Artificial intelligence (AI) has emerged as a promising tool for predicting COVID-19 from medical images. In this paper, we propose a novel continual learning (also known as incremental learning)-based approach and present the design and implementation of a smartphone application for screening COVID-19. Our approach demonstrates the ability to adapt to evolving datasets, including data collected from different locations or hospitals, varying virus strains, and diverse clinical presentations, without retraining from scratch. We have evaluated state-of-the-art continual learning methods for detecting COVID-19 from chest X-rays and selected the best-performing model for our mobile app. We evaluated various deep learning architectures to select the best-performing one as a foundation model for continual learning. Both regularization and memory-based methods for continual learning were tested, using different memory sizes to develop the optimal continual learning model for our app. DenseNet161 emerged as the best foundation model with 96.87% accuracy, and Learning without Forgetting (LwF) was the top continual learning method with an overall performance of 71.99%. The mobile app design considers both patient and doctor perspectives. It incorporates the continual learning DenseNet161 LwF model on a cloud server, enabling the model to learn from new instances of chest X-rays and their classifications as they are submitted. The app is designed, implemented, and evaluated to ensure it provides an efficient tool for COVID-19 screening. The app is available to download from https://github.com/DannyFGitHub/COVID-19PneumoCheckApp .

• Robust COVID-19 cough classification using neural decision tree and forest. • Experimentation across five datasets and their merged set highlights dataset diversity. • Cross-datasets analyses show demographic variability in COVID-19 cough sounds. • RFECV and Bayesian optimization improved feature selection and model performance. • SMOTE oversampling and threshold moving enhanced data balance and classification. This research presents a robust approach to classifying COVID-19 cough sounds using cutting-edge machine learning techniques. Leveraging deep neural decision trees and deep neural decision forests, our methodology demonstrates consistent performance across diverse cough sound datasets. We begin with a comprehensive extraction of features to capture a wide range of audio features from individuals, whether COVID-19 positive or negative. To determine the most important features, we use recursive feature elimination along with cross-validation. Bayesian optimization fine-tunes hyper-parameters of deep neural decision tree and deep neural decision forest models. Additionally, we integrate the synthetic minority over-sampling technique during training to ensure a balanced representation of positive and negative data. Model performance refinement is achieved through threshold optimization, maximizing the ROC-AUC score. Our approach undergoes a comprehensive evaluation in five datasets: Cambridge (asymptomatic and symptomatic), Coswara, COUGHVID, Virufy, and the combined Virufy with the NoCoCoDa dataset. Consistently outperforming state-of-the-art methods, our proposed approach yields notable AUC scores of 0.97, 0.98, 0.92, 0.93, 0.99, and 0.99, alongside remarkable precision scores of 1, 1, 0.72, 0.93, 1, and 1 across the respective datasets. Merging all datasets into a combined dataset, our method, using a deep neural decision forest classifier, achieves an accuracy of 0.97, AUC of 0.97, precision of 0.95, recall of 0.96, F1-score of 0.96, and specificity score of 0.97. Also, our study includes a comprehensive cross-datasets analysis, revealing demographic and geographic differences in the cough sounds associated with COVID-19. These differences highlight the challenges in transferring learned features across diverse datasets and underscore the potential benefits of dataset integration, improving generalizability and enhancing COVID-19 detection from audio signals. The code used to generate the reported results is available at https://github.com/Rofiquldk1/COVID-19-Detection-from-Cough-Sound

