for

A Learn-to-Rank Approach to Medicine Selection for Patient Treatments

Aim/Purpose: This research utilized a learn-to-rank algorithm to provide medical recommendations to prescribers. The algorithm has been utilized in other domains, such as information retrieval and recommender systems. Background: Ranking the possible medical treatments according to diagnoses of the medical cases is very beneficial for doctors, especially during the coding process. Methodology: We developed two deep learning pointwise learn-to-rank models within one prediction pipeline: one for predicting the top possible active ingredients from disease features, the other for ranking actual medicines codes from diseases and the ingredients features. Contribution: A new learn-to-rank deep learning model has been developed to rank medical procedures based on datasets collected from insurance companies. Findings: We ran 18 cross-validation trials on a confidential dataset from an insurance company. We obtained an average normalized discounted cumulative gain (NDCG@8) of 74% with a 5% standard deviation as a result of all 18 experiments. Our approach outperformed a known approach used in the information retrieval domain in which data is represented in LibSVM format. Then, we ran the same trials using three learn-to-rank models – pointwise, pairwise, and listwise – which yielded average NDCG@8 of 71%, 72%, and 72%, respectively. Recommendations for Practitioners: The proposed model provides an insightful approach to helping to manage the patient’s treatment process. Recommendation for Researchers: This research lays the groundwork for exploring various applications of data science techniques and machine learning algorithms in the medical field. Future studies should focus on the significant potential of learn-to-rank algorithms across different medical domains, including their use in cost-effectiveness models. Emphasizing these algorithms could enhance decision-making processes and optimize resource allocation in healthcare settings. Impact on Society: This will help insurance companies and end users reduce the cost associated with patient treatment. It also helps doctors to choose the best procedure and medicines for their patients. Future Research: Future research is required to investigate the impact of medicine data at a granular level.




for

Modeling the Predictors of M-Payments Adoption for Indian Rural Transformation

Aim/Purpose: The last decade has witnessed a tremendous progression in mobile penetration across the world and, most importantly, in developing countries like India. This research aims to investigate and analyze the factors influencing the adoption of mobile payments (M-payments) in the Indian rural population. This, in turn, would bring about positive changes in the lives of people in these countries. Background: A conceptual framework was worked upon using UTAUT as a foundation, which included constructs, namely, facilitating conditions, social influences, performance expectancy, and effort expectancy. The model was further extended by incorporating the awareness construct of m-payments to make it more comprehensive and to understand behavioral intentions and usage behavior for m-payments in rural India. Methodology: A questionnaire-based study was conducted to collect primary data from 410 respondents residing in rural areas in the state of Punjab. Convenience sampling was conducted to collect the data. Structural equation modeling was used to conduct statistical analysis, including exploratory and confirmatory factor analyses. Contribution: A new conceptual model for M-payments adoption in rural India was developed based on the study’s findings. Using the findings of the study, marketers, policymakers, and academicians can gain insight into the factors that motivate the rural population to use M-payments. Findings: The study has found that M-payment Awareness (AW) is the strongest factor within the proposed model for deeper diffusion of M-payments in rural areas in the state of Punjab. Performance expectancy (PE), effort expectancy (EE), social influences (SI), and facilitating conditions (FC) are also positively and significantly related to behavioral intentions for using M-payments among the Indian rural population in the state of Punjab. Recommendations for Practitioners: M-payments are emerging as a new mode of transactions among the Indian masses. The government needs to play a pivotal role in advocating the benefits linked with the usage of M-payments by planning financial literacy and awareness campaigns, promoting transparency and accountability of the intermediaries, and reducing transaction costs of using M-payments. Mobile manufacturing companies should come up with devices that are easy to use and incorporate multilanguage mobile applications, especially for rural areas, as India is a multi-lingual country. A robust regulatory framework will not only shape consumer trust but also prevent privacy breaches. Recommendation for Researchers: It is recommended that a comparative study among different M-payment platforms be conducted by exploring constructs such as usefulness and ease of use. However, the vulnerability of data leakage may result in insecurity and skepticism about its adoption. Impact on Society: India’s rural areas have immense potential for adoption of M-payments. Appropriate policies, awareness drives, and necessary infrastructure will boost faster and smoother adoption of M-payments in rural India to thrive in the digital economy. Future Research: The adapted model can be further tested with moderating factors like age, gender, occupation, and education to understand better the complexities of M-payments, especially in rural areas of India. Additionally, cross-sectional studies could be conducted to evaluate the behavioral intentions of different sections of society.




for

Investigating Intention to Invest in Online Peer-to-Peer Lending Platforms Among the Bottom 40 Group in Malaysia

