The onset of the COVID-19 pandemic prompted a rapid adoption of video conferencing tools to support teaching and learning. Even though the pandemic has receded, these tools have cemented their place in the education landscape. This research aims to review the literature on the use of videoconferencing tools, specifically Zoom, in the context of higher education (HE). Articles were retrieved from two databases: Scopus and ERIC (Education Resources Information Center). The final sample included a total of 84 articles. The review uncovers six benefits of using Zoom in HE: Social presence, flexibility and accessibility, facilitating synchronous remote academic advising, conducting assessments, perceived usability, and inclusivity. It also identified seven challenges: Engagement issues, simulating hands-on experience, technical issues, privacy and security issues, familiarity issues, instructor difficulty, and Zoom fatigue. Several implications of these findings for future educational practices are discussed.
Millions are fighting to maintain mental well-being in our high-pressure world. Stress, isolation, and mental illnesses can all be relentless foes, and neglecting them can have a domino effect on physical health, leading to a decline in overall well-being. Technological advances, particularly spatial computing, offer promising solutions to this problem. Spatial computing's main feature is the contextualization of digital information within users' physical environments. Past studies have shown that this feature can support mental health by enabling remote therapy delivery and providing immersive therapeutic environments anywhere, at any time. While most studies on spatial computing in mental health highlight its benefits, the potential drawbacks and societal integration challenges remain largely unexplored. This work is a systematic literature review aimed at identifying opportunities for integrating spatial computing into mental healthcare, as well as the challenges it could face upon entering society and any potentially negative impacts it may have. The findings demonstrate that spatial computing can improve mental well-being through enhanced accessibility and immersive experiences. However, integration challenges and potentially negative impacts of spatial computing were also concerning, such as high costs and the risk of addiction. Due to the lack of extensive studies on these aspects, the findings on integration challenges and potential negative impacts require further investigation. Future research and real-world applications of spatial computing are necessary to validate this study's findings and uncover any underlying effects.
Data drift, caused by changes in environment from training, can significantly degrade the performance of neural networks in real-world computer vision tasks. Traditional drift detection methods often focus on a single type of drift, limiting their robustness in complex scenarios. In this paper, we extend an existing method that estimates the degree of a single drift for an image classification model by targeting multi-drift problem for which a set of answer keys is prepared for combination of drift levels and the answer keys are used to predict the magnitudes of drift combination applied to the images. We evaluated our extended method using the CIFAR-10 dataset with two drifts of various levels of Gaussian noise and rain effects. Experimental results confirmed that predicting multiple drifts together effectively prevents the overestimation of drift magnitudes. Additionally, the proposed drift detection model further improves accuracy of drift magnitude estimation.
This study aims to classify anxiety disorders in adolescents using three machine learning algorithms, namely naive bayes, C4.5, and K-Nearest Neighbor (K-NN). The data used was taken from the DASS-21 questionnaire, which contains 418 samples with 11 attributes, including age, gender, and seven anxiety-related questions. The algorithm was tested using the holdout technique with 80:20, 70:30, and 60:40 data splits, as well as the k-fold cross-validation technique. The results showed that the C4.5 algorithm performed best with 100% accuracy in the holdout technique, followed by naive bayes with 99% accuracy, and K-NN with 94% accuracy. In cross-validation testing, C4.5 also showed the highest accuracy of 98%, while naive bayes and K-NN achieved 89% and 92% respectively. This study concludes that the C4.5 algorithm is superior in classifying anxiety compared to naive bayes and K-NN, so it can be relied upon for machine learning-based diagnostic applications in supporting the detection of anxiety disorders efficiently.
Due to the persisting environmental concerns and resource depletion caused by traditional energy sources, people have begun to seek renewable sources of energy and develop energy harvesting tools. The application of the piezoelectric effect has been utilized by many large and small-scale energy harvesters as a way to combat the energy crisis many communities face today. Although underutilized, piezoelectric's potential for energy harvesting remains; with advancements in material science and energy-harvesting techniques, it can be more widely and effectively deployed. Compared to piezoelectric films, solar films are more utilized in many fields; nonetheless, their application in a residential setting remains uncommon. This paper focuses on developing a plant-shaped hybrid energy harvester with Platanus Acerifolia Leaf designed to cater to wind resistance. In controlled experiments, the piezoelectric leaves were shown to produce 2.268V at the wind speed of 9 m/s, while the solar films generated 4.2V at an illuminance of 2000 lux. In an uncontrolled residential test, the hybrid system produced a constant output voltage of 3.6V with a total power generation of 343.6 mW over 9 hours resulting in a total energy collection of 3,092.4 mWh. This enhanced plantshaped energy harvester is capable of providing an additional source of energy for many, especially in a residential setting. Moreover, it contributes to and promotes the field of material science and the utilization of energy-harvesting techniques.