Abstract Accurate, non-destructive assessment of egg quality is critical for ensuring food safety, maintaining product standards, and operational efficiency in commercial poultry production. This paper introduces ELMF4EggQ , an ensemble learning framework that employs multimodal feature fusion to classify egg grade and freshness using only external attributes – image, shape, and weight. A novel, publicly available dataset of 186 brown-shelled eggs was constructed, with egg grade and freshness levels determined through laboratory-based expert assessments involving internal quality measurements, such as yolk index and Haugh unit. To the best of our knowledge, this is the first study to apply machine learning methods for internal egg quality assessment using only external, non-invasive features, and the first to release a corresponding labeled dataset. The proposed framework integrates deep features extracted from external egg images with structural characteristics such as egg shape and weight, enabling a comprehensive representation of each egg. Image feature extraction is performed using top-performing pre-trained CNN models (ResNet152, DenseNet169, and ResNet152V2), followed by principal component analysis (PCA)-based dimensionality reduction, synthetic minority oversampling technique (SMOTE) augmentation, and classification using multiple machine learning algorithms. An ensemble voting mechanism combines predictions from the best-performing classifiers to enhance overall accuracy. Experimental results demonstrate that the multimodal approach significantly outperforms the image-only and tabular-only (shape and weight) baselines, with the multimodal ensemble approach achieving an accuracy of 82.24% (SD $$\pm {2.41\%}$$ ) in grade classification and 70.41% (SD $$\pm {3.20\%}$$ ) in freshness prediction. The framework demonstrates strong potential for real-time, low-cost deployment in commercial egg processing environments. It highlights the feasibility of using computer vision and lightweight structural inputs for scalable, non-invasive egg quality evaluation. All code and data are publicly available at https://github.com/Kenshin-Keeps/Egg_Quality_Prediction_ELMF4EggQ , promoting transparency, reproducibility, and further research in this domain.

Abstract Maintaining constant vigilance over arterial blood pressure (ABP) is crucial for diagnosing hypertension and other critical cardiovascular diseases. While traditional cuff-based approaches are non-invasive, they have limitations in providing continuous blood pressure monitoring. In contrast, complex ABP monitoring systems, while accurate, are primarily suitable for clinical settings due to their intrusive nature. This study introduces a groundbreaking method for generating arterial blood pressure (ABP) waveforms using remote radar signals and deep learning (DL) techniques. This approach eliminates the need for invasive procedures, wearable biosensors, and costly equipment typically associated with ABP recording. We introduce MultiResLinkNet, a segmentation model based on a one-dimensional convolutional neural network (1D CNN), specifically designed to synthesize arterial blood pressure (ABP) directly from raw radar waveforms. We trained and evaluated the end-to-end DL framework using a publicly available benchmark radar dataset containing raw radar data and corresponding physiological signals from 30 subjects across various scenarios, including Resting, Valsalva, Apnea, Tilt-up, and Tilt-down. The proposed MultiResLinkNet excelled in ABP segmentation, outperforming state-of-the-art networks in combined and individual scenarios, and produced the best average temporal and spectral correlations as well as the lowest temporal and spectral errors in nearly all scenarios’ data. Furthermore, qualitative evaluation demonstrated a strong resemblance between the synthesized and ground truth ABP waveforms. Our novel approach enables remote monitoring of critical patients continuously, especially those undergoing surgery, by predicting ABP waveforms from non-contact radar signals. This breakthrough offers significant advantages, facilitating continuous ABP monitoring without the need for invasive procedures or cumbersome wearable sensors.

The challenge of designing against child sexual abuse becomes more complicated in conservative societies where talking about sex is tabooed. Our mix-method study, comprised of an online survey, five FGDs, and 20 semi-structured interviews in Bangladesh, investigates the common nature, location, and time of the abuse, post-incident support, and possible combating strategies. Besides revealing important facts, our findings highlight the need of decentering the design from the victims (children and/or guardians) to the community. Hence, building on the theory of transformative justice, we prototyped and evaluated ‘ShishuShurokkha’ – an online tool that involves the whole community by allowing anonymous bystander reporting, visualizing case-maps, connecting with legal, medical, and social support, and raising awareness. The evaluation of ShishuShurokkha shows the promise for such a communal approach toward combating child sexual abuse, and highlights the needs for sincere involvement of the government, NGOs, the legal, educational, and religious services in this.