Aim/Purpose: This study investigates the intention to invest in online peer-to-peer (P2P) lending platforms among the bottom 40% (B40) Malaysian households by income. Background: The B40 group citizens earn less than USD 1,096.00 (i.e., RM 4,850.00) in monthly household income, thereby possessing relatively small capital investments suitable for online P2P lending. Methodology: Drawing on the technology acceptance model (TAM), this research developed and tested the relevant hypotheses with data collected from 216 respondents. The partial least square structural equation modelling (PLS-SEM) technique was employed to analyse the collected data. Contribution: This study contributes to the body of knowledge on financial inclusion by demonstrating the relevance of modified TAM in explaining the intention to invest in online P2P lending platforms among investors with lower disposable income (i.e., the B40 group in Malaysia). Findings: The findings revealed that information quality, perceived risk, and perceived ease of use are relevant to B40 investment intention in P2P online lending platforms. However, contrary to expectations, trust and financial literacy are insignificant predictors of B40 investment intention. Recommendations for Practitioners: The P2P lending platform operators could enhance financial inclusion among the B40 group by ensuring borrowers provide sufficient, relevant, and reliable information with adequate security measures to minimise risk exposure. The financial regulators should also conduct periodic audits to ensure that the operators commit to enhancing information quality, platform security, and usability. Recommendation for Researchers: The intention to invest in online P2P lending platforms among the B40 group could be enhanced by improving information quality and user experience, addressing perceived risks, reassessing trust-building strategies and financial literacy initiatives, and adopting holistic, interdisciplinary approaches. These findings suggest targeted strategies to enhance financial inclusion and investment participation among B40 investors. Impact on Society: The study’s findings hold significant implications for financial regulators and institutions, such as the Securities Commission Malaysia, Bank Negara Malaysia, commercial and investment banks, and insurance companies. By focusing on these key determinants, policymakers can design targeted interventions to improve the accessibility and attractiveness of P2P lending platforms for B40 investors. Enhanced information quality and ease of use can be mandated through regulatory frameworks, while effective risk communication and mitigation strategies can be developed to build investor confidence. These measures can collectively promote financial growth and inclusion, supporting broader economic development goals. Future Research: Future research could expand the sample size to consider older B40 individuals across different countries and use a longitudinal survey to assess the actual investment decision of the B40 investors.




for

Recommendation System for an Online Shopping Pay-Later System Using a Multistage Approach: A Case Study from Indonesia

Aim/Purpose: In this study, we developed a recommendation system model designed to support decision-makers in identifying consumers eligible for pay-later options via consensus-based decision-making. This approach was chosen due to the high and complex risks involved, such as delayed payments, challenges in reaching consumers, and issues of bad credit. Background: The “pay-later” option, which allows consumers to postpone payment for e-commerce purchases, offers convenience and flexibility but also introduces several challenges: (i) by enabling payment deferral, merchants face financial risks, including potential delays or defaults in payment, adversely affecting their cash flow and profitability; and (ii) this payment delay can also heighten the risk of fraud, including identity theft and unauthorized transactions. Methodology: This study initiated a risk analysis utilizing the ROAD process. Considering contemporary economic developments and advancements in neural networks, integrating these networks into risk assessment has become crucial. Consequently, model development involved the amalgamation of three deep learning methods – CNN (Convolutional Neural Networks), RNN (Recurrent Neural Networks), and LSTM (Long Short-Term Memory) – to address various risk alternatives and facilitate multi-stage decision-making recommendations. Contribution: Our primary contribution is threefold. First, our study identified potential consumers by prioritizing those with the smallest associated problem consequence values. Second, we achieved an optimal recall value using a candidate generator. Last, we categorized consumers to assess their eligibility for pay-later rights. Findings: The findings from this study indicate that our multi-stage recommendation model is effective in minimizing the risk associated with consumer debt repayment. This method of consumer selection empowers policymakers to make informed decisions regarding which consumers should be granted pay-later privileges. Recommendations for Practitioners: This recommendation system is proposed to several key parties involved in the development, implementation, and use of pay-later systems. These parties include E-commerce Executive Management for financial analysis and risk evaluation, the Risk Management Team to assess and manage risks related to users utilizing Pay-Later services, and Sales Managers to integrate Pay-Later services into sales strategies. Recommendation for Researchers: Advanced fraud detection mechanisms were implemented to prevent unauthorized transactions effectively. The goal was to cultivate user confidence in the safety of their financial data by ensuring secure payment processing. Impact on Society: Ensuring consumers understand the terms and conditions of pay-later arrangements, including interest rates, repayment schedules, and potential fees, is crucial. Providing clear and transparent information, along with educating consumers about their financial responsibilities, helps prevent misunderstandings and disputes. Future Research: Our future development plans involve the ongoing assessment of the system’s performance to enhance prediction accuracy. This includes updating models and criteria based on feedback and changes in economic or market conditions. Upholding compliance with security and data privacy regulations necessitates the implementation of protective measures to safeguard consumer information. The implementation of such a system requires careful consideration to ensure fairness and adherence to legal standards. Additionally, it is important to acknowledge that algorithms and models may evolve over time through the incorporation of additional data and continuous evaluations.




for

The Relationship Between Electronic Word-of-Mouth Information, Information Adoption, and Investment Decisions of Vietnamese Stock Investors