This project aimed to implement a prototype capable of real-time monitoring and control of various parameters in information management setting with Internet of Things (IoT). Research methodology involved designing and implementing the system prototype to resemble a supply container or transportation truck with installed sensors controlled by a microcontroller. Main components used temperature sensors (AHT20), RFID reader for access control, GPS module for location tracking, and a web/mobile dashboard for remote monitoring. Additionally, a robust database system was developed to store sensor readings and transaction history, ensuring data integrity and facilitating seamless communication between components for real-time data transmission. Key findings indicate that the system effectively monitors temperature, manages access control through RFID authentication, tracks real-time location via GPS, and provides a user-friendly dashboard interface for data visualization.
Cyberbullying is now a serious issue in contemporary times. The impact of cyberbullying on youngsters is of enormous concern in the public health sector as well as in the cybersecurity field. Several researchers have discussed the impact of cyberbullying from diverse perspectives, however, there is still a large gap in finding the paramount impact of cyberbullying in contemporary times, as well as providing solutions to eradicate this menace, especially as it deals with the negative effects of technology which is emerging daily. This study reviews via a scoping evidence approach suggesting the impact of cyberbullying on youngsters, focusing on the psychological, emotional, physical health, and academic implications. Findings from studies revealed that most youngsters face more of the psychological and emotional consequences of cyberbullying as compared to others discussed in the literature. Interestingly, there were also revelations of how cyberbullying had negative influences on the physical health and academic achievements of youngsters. These findings thus instigate the need for novel approaches that can help curb the rampancy of cyberbullying as well as ensure that decision-makers and stakeholders in the respective institutions put these impacts of cyberbullying into consideration when designing cybersecurity policies to ensure that victims of cyberbullying are assisted to recover, put in place technological solutions that can track and punish offenders and also reduce the rate of cyberbullying victimization.
The pervasive spread of dis/misinformation in Indonesia, particularly via social media and instant messaging applications, has emerged as a major concern due to its societal challenges. This study investigates the roles of various stakeholders and the interplay between stakeholders in combating dis/misinformation in Indonesia. By understanding the roles of various stakeholders, this research provides insights for improving strategies to counter dis/misinformation. Using a qualitative research design, data was collected through interviews with stakeholders and analyzed using coding and thematic analysis. The results highlight the use of technologies such as WhatsApp Business API, Hoax Buster Tools, and AI-based verification tools by organizations like Mafindo and Cekfakta.com. This study also explores the importance of public engagement and digital literacy campaigns in Indonesia, including grassroots initiatives like Lentera Litera and TurnbackHoax.id. This research underscores the roles and challenges faced by journalists, media organizations, fact-checking platforms, and NGOs in enhancing digital literacy and efforts in combating dis/misinformation. The findings suggest that a multi-faceted approach involving regulation, technology, public engagement, and international collaboration is essential for combating dis/misinformation effectively. This research enhances understanding of dis/misinformation dynamics and provides actionable recommendations for Indonesian stakeholders, emphasizing the need for collaborative efforts, technology use, and public engagement to create a truthful information ecosystem.
Drug-target Interaction (DTI) prediction provides strong support for drug repurposing. Most of the existing models are designed for specific protein / drug datasets to identify decoys, but there is still room for improvement in the generalization capability. Therefore, a universal DTI framework supporting Multiple Data Foundations with Alternative Training and Contrastive Learning (MuFAl) is proposed, learning diverse information such as protein structure and sequence from heterogeneous data sources. MuFAl makes predictions based on the representation distance in vector spaces, with improved generalization capability brought by alternative training and fine-grained recognition ability of decoys provided by contrastive learning. This model could be used for scenarios including large-scale heterogeneous data processing or zero-shot predictions. Experimental results indicate that compared with some state-of-the-art models, MuFAl retains the fine-grained feature extraction ability while improving the generalization ability.
The business sector generates vast amounts of data daily through contracts, reports, and tenders, containing valuable insights that need extraction. Despite advancements in AI for information extraction, extracting information from tenders remains underexplored. This study focuses on extracting and classifying named entities from unstructured tender documents to improve tender management. We developed a document segmentation approach, utilizing various text extraction tools on tender PDF tables. A custom text analyzer was created for text normalization, keyword identification, and segmentation into header, body, and footer. By discarding body text and combining header and footer, we reduced text complexity. Challenges due to unstructured PDF tables were addressed using rules and regular expressions to extract and classify tender named entities. This method enhances the usability and analysis of tender documents by accurately identifying and categorizing tender entities.
