback, and intelligent music production demonstrate the transformative potential of AI. This study will illuminate the obstacles AI faces in music education. Ethical considerations, data privacy, algorithmic bias, and human competence must be thoroughly investigated. In addition, the analysis would identify knowledge deficits for future research and development. This research could assist educators, researchers, and policymakers utilize AI in music education by conducting a comprehensive literature review. This work can assist in the development of AI-based instruments, the improvement of pedagogy, and the promotion of music education
Multi-label Aspect Dangerous Speech Classification Using Keyword-Driven Ensemble Classifier on Imbalanced Data
Get PDFThis study aims to detect various aspects of dangerous speech on social media, particularly Twitter, which has the potential to incite violence and increase prejudice against specific communities. The research dataset includes tweets containing dangerous speech related to the Indonesian government from 2019 to 2022. Researchers manually labeled the data based on seven aspects of hazardous speech, including social and historical context, dehumanization, accusations in the mirror, threats against women/children, questioning in-group loyalty, and threats against groups. The study employs a multi-label classification method to handle these aspects, which appear simultaneously in a single text. The main challenges include data imbalance, ambiguity, and the informal language frequently appearing in tweets. This study introduces a Keyword-Driven Ensemble Classifier (KDEC), a new ensemble model that leverages the strengths of SVC, Logistic Regression, IndoBERTweet, and specific keyword lists for each label. Researchers designed KDEC based on the best results from machine learning and deep learning methods tested in this study. The research team tested the model on small and large datasets, conducting trials involving seven and four-label classifications. The results show that KDEC, with label reduction and keyword support, effectively addresses data imbalance, resolves label overlap, and achieves 92% accuracy for seven-label classification and 88% for four-label classification. The findings of this research are highly relevant for hate speech analysis across various platforms and languages, particularly in understanding context and conveyed messages. Additionally, this study provides valuable insights into managing harmful content in online government-related discussions. This method identifies dangerous speech on a larger scale and supports data-driven social media content regulation decision-making
Indonesian Word Sound Recognition Using Convolutional Neural Network Method
Get PDFAccess to education, particularly in a university environment, is essential for deaf and hard-of-hearing students as more of them pursue higher education. At UIN Sunan Kalijaga the current challenges are a limited number of sign language interpreters and translating technical terminology in lectures. Many methods are available for speech recognition, but research on how well this method performs in Indonesian has not been published, especially in education-level recognizers. This experimental study aims to investigate if Indonesian words can be recognized through Convolutional Neural Networks (CNN) and to find out the Data Ratio for Training, Validation, and Testing set to get the best performance. The study used a dataset of 4 Indonesian words with the total voice sample, each with 50 voice samples from young adults aged 19-23. Audio data is preprocessed into spectrograms, inputs to the CNN model using TensorFlow. The CNN Model had a 90% accuracy with a 60:20:20 ratio between training, validation, and test data. The other ratios (70:15:15 and 80:10:10) provided accuracy ranges of between 80% to 90%. This study shows that CNNs are the best for Indonesian word recognition and that the data ratio of 60:20:20 is optimal. This result has valuable benefits, such as using voice-to-text over lectures to enhance the ease of learning and education in Indonesia. Further studies should be conducted using different neural network approaches; the denoise approach is also necessary to increase accuracy