Aim/Purpose: This study investigates the relationship between Electronic Word-of-Mouth (EWOM), Information Adoption, and the stock investment of Vietnamese investors. Background: Misinformation spreads online, and a lack of strong information analysis skills can lead Vietnamese investors to make poor stock choices. By understanding how online conversations and information processing influence investment decisions, this research can help investors avoid these pitfalls. Methodology: This study applies Structural Equation Modelling (SEM) to investigate how non-professional investors react to online information and which information factors influence their investment decisions. The final sample includes 512 investors from 18 to 65 years old from various professional backgrounds (including finance, technology, education, etc.). We conducted a combined online and offline survey using a convenience sampling method from August to November 2023. Contribution: This study contributes to the growing literature on Electronic Word-of-Mouth (EWOM) and its impact on investment decisions. While prior research has explored EWOM in various contexts, we focus on Vietnamese investors, which can offer valuable insights into its role within a developing nation’s stock market. Investors, particularly those who are new or less experienced, are often susceptible to the influence of EWOM. By examining EWOM’s influence in Vietnam, this study sheds light on a crucial factor impacting investment behavior in this emerging market. Findings: The results show that EWOM has a moderate impact on the Information Adoption and investment decisions of Vietnamese stock investors. Information Quality (QL) is the factor that has the strongest impact on Information Adoption (IA), followed by Information Credibility (IC) and Attitude Towards Information (AT). Needs for Information (NI) only have a small impact on Information Adoption (IA). Finally, Information Adoption (IA) has a limited influence on investor decisions in stock investment. We also find that investors need to verify information through official sites before making investment decisions based on posts in social media groups. Recommendations for Practitioners: The findings suggest that state management and media agencies need to coordinate to improve the quality of EWOM information to protect investors and promote the healthy development of the stock market. Social media platform managers need to moderate content, remove false information, prioritize displaying authentic information, cooperate with experts, provide complete information, and personalize the experience to enhance investor trust and positive attitude. Securities companies need to provide complete, accurate, and updated information about the market and investment products. They can enhance investor trust and positive attitude by developing news channels, interacting with investors, and providing auxiliary services. Listed companies need to take the initiative to improve the quality of information disclosure and ensure clarity, comprehensibility, and regular updates. Use diverse communication channels and improve corporate governance capacity to increase investor trust and positive attitude. Investors need to seek information from reliable sources, compare information from multiple sources, and carefully check the source and author of the information. They should improve their investment knowledge and skills, consult experts, define investment goals, and build a suitable investment portfolio. Recommendation for Researchers: This study synthesized previous research on EWOM, but there is still a gap in the field of securities because each nation has its laws, regulations, and policies. The relationships between the factors in the model are not yet clear, and there is a need to develop a model with more interactive factors. The research results need to be further verified, and more research can be conducted on the influence of investor psychology, investment experience, etc. Impact on Society: This study finds that online word-of-mouth (EWOM) can influence Vietnamese investors’ stock decisions, but information quality is more important. Policymakers should regulate EWOM accuracy, fund managers should use social media to reach investors, and investors should diversify their information sources. Future Research: This study focuses solely on the stock market, while individual investors in Vietnam may engage in various other investment forms such as gold, real estate, or cryptocurrencies. Therefore, future research could expand the scope to include other investment types to gain a more comprehensive understanding of how individual investors in Vietnam utilize electronic word-of-mouth (EWOM) and adopt information in their investment decision-making process. Furthermore, while these findings may apply to other emerging markets with similar levels of financial literacy as Vietnam, they may not fully extend to countries with higher financial literacy rates. Hence, further studies could be conducted in developed countries to examine the generalizability of these findings. Finally, future research could see how EWOM’s impact changes over a longer period. Additionally, a more nuanced understanding of the information adoption process could be achieved by developing a research model with additional factors.




for

Revolutionizing Autonomous Parking: GNN-Powered Slot Detection for Enhanced Efficiency