Given the prevalence of new concepts, technical terms, and the rise of e-book reading in scholarly papers, research has been conducted to enhance scholarly browsing by providing access to key terms, cited texts, and definitions of technical terms. We propose a method to extract various types of information from scholarly papers using ChatGPT and provide readers with helpful information for scholarly browsing. The proposed interface involves five key elements: a keyword tag cloud, a flowchart outlining the methodology, figure and table descriptions, a paper summary, and summaries of each section. These elements are presented to the reader alongside the PDF file of a paper. In our experiments, we measured the time required to answer questions about the paper's content and the percentage of correct responses to evaluate both the speed and depth of comprehension. The impact of the reading comprehension support interface on usability, understanding depth, and time to understand the paper was assessed via a questionnaire. Our experimental results demonstrated that while the interface did not reduce comprehension time, seven out of ten subjects reported slightly faster comprehension, with none experiencing slower comprehension. Half of the subjects indicated their understanding was somewhat deeper, and none reported a decline. Furthermore, nine out of ten subjects found the paper summary and section summaries to enhance their understanding.
Various forms of communication will shape the level of intimacy of interpersonal relationships between husband and wife. Interpersonal communication is also inseparable from digital media such as Whatsapp application. By using Whatsapp, communication between husband and wife can become more intense. However, it is not uncommon to find that language through chat can be a means of miscommunication. The amount of noise in digital communication has been proven to often trigger commotion in communication between husband and wife. Behaviorism theory talks about all human behavior as the result of a learning process from life experiences or educational backgrounds. This study aims to describe the interpersonal communication management of husbands and wives who communicate a lot through digital media appropriately. This interpersonal communication management is expected to minimize the occurrence of miscommunication between husband and wife who both work and have a marriage age of 1–10 years. The research method used is a qualitative method by conducting depth interviews with 5 couples & open ended survey with 15 respondents. This study found the topic and context of WhatsApp use in working couples, frequently used features and the advantages and potential conflicts of WhatsApp usage in working couples.
The rise of advanced technologies, such as AI- driven chatbots, enables SMEs in e-marketplaces to provide responsive and efficient customer support, improve engagement, and streamline services. However, customers increasingly express concerns about AI-supported chatbot services, which affects their willingness to engage with these technologies. Consequently, this study aims to examine the factors that influence customers’ behavioral intentions to use AI and their intentions for the continued use of AI-supported chatbots. Using a purposive sampling technique, the study collected data from 152 respondents through an online questionnaire. To analyze and predict the findings from the collected data, the study employed PLS-SEM as its statistical approach. The results indicate that information quality, system quality, and service quality significantly influence trust. Furthermore, service quality, perceived ease of use, confirmation of expectations, and perceived usefulness affect user satisfaction. Additionally, confirmation of expectations impacts perceived usefulness, and user satisfaction influences the behavioral intention to use AI chatbots. The findings also reveal that trust, user satisfaction, and perceived usefulness effectively enhance the intention to continue using AI-driven chatbots. However, information quality and system quality do not correlate with user satisfaction, and confirmation of expectations does not relate to user satisfaction. These findings contribute valuable insights to the existing literature on AI- driven chatbot services. Furthermore, stakeholders involved with AI-driven chatbot services for SMEs will gain an understanding of how to enhance user-friendly chatbot services.
Logical thinking is essential for organizing one's thoughts and fostering the generation of diverse and innovative ideas. However, acquiring logical thinking skills is not straightforward. This is because individuals often lead others to the conclusions they desire, and emotions can also influence one's reasoning abilities. Therefore, we have designed a web application that encourages users to engage in logical thinking with minimal mental burden, and propose it in this paper. The key feature of this application utilizes human-shaped pictograms. This paper also indicates why the application adopts human-shaped pictograms.
The development of Linguistic Ontology is essential to advance the Indonesian ontology-based question-answering (QA) system. This paper describes the development of Indonesian linguistic ontology (ILO), including design, construction, and evaluation. ILO is developed as a knowledge base that represents linguistic aspects aimed at overcoming the challenges still faced by Indonesian ontology-based QA systems. The test results show that ILO can effectively handle complex sentence structures, recognize sentence pattern variations, and generate appropriate SPARQL queries. ILO can improve natural language understanding in Indonesian ontology-based QA systems. However, ILO, which is entirely manually curated, has drawbacks. Therefore, future work should focus on implementing ILO in QA systems in various domains and exploring semi-automatic methods to improve scalability and adaptability.