Having precise specifications of service APIs is essential for many Software Engineering activities. Unfortunately, available documentation of services is often inadequate and/or imprecise and, hence, cannot be fully relied upon. Generating service documentation manually is a tedious and error-prone task, especially in light of changes to services. Therefore, there is a need for automated support in generating service documentation. In this work, we present a novel approach to infer the API of a service by analyzing recorded messages sent to and received from this service. Our approach includes a novel, two-level clustering technique to cluster messages, a step that many existing approaches to infer message formats fail to perform precisely in the presence of significant variation of payload information of the available messages. We have evaluated our approach on message traces from four different real-world services. The experimental result shows that our approach is more effective than existing techniques in extracting correct message formats from recorded messages.

HCI has a dearth of knowledge in understanding how religiosity, spirituality, and ideological values and practices shape the notion of privacy and guide information practices worldwide. In this paper, we fill this gap by reporting our findings from an eight-month-long ethnographically informed study in Bangladeshi Islamic communities. We report how the Islamic spirit of purdah, amanah, gheebat, riya, and buhtan represent the notion of privacy and guide privacy practices among “pious” Bangladeshi Muslims. We further discuss how sacred values generate norms and customs associated with privacy and surveillance. Finally, we recommend how a nuanced understanding of divine interests, identity performance, family surveillance, and spatial privacy norms help designing for inclusive privacy in the Global South. This paper makes a novel contribution to HCI by providing a new analytical perspective to understand privacy and design privacy-preserving technologies and tools for regions where religiosity, spirituality, and sacred values play a dominant role.

The impact of road traffic incidents (e.g., road accidents, vehicle breakdowns) have become progressively worse over the years, being a major cause of many adverse issues such as serious injury, economic loss, and lifelong disabilities. Thus, it is essential to acknowledge these issues and proactively construct appropriate solutions to mitigate the impact of these issues in the future. This study outlines the history of traffic incident research and covers several solutions such as machine learning, mathematical modeling, and visualization system to traffic incident analysis. In this paper, we design a unique visualization system, ExTraVis, for incident data exploration and analysis that can be used to help traffic management controllers, aid to make decisions, and help them to understand how past incidents affected and where incidents may occur. The key features of this system are visual exploration and analysis to overcome the problems linked with road traffic incidents and to encourage future work and improvements. Additionally, we gather various custom queries for free text search feature. We find that people ask questions and our system provide 90% correct visual insights. Finally, we demonstrate the effectiveness and robustness of ExTraVis by comparing with three different incident visualization dashboards and a user study.

Today Information in the world wide web is overwhelmed by unprecedented quantity of data on versatile topics with varied quality. However, the quality of information disseminated in the field of medicine has been questioned as the negative health consequences of health misinformation can be life-threatening. There is currently no generic automated tool for evaluating the quality of online health information spanned over broad range. To address this gap, in this paper, we applied data mining approach to automatically assess the quality of online health articles based on 10 quality criteria. We have prepared a labelled dataset with 53012 features and applied different feature selection methods to identify the best feature subset with which our trained classifier achieved an accuracy of [Formula: see text] varied over 10 criteria. Our semantic analysis of features shows the underpinning associations between the selected features & assessment criteria and further rationalize our assessment approach. Our findings will help in identifying high quality health articles and thus aiding users in shaping their opinion to make right choice while picking health related help from online.

For an unfamiliar Application Programming Interface (API), software developers often access the official documentation to learn its usage, and post questions related to this API on social question and answering (Q&A) sites to seek solutions. The official software documentation often captures the information about functionality and parameters, but lacks detailed descriptions in different usage scenarios. On the contrary, the discussions about APIs on social Q&A sites provide enriching usages. Moreover, existing code search engines and information retrieval systems cannot effectively return relevant software documentation when the issued query does not contain code snippets or API-like terms. In this paper, we present CnCxL2RCnCxL2R, a software documentation recommendation strategy incorporating the content of official documentation and the social context on Q&A into a learning-to-rank schema. In the proposed strategy, the content, local context and global context of documentation are considered to select candidate documents. Then four types of features are extracted to learn a ranking model. We conduct a large-scale automatic evaluation on Java documentation recommendation. The results show that CnCxL2RCnCxL2R achieves state-of-the-art performance over the eight baseline models. We also compare the CnCxL2RCnCxL2R with Google search. The results show that CnCxL2RCnCxL2R can recommend more relevant software documentation, and can effectively capture the semantic between the high-level intent in developers' queries and the low-level implementation in software documentation.