Aim/Purpose: Accurate detection of vacant parking spaces is crucial for autonomous parking. Deep learning, particularly Graph Neural Networks (GNNs), holds promise for addressing the challenges of diverse parking lot appearances and complex visual environments. Our GNN-based approach leverages the spatial layout of detected marking points in around-view images to learn robust feature representations that are resilient to occlusions and lighting variations. We demonstrate significant accuracy improvements on benchmark datasets compared to existing methods, showcasing the effectiveness of our GNN-based solution. Further research is needed to explore the scalability and generalizability of this approach in real-world scenarios and to consider the potential ethical implications of autonomous parking technologies. Background: GNNs offer a number of advantages over traditional parking spot detection methods. Unlike methods that treat objects as discrete entities, GNNs may leverage the inherent connections among parking markers (lines, dots) inside an image. This ability to exploit spatial connections leads to more accurate parking space detection, even in challenging scenarios with shifting illumination. Real-time applications are another area where GNNs exhibit promise, which is critical for autonomous vehicles. Their ability to intuitively understand linkages across marking sites may further simplify the process compared to traditional deep-learning approaches that need complex feature development. Furthermore, the proposed GNN model streamlines parking space recognition by potentially combining slot inference and marking point recognition in a single step. All things considered, GNNs present a viable method for obtaining stronger and more precise parking slot recognition, opening the door for autonomous car self-parking technology developments. Methodology: The proposed research introduces a novel, end-to-end trainable method for parking slot detection using bird’s-eye images and GNNs. The approach involves a two-stage process. First, a marking-point detector network is employed to identify potential parking markers, extracting features such as confidence scores and positions. After refining these detections, a marking-point encoder network extracts and embeds location and appearance information. The enhanced data is then loaded into a fully linked network, with each node representing a marker. An attentional GNN is then utilized to leverage the spatial relationships between neighbors, allowing for selective information aggregation and capturing intricate interactions. Finally, a dedicated entrance line discriminator network, trained on GNN outputs, classifies pairs of markers as potential entry lines based on learned node attributes. This multi-stage approach, evaluated on benchmark datasets, aims to achieve robust and accurate parking slot detection even in diverse and challenging environments. Contribution: The present study makes a significant contribution to the parking slot detection domain by introducing an attentional GNN-based approach that capitalizes on the spatial relationships between marking points for enhanced robustness. Additionally, the paper offers a fully trainable end-to-end model that eliminates the need for manual post-processing, thereby streamlining the process. Furthermore, the study reduces training costs by dispensing with the need for detailed annotations of marking point properties, thereby making it more accessible and cost-effective. Findings: The goal of this research is to present a unique approach to parking space recognition using GNNs and bird’s-eye photos. The study’s findings demonstrated significant improvements over earlier algorithms, with accuracy on par with the state-of-the-art DMPR-PS method. Moreover, the suggested method provides a fully trainable solution with less reliance on manually specified rules and more economical training needs. One crucial component of this approach is the GNN’s performance. By making use of the spatial correlations between marking locations, the GNN delivers greater accuracy and recall than a completely linked baseline. The GNN successfully learns discriminative features by separating paired marking points (creating parking spots) from unpaired ones, according to further analysis using cosine similarity. There are restrictions, though, especially where there are unclear markings. Successful parking slot identification in various circumstances proves the recommended method’s usefulness, with occasional failures in poor visibility conditions. Future work addresses these limitations and explores adapting the model to different image formats (e.g., side-view) and scenarios without relying on prior entry line information. An ablation study is conducted to investigate the impact of different backbone architectures on image feature extraction. The results reveal that VGG16 is optimal for balancing accuracy and real-time processing requirements. Recommendations for Practitioners: Developers of parking systems are encouraged to incorporate GNN-based techniques into their autonomous parking systems, as these methods exhibit enhanced accuracy and robustness when handling a wide range of parking scenarios. Furthermore, attention mechanisms within deep learning models can provide significant advantages for tasks that involve spatial relationships and contextual information in other vision-based applications. Recommendation for Researchers: Further research is necessary to assess the effectiveness of GNN-based methods in real-world situations. To obtain accurate results, it is important to employ large-scale datasets that include diverse lighting conditions, parking layouts, and vehicle types. Incorporating semantic information such as parking signs and lane markings into GNN models can enhance their ability to interpret and understand context. Moreover, it is crucial to address ethical concerns, including privacy, potential biases, and responsible deployment, in the development of autonomous parking technologies. Impact on Society: Optimized utilization of parking spaces can help cities manage parking resources efficiently, thereby reducing traffic congestion and fuel consumption. Automating parking processes can also enhance accessibility and provide safer and more convenient parking experiences, especially for individuals with disabilities. The development of dependable parking capabilities for autonomous vehicles can also contribute to smoother traffic flow, potentially reducing accidents and positively impacting society. Future Research: Developing and optimizing graph neural network-based models for real-time deployment in autonomous vehicles with limited resources is a critical objective. Investigating the integration of GNNs with other deep learning techniques for multi-modal parking slot detection, radar, and other sensors is essential for enhancing the understanding of the environment. Lastly, it is crucial to develop explainable AI methods to elucidate the decision-making processes of GNN models in parking slot detection, ensuring fairness, transparency, and responsible utilization of this technology.




for

Learning-Based Models for Building User Profiles for Personalized Information Access

Aim/Purpose: This study aims to evaluate the success of deep learning in building user profiles for personalized information access. Background: To better express document content and information during the matching phase of the information retrieval (IR) process, deep learning architectures could potentially offer a feasible and optimal alternative to user profile building for personalized information access. Methodology: This study uses deep learning-based models to deduce the domain of the document deemed implicitly relevant by a user that corresponds to their center of interest, and then used predicted domain by the best given architecture with user’s characteristics to predict other centers of interest. Contribution: This study contributes to the literature by considering the difference in vocabulary used to express document content and information needs. Users are integrated into all research phases in order to provide them with relevant information adapted to their context and their preferences meeting their precise needs. To better express document content and information during this phase, deep learning models are employed to learn complex representations of documents and queries. These models can capture hierarchical, sequential, or attention-based patterns in textual data. Findings: The results show that deep learning models were highly effective for building user profiles for personalized information access since they leveraged the power of neural networks in analyzing and understanding complex patterns in user behavior, preferences, and user interactions. Recommendations for Practitioners: Building effective user profiles for personalized information access is an ongoing process that requires a combination of technology, user engagement, and a commitment to privacy and security. Recommendation for Researchers: Researchers involved in building user profiles for personalized information access play a crucial role in advancing the field and developing more innovative deep-based networks solutions by exploring novel data sources, such as biometric data, sentiment analysis, or physiological signals, to enhance user profiles. They can investigate the integration of multimodal data for a more comprehensive understanding of user preferences. Impact on Society: The proposed models can provide companies with an alternative and sophisticated recommendation system to foster progress in building user profiles by analyzing complex user behavior, preferences, and interactions, leading to more effective and dynamic content suggestions. Future Research: The development of user profile evolution models and their integration into a personalized information search system may be confronted with other problems such as the interpretability and transparency of the learning-based models. Developing interpretable machine learning techniques and visualization tools to explain how user profiles are constructed and used for personalized information access seems necessary to us as a future extension of our work.