In the rapidly envolving digital landscape, small and medium enterprises (SMEs _ face challenges in effectively engaging with consumers through social media due to limited resources and technological adoption. This study examines the impact of information systems ton consumer engagement in social media marketing (SM) for SMEs in Indonesia, with a focus on the adoption of SMM as a primary variable. Using purposive sampling, 129 SMEs from the Jabodetabek region were selected for cross-sectional preliminary research conducted between May and June 2024. Data were collected through a Google Form survey and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) with Smart PLS 3.0. The analysis involved bootstrapping and computation of PLS scores to assess relationships among nine variables: competitor pressure, affordable marketing cost, user-generated content, belief in information, customer engagement, sales performance, and overall marketing effectiveness. The results indicate that social media marketing significantly enhances the marketing effectiveness of SMEs in Indonesia, improving both market efficiency and performance. All hypothesized relationships were statistically significant, except for the relationship between competitor pressure and social media marketing adoption (H1), which was found to be non-significant. These findings suggest that while competitor pressure may not drive the adoption of SMM, other factors such as affordable marketing cost, user-generated content, and customer relationships play a critical role in the success of SMEs' social media strategies.
The course knowledge graph aims to provide personalized learning recommendation and learning path map, and solve the problems of “information overload” and “knowledge roaming” among learners. This paper proposed a method for constructing a course knowledge graph. The method is based on the dependency of knowledge points. The research object is the “Mobile APP Application Development” course. This method uses the SMRK technique, which includes Scope, Module, Relationship, and Knowledge Point. The knowledge graph illustrates the specific characteristics of the course knowledge. It also considers the unique structure and interconnections within computer science during the construction process. In this course graph, the importance of learners' cognitive level and the necessity of visual representation of knowledge concepts, attributes and relationships are considered to improve learning efficiency and reduce redundant learning. Methods for constructing knowledge graphs include the use of manual extraction tools and ontology building tools, supplemented by entity and relation extraction algorithms. The proposed construction method for a course knowledge graph, which is based on the dependencies among knowledge points, illustrates three types of relationships between these knowledge points. This knowledge graph offers effective technical support for learners in their subsequent studies.
Medical images often exhibit low quality, which adversely affects the performance of advanced algorithms such as image segmentation, image fusion and detection, thereby necessitating image enhancement. In this work, a novel Lapla-cian attention residual U-net (LaRU-Net) model is proposed to enhance the visual quality of magnetic resonance (MR) and computed tomography (CT) images. The LaRU-Net model is built on the architecture of U-Net with a novel integration of an enhancement block. In addition, the attention-augmented skip connections and residual learning are employed in the encoder-decoder framework for improving feature propagation. The enhancement block with laplacian filtering and an attention mechanism aids the LaRU-Net to emphasize anatomical details and learn from enhanced features. A joint loss function is also introduced which is a weighted combination of three distinct measures. The performance of the proposed method is compared with the state-of-the-art methods utilizing eight image quality assessment (IQA) metrics. The quantitative evaluation shows that the LaRU-Net outperforms existing methods.
The field of agriculture has slowly been integrating the use of technology with their daily operations to its advantage. Sunflower seeds, which are a popular mass-produced agricultural product, also employ these emerging techniques. This research aims to introduce a computer vision-based system designed to automate the classification and counting of sunflower seeds by leveraging YOLOv8 and DeepSORT algorithms. The system aims to identify and quantify three distinct seed varieties: Giant, Dwarf F1, and Mammoth Grey seeds using a Raspberry Pi-powered hardware setup. This setup was also used to capture high-quality images of the sunflower seeds which made up the dataset used. The researchers will utilize, along with the Raspberry Pi. a USB webcam to capture the video feed of the sunflower seeds and an LCD monitor to display the seed count and seed variety. Upon running multiple trials and utilizing a confusion matrix to analyse the system accuracy, the system was able to achieve an overall accuracy of 91.11% for its classification and 97.56% for its counting accuracy.
The medicine used today is often referred to as “one-size-fits-all” medicine, designed to treat the average person. However, such approach could lead to ineffective or suboptimal results, or even serious side effects for patients who deviate significantly from the average. To alleviate such problem, the concept of precision medicine has emerged, aiming to provide treatment tailored to the individual patient's characteristics. In order for precision medicine to be effective, some requirements such as comprehensive data collection, data integration, and data analysis should be satisfied. In this paper, we propose health digital twin (DT) based on clinical personal knowledge graph (PKG). Since digital twin is continuously synchronized with patient's body condition through collected data, it is possible to describe his/her body function accurately. The proposed PKG is based on the semantics of SNOMED CT and LOINC, which satisfies the requirements. In our approach, we added chronological collection of PKGs to show the changes of the health conditions of the patients to improve accuracy.