Typically, people perform visual data analysis using mouse and touch interactions. While such interactions are often easy to use, they can be inadequate for users to express complex information and may require many steps to complete a task. Recently natural language interaction has emerged as a promising technique for supporting exploration with visualization, as the user can express a complex analytical question more easily. In this paper, we investigate how to synergistically combine language and mouse-based direct manipulations so that weakness of one modality can be complemented by the other. To this end, we have developed a novel system, named Multimodal Interactions System for Visual Analysis (MIVA), that allows user to provide input using both natural language (e.g., through speech) and direct manipulation (e.g., through mouse or touch) and presents the answer accordingly. To answer the current question in the context of past interactions, the system incorporates previous utterances and direct manipulations made by the user within a finite-state model. We tested the applicability of MIVA on several dashboards including a COVID-19 dashboard that visualizes coronavirus cases around the globe. Our demonstration provides initial indication that the MIVA system enhances the flow of visual analysis by enabling fluid, iterative exploration and refinement of data in a dashboard with multiple-coordinated views.

APIs play a significant role in the sharing, utilization and integration of information and service assets for enterprises, delivering significant business value. However, the documentation of service APIs can often be incomplete, ambiguous, or even non-existent, hindering API-based application development efforts. In this paper, we introduce an approach to automatically mine the fine-grained message formats required in defining the APIs of services and applications from their interaction traces, without assuming any prior knowledge. Our approach includes three major steps with corresponding techniques: (1) classifying the interaction messages of a service into clusters corresponding to message types, (2) identifying the keywords of messages in each cluster, and (3) extracting the format of each message type. We have applied our approach to network traces collected from four real services which used the following application protocols: REST, SOAP, LDAP and SIP. The results show that our approach achieves much greater accuracy in extracting message formats for service APIs than current state-of-art approaches.

Mobile phones can record individual’s daily behavioral data as a time-series. In this paper, we present an effective time-series segmentation technique that extracts optimal time segments of individual’s similar behavioral characteristics utilizing their mobile phone data. One of the determinants of an individual’s behavior is the various activities undertaken at various times-of-the-day and days-of-the-week. In many cases, such behavior will follow temporal patterns. Currently, researchers use either equal or unequal interval-based segmentation of time for mining mobile phone users’ behavior. Most of them take into account static temporal coverage of 24-h-a-day and few of them take into account the number of incidences in time-series data. However, such segmentations do not necessarily map to the patterns of individual user activity and subsequent behavior because of not taking into account the diverse behaviors of individuals over time-of-the-week. Therefore, we propose a behavior-oriented time segmentation (BOTS) technique that takes into account not only the temporal coverage of the week but also the number of incidences of diverse behaviors dynamically for producing similar behavioral time segments over the week utilizing time-series data. Experiments on the real mobile phone datasets show that our proposed segmentation technique better captures the user’s dominant behavior at various times-of-the-day and days-of-the-week enabling the generation of high confidence temporal rules in order to mine individual mobile phone users’ behavior.

Graphical models for business processes are very large and cumbersome to build. Reusable process patterns can make this modeling task much easier. While using reusable components is a well-explored subject in software engineering, not much has been done in the context of business process modeling. In this paper, we will present an extension to Business Process Model and Notation (BPMN), the standard notation for modeling business processes, in the form of reusable Process Patterns. We introduce a type system for these patterns and use it to define a valid embedding of a process pattern in a larger model. We also introduce the formal notations and show that business processes modeled using our extended notation can be translated to BPMN. We present a case study to demonstrate the applicability of the process pattern and further quantify its characteristics using a set of criteria. We also implement a modeling tool for users to model business process using process patterns.