for

Emphasizing Data Quality for the Identification of Chili Varieties in the Context of Smart Agriculture

Aim/Purpose: This research aims to evaluate models from meta-learning techniques, such as Riemannian Model Agnostic Meta-Learning (RMAML), Model-Agnostic Meta-Learning (MAML), and Reptile meta-learning, to obtain high-quality metadata. The goal is to utilize this metadata to increase accuracy and efficiency in identifying chili varieties in smart agriculture. Background: The identification of chili varieties in smart agriculture is a complex process that requires a multi-faceted approach. One challenge in chili variety identification is the lack of a large and diverse dataset. This can be addressed using meta-learning techniques, which allow the model to leverage knowledge learned from other related tasks or artificially expand the dataset by applying transformations to existing data. Another challenge is the variation in growing conditions, which can affect the appearance of chili varieties. Meta-learning techniques can help address this challenge by allowing the model to adapt to variations in growing conditions with task-specific embeddings and optimizations. With the help of meta-learning techniques, such as data augmentation, data characterization, selection of datasets, and performance estimation, quality metadata for accurate identification of chili varieties can be achieved even in the presence of limited data and variations in growing conditions. Furthermore, the use of meta-learning techniques in chili variety identification can also assist in addressing challenges related to the computational complexity of the task. Methodology: The research approach employed is quantitative, specifically comparing three models from meta-learning techniques to determine which model is most suitable for our dataset. Data was collected from the variety assembly garden in the form of images of chili leaves using a mobile device. The research successfully gathered 1,974 images of chili leaves, with 697 images of large red chilies, 649 images of curly red chilies, and 628 images of cayenne peppers. These chili leaf images were then processed using augmentation techniques. The results of image data augmentation were categorized based on leaf characteristics (such as oval, lancet, elliptical, serrated leaf edges, and flat leaf edges). Subsequently, training and validation utilized three models from meta-learning techniques. The final stage involved model evaluation using 2-way and 3-way classification, as well as 5-shot and 10-shot learning scenarios to select the dataset with the best performance. Contribution: Improving classification accuracy, with a focus on ensuring high-quality data, allows for more precise identification and classification of chili varieties. Enhancing model training through an emphasis on data quality ensures that the models receive reliable and representative input, leading to improved generalization and performance in identifying chili varieties. Findings: With small collections of datasets, the authors have used data augmentation and meta-learning techniques to overcome the challenges of limited data and variations in growing conditions. Recommendations for Practitioners: By leveraging the knowledge and adaptability gained from meta-learning, accurate identification of chili varieties can be achieved even with limited data and variations in growing conditions. The use of meta-learning techniques in chili variety identification can greatly improve the accuracy and reliability of the identification process. Recommendation for Researchers: Using meta-learning techniques, such as transfer learning and parameter optimization, researchers can overcome challenges related to limited data and variations in growing conditions in chili variety identification. Impact on Society: The findings from this research can help identify superior chili seeds, thereby motivating farmers to cultivate high-quality chilies and achieve bountiful harvests. Future Research: We intend to verify our approach on a more extensive array of datasets and explore the implementation of more resilient regularization techniques, going beyond image augmentation, within the meta-learning techniques. Furthermore, our goal is to expand our research to encompass the automatic learning of parameters during training and tackle issues associated with noisy labels. Building on the insights gained from our observed outcomes, a future objective is to enhance the refinement of model-agnostic meta-learning techniques that can effectively adapt to intricate task distributions with substantial domain gaps between tasks. To realize this aim, our proposal involves devising model-agnostic meta-learning techniques specifically designed for multi-modal scenarios.




for

Using Social Media Applications for Accessing Health-related Information: Evidence from Jordan

Aim/Purpose: This study examined the use of Social Media Applications (SMAs) for accessing health-related information within a heterogeneous population in Jordan. The objective of this study was therefore threefold: (i) to investigate the usage of SMAs, including WhatsApp, Twitter, YouTube, Snapchat, Instagram, and Facebook, for accessing health-related information; (ii) to examine potential variations in the use of SMAs based on demographic and behavioral characteristics; and (iii) to identify the factors that can predict the use of SMAs. Background: There has been limited focus on investigating the behavior of laypeople in Jordan when it comes to seeking health information from SMAs. Methodology: A cross-sectional study was conducted among the general population in Jordan using an online questionnaire administered to 207 users. A purposive sampling technique was employed, wherein all the participants actively sought online health information. Descriptive statistics, t-tests, and regression analyses were utilized to analyze the collected data. Contribution: This study adds to the existing body of research on health information seeking from SMAs in developing countries, with a specific focus on Jordan. Moreover, laypeople, often disregarded by researchers and health information providers, are the most vulnerable individuals who warrant greater attention. Findings: The findings indicated that individuals often utilized YouTube as a platform to acquire health-related information, whereas their usage of Facebook for this purpose was less frequent. Participants rarely utilized Instagram and WhatsApp to obtain health information, while Twitter and Snapchat were very seldom used for this purpose. The variable of sex demonstrated a notable positive correlation with the utilization of YouTube and Twitter for the purpose of finding health-related information. Conversely, the variable of nationality exhibited a substantial positive correlation with the utilization of Facebook, Instagram, and Twitter. Consulting medical professionals regarding information obtained from the Internet was a strong indicator of using Instagram to search for health-related information. Recommendations for Practitioners: Based on the empirical results, this study provides feasible recommendations for the government, healthcare providers, and developers of SMAs. Recommendation for Researchers: Researchers should conduct separate investigations for each application specifically pertaining to the acquisition of health-related information. Additionally, it is advisable to investigate additional variables that may serve as predictors for the utilization of SMAs. Impact on Society: The objective of this study is to enhance the inclination of the general public in Jordan to utilize SMAs for health-related information while also maximizing the societal benefits of these applications. Future Research: Additional research is required to examine social media’s usability (regarding ease of use) and utility (comparing advantages to risks) in facilitating effective positive change and impact in healthcare.




for

Automatic pectoral muscles and artefacts removal in mammogram images for improved breast cancer diagnosis

Breast cancer is leading cause of mortality among women compared to other types of cancers. Hence, early breast cancer diagnosis is crucial to the success of treatment. Various pathological and imaging tests are available for the diagnosis of breast cancer. However, it may introduce errors during detection and interpretation, leading to false-negative and false-positive results due to lack of pre-processing of it. To overcome this issue, we proposed a effective image pre-processing technique-based on Otsu's thresholding and single-seeded region growing (SSRG) to remove artefacts and segment the pectoral muscle from breast mammograms. To validate the proposed method, a publicly available MIAS dataset was utilised. The experimental finding showed that proposed technique improved 18% breast cancer detection accuracy compared to existing methods. The proposed methodology works efficiently for artefact removal and pectoral segmentation at different shapes and nonlinear patterns.




for

IRNN-SS: deep learning for optimised protein secondary structure prediction through PROMOTIF and DSSP annotation fusion

DSSP stands as a foundational tool in the domain of protein secondary structure prediction, yet it encounters notable challenges in accurately annotating irregular structures, such as β-turns and γ-turns, which constitute approximately 25%-30% and 10%-15% of protein turns, respectively. This limitation arises from DSSP's reliance on hydrogen-bond analysis, resulting in annotation gaps and reduced consensus on irregular structures. Alternatively, PROMOTIF excels at identifying these irregular structure annotations using phi-psi information. Despite their complementary strengths, previous methodologies utilised DSSP and PROMOTIF separately, leading to disparate prediction methods for protein secondary structures, hampering comprehensive structure analysis crucial for drug development. In this work, we bridge this gap using an annotation fusion approach, combining DSSP structures with beta, and gamma turns. We introduce IRNN-SS, a model employing deep inception and bidirectional gated recurrent neural networks, achieving 77.4% prediction accuracy on benchmark datasets, outpacing current models.




for

Feature analytics of asthma severity levels for bioinformatics improvement using Gini importance

In the context of asthma severity prediction, this study delves into the feature importance of various symptoms and demographic attributes. Leveraging a comprehensive dataset encompassing symptom occurrences across varying severity levels, this investigation employs visualisation techniques, such as stacked bar plots, to illustrate the distribution of symptomatology within different severity categories. Additionally, correlation coefficient analysis is applied to quantify the relationships between individual attributes and severity levels. Moreover, the study harnesses the power of random forest and the Gini importance methodology, essential tools in feature importance analytics, to discern the most influential predictors in asthma severity prediction. The experimental results bring to light compelling associations between certain symptoms, notably 'runny-nose' and 'nasal-congestion', and specific severity levels, elucidating their potential significance as pivotal predictive indicators. Conversely, demographic factors, encompassing age groups and gender, exhibit comparatively weaker correlations with symptomatology. These findings underscore the pivotal role of individual symptoms in characterising asthma severity, reinforcing the potential for feature importance analysis to enhance predictive models in the realm of asthma management and bioinformatics.




for

Optimisation with deep learning for leukaemia classification in federated learning

The most common kind of blood cancer in people of all ages is leukaemia. The fractional mayfly optimisation (FMO) based DenseNet is proposed for the identification and classification of leukaemia in federated learning (FL). Initially, the input image is pre-processed by adaptive median filter (AMF). Then, cell segmentation is done using the Scribble2label. After that, image augmentation is accomplished. Finally, leukaemia classification is accomplished utilising DenseNet, which is trained using the FMO. Here, the FMO is devised by merging the mayfly algorithm (MA) and the fractional concept (FC). Following local training, the server performs local updating and aggregation using a weighted average by RV coefficient. The results showed that FMO-DenseNet attained maximum accuracy, true negative rate (TNR) and true positive rate (TPR) of 94.3%, 96.5% and 95.3%. Moreover, FMO-DenseNet gained minimum mean squared error (MSE) and root mean squared error (RMSE) of 5.7%, 9.2% and 30.4%.




for

International Journal of Bioinformatics Research and Applications




for

TRACC: tiered real-time anonymised chain for contact-tracing

Epidemiologists recommended contact-tracing as an effective control measure for the global infection like COVID-19 pandemic. Despite its effectiveness in infection containment, it has many limitations such as labour-intensive process, prone to human errors and most importantly, user privacy concerns. To address these shortcomings, we proposed location-aware blockchain-based hierarchical contact-tracing framework for anonymised data collection and processing. This infectious disease control framework serves both the infected users with localised alerts as well as stakeholders such as city officials and health workers with health statistics. Our proposed solution uses hierarchical network design that offloads individual infection block data to create hospital and city-level 'chains' for generating macro-level infection statistics. Results demonstrate that our system can represent the dynamic complexities of contract tracing in highly infection situations. Overall, our design emphasises on data processing and verification mechanism for large volume of infection data over a significant period of time for active risk assessment.




for

Coevolution of trust dynamics and formal contracting in governing inter-organisation exchange

Recently, interest in the correlation between 'contract' in transaction governance and 'trust' in relational governance mechanisms has been growing. This study focuses on issues related to the evolution of contract and inter-organisational trust dynamics in transaction governance and uses mixed research method to investigate sectors related to transaction governance in Taiwan's electronics industry. The study finds higher flexibility in contract implementation to be a promoter of trust between two parties in a relationship, thereby promoting project execution efficiency in the case of Taiwanese firms. Organisational management differs between the East and West; therefore, Western firms should understand how various contractual provisions can be used to accommodate different transactions when cooperating with Taiwanese electronics companies.




for

Talent development for the knowledge economy

The world's economies are attempting to transform themselves to have a greater focus on developing knowledge as a commodity through innovation. Innovation starts with a creative activity that yields an invention but is augmented through a systematic value driven knowledge management system to yield new knowledge that can create a competitive advantage. To succeed in such an economy, organisations must have or develop the talent that can produce and use information effectively, they must have an ambidextrous organisational structure that allows them to innovate and produce simultaneously, and they must have an innovation management system to sustain effective innovation. In this paper we show how to augment existing university courses to simultaneously develop subject matter and innovation skills in students. We also suggest the incorporation of the new Innovations Management System Standard Series ISO 56000 into business curricula to better prepare students to function in the knowledge economy.




for

An integrated framework for the alignment of stakeholder expectations with student learning outcomes

In this paper, two hypothetical frameworks are proposed through the application of quality function deployment (QFD) to integrate the current institutional level and program level student learning focus areas with the relevant institutional and program specific stakeholder expectations. A generic skillset proficiency expected of all the graduating students at the institutional level by the stakeholders is considered in the first QFD application example and a program specific knowledge proficiency expected at the program level by the stakeholders is considered in the second QFD application example. Operations management major/option is considered for illustration purposes at the program level. In addition, an assurance of learning based approach rooted in continuous improvement philosophy is proposed to align the stakeholder expectations with the relevant student learning outcomes at different learning tiers.




for

International Journal of Information and Operations Management Education




for

Hybrid encryption of Fernet and initialisation vector with attribute-based encryption: a secure and flexible approach for data protection

With the continuous growth and importance of data, the need for strong data protection becomes crucial. Encryption plays a vital role in preserving the confidentiality of data, and attribute-based encryption (ABE) offers a meticulous access control system based on attributes. This study investigates the integration of Fernet encryption with initialisation vector (IV) and ABE, resulting in a hybrid encryption approach that enhances both security and flexibility. By combining the advantages of Fernet encryption and IV-based encryption, the hybrid encryption scheme establishes an effective and robust mechanism for safeguarding data. Fernet encryption, renowned for its simplicity and efficiency, provides authenticated encryption, guaranteeing both the confidentiality and integrity of the data. The incorporation of an initialisation vector (IV) introduces an element of randomness into the encryption process, thereby strengthening the overall security measures. This research paper discusses the advantages and drawbacks of the hybrid encryption of Fernet and IV with ABE.




for

Resource monitoring framework for big raw data processing

Scientific experiments, simulations, and modern applications generate large amounts of data. Analysing resources required to process such big datasets is essential to identify application running costs for cloud or in-house deployments. Researchers have proposed keeping data in raw formats to avoid upfront utilisation of resources. However, it poses reparsing issues for frequently accessed data. The paper discusses detailed comparative analysis of resources required by in-situ engines and traditional database management systems to process a real-world scientific dataset. A resource monitoring framework has been developed and incorporated into the raw data query processing framework to achieve this goal. The work identified different query types best suited to a given data processing tool in terms of data to result time and resource requirements. The analysis of resource utilisation patterns has led to the development of query complexity aware (QCA) and resource utilisation aware (RUA) data partitioning techniques to process big raw data efficiently. Resource utilisation data have been analysed to estimate the data processing capacity of a given machine.




for

Characteristics of industrial service ecosystem practices for industrial renewal

The emergence of service ecosystems can accelerate the industrial renewal required because of urgent global challenges. However, existing research has not sufficiently grasped the social dynamics of coevolution in ecosystems that enhance industrial renewal. This study aimed to advance ecosystem research through a practice lens and to present the key characteristics of industrial service ecosystem practice involved in industrial renewal. Consequently, its three characteristics - <i>accomplishment</i>, <i>attractiveness</i> and <i>actionability</i> - were configured based on an abductive study derived from the ecosystem literature, three practice-oriented approaches to learning, and two case ecosystem examinations. These features created the logic for resource integration and enhanced ecosystems to evolve as units, thus exceeding the actors' independent avenues of renewal. The findings of this study provided a deeper understanding of the coevolution in ecosystems needed to accelerate industrial renewal as well as a novel conceptualisation of an <i>ecosystem-as-practice</i> for further studies.




for

Data as a potential path for the automotive aftersales business to remain active through and after the decarbonisation

This study aims to identify and understand the perspectives of automotive aftersales stakeholders regarding current challenges posed by decarbonisation strategies. It examines potential responses that the automotive aftersales business could undertake to address these challenges. Semi-structured interviews were undertaken with automotive industry experts from Europe and Latin America. This paper focuses primarily on impacts of decarbonisation upon automotive aftersales and the potential role of data in that business. Results show that investment in technology will be a condition for businesses that want to remain active in the industry. Furthermore, experts agree that incumbent manufacturers are not filling the technology gap that the energy transition is creating in the automotive sector, a consequence of which will be the entrance of new players from other sectors. The current aftersales businesses will potentially lose bargaining control. Moreover, policy makers are seen as unreliable leaders of the transition agenda.




for

Perceived service process in e-service delivery system: B2C online retailers performance ranking by TOPSIS

Significant work in service domain has focused on customer journey within e-service delivery system process (e-SDSP). Few studies have focused on process-centric approach to customer journey during delivery of e-services. This study aims to investigate the performance assessment of three online retailers (alternatives) using perceived service process during different stages of e-SDSP as a criterion for decision-making. TOPSIS is used in this paper to rate and evaluate multiple online retailers. Based on perceived service process as the criterion, results show that online retailer-2 outperforms other two online retailers. This study is one of the first to rate online retailers by utilising customer-perceived service process (latent variables) as a decision-making criterion throughout e-SDSP. The finding suggests that perceived searching process is the most essential criterion for decision-making, followed by the perceived after-sales service process, the perceived agreement process, and the perceived fulfilment process. Implications, limitations, and future scope are also discussed.




for

Modeling the Organizational Aspects of Learning Objects in Semantic Web Approaches to Information Systems




for

Developing Learning Objects for Secondary School Students: A Multi-Component Model




for

Learning Objects, Learning Object Repositories, and Learning Theory: Preliminary Best Practices for Online Courses




for

Teaching, Designing, and Sharing: A Context for Learning Objects




for

Practical Guidelines for Learning Object Granularity from One Higher Education Setting




for

A Framework for Metadata Creation Tools




for

Scoping and Sequencing Educational Resources and Speech Acts: A Unified Design Framework for Learning Objects and Educational Discourse




for

Decoupling the Information Application from the Information Creation: Video as Learning Objects in Three-Tier Architecture




for

Addressing the eLearning Contradiction: A Collaborative Approach for Developing a Conceptual Framework Learning Object




for

Learning Objects: Using Language Structures to Understand the Transition from Affordance Systems to Intelligent Systems




for

A Study of the Design and Evaluation of a Learning Object and Implications for Content Development




for

The OSEL Taxonomy for the Classification of Learning Objects




for

A New Learning Object Repository for Language Learning: Methods and Possible Outcomes




for

Guidelines and Standards for the Development of Fully Online Learning Objects




for

A Cognitive and Logic Based Model for Building Glass-Box Learning Objects




for

The Present and Future of Standards for E-Learning Technologies




for

Learning Objects and E-Learning: an Informing Science Perspective




for

Reading in A Digital Age: e-Books Are Students Ready For This Learning Object?




for

An Engagement Model for Learning: Providing a Framework to Identify Technology Services




for

An Integrated Approach for Automatic Aggregation of Learning Knowledge Objects




for

Practical E-Learning for the Faculty of Mathematics and Physics at the University of Ljubljana




for

Contextual Inquiry: A Systemic Support for Student Engagement through Reflection




for

Models for Sustainable Open Educational Resources




for

Learning Object Patterns for Programming




for

Ontology-Driven E-Learning System Based on Roles and Activities for Thai Learning Environment




for

Learning Pod: A New Paradigm for Reusability of Learning Objects