Mobile phone log data (e.g., phone call log) is not static as it is progressively added to day-by-day according to individual's diverse behaviors with mobile phones. Since human behavior changes over time, the most recent pattern is more interesting and significant than older ones for predicting individual's behavior. The goal of this poster paper is to identify the recent behavioral data length dynamically from the entire phone log for recency-based behavior modeling. To the best of our knowledge, this is the first dynamic recent log-based study that takes into account individual's recent behavioral patterns for modeling their phone call behaviors.

Knowledge graph is useful for many different domains like search result ranking, recommendation, exploratory search, etc. It integrates structural information of concepts across multiple information sources, and links these concepts together. The extraction of domain specific relation triples (subject, verb phrase, object) is one of the important techniques for domain specific knowledge graph construction. In this research, an automatic method named HDSKG is proposed to discover domain specific concepts and their relation triples from the content of webpages. We incorporate the dependency parser with rule-based method to chunk the relations triple candidates, then we extract advanced features of these candidate relation triples to estimate the domain relevance by a machine learning algorithm. For the evaluation of our method, we apply HDSKG to Stack Overflow (a Q&A website about computer programming). As a result, we construct a knowledge graph of software engineering domain with 35279 relation triples, 44800 concepts, and 9660 unique verb phrases. The experimental results show that both the precision and recall of HDSKG (0.78 and 0.7 respectively) is much higher than the openIE (0.11 and 0.6 respectively). The performance is particularly efficient in the case of complex sentences. Further more, with the self-training technique we used in the classifier, HDSKG can be applied to other domain easily with less training data.

Mobile or cellular phones can record various types of context data related to a user's phone call activities. In this paper, we present an approach to discovering individualized behavior rules for mobile users from their phone call records, based on the temporal context in which a user accepts, rejects or misses a call. One of the determinants of an individual's phone behavior is the various activities undertaken at various times of a day and days of the week. In many cases, such behavior will follow temporal patterns. Currently, researchers modeling user behavior using temporal context statically segment time into arbitrary categories (e.g., morning, evening) or periods (e.g., 1 hour). However, such time categorization does not necessarily map to the patterns of individual user activity and subsequent behavior. Therefore, we propose a behavior-oriented time segmentation (BOTS) technique that dynamically identifies diverse time segments for an individual user's behaviors based on the phone call records. Experiments on real datasets show that our proposed technique better captures the user's dominant call response behavior at various times of the day and week, thereby enabling more appropriate rules to be created for the purpose of automated handling of incoming calls, in an intelligent call interruption management system.

The electronic calendar usually serves as a personal organizer and is a valuable resource for managing daily activities or schedules of the users. Naturally, a calendar provides various contextual information about individual's scheduled events/appointments, e.g., meeting. A number of researchers have utilized such information to predict human behavior for mobile communication, by assuming a predefined event-behavior mapping which is static and non-personalized. However, in the real world, people differ from each other in how they respond to incoming calls during their scheduled events, even a particular individual may respond differently subject to what type of event is scheduled in the calendar. Thus a static behavioral model does not necessarily map to calendar schedules and corresponding phone call response behavior of individuals. Therefore, we propose an evidencebased behavioral model (EBM) that dynamically identifies the actual call response behavior of individuals for various calendar events based on their mobile phone log that records the data related to a user's phone call activities. Experiments on real datasets show that our proposed technique better captures the user's call response behavior for various calendar events, thereby enabling more appropriate rules to be created for the purpose of automated handling of incoming calls in an intelligent call interruption management system.

With the convergence of pervasive mobile computing and social networking, interest has grown significantly in software systems and applications that are aware of users' social context to make pervasive applications more intelligent and accessible. Thus, socially-aware systems have further advanced context-aware systems taking account of human social context such as social relationships to enable the attainment of users' tasks in different domains. However, social context-awareness introduces a variety of software engineering challenges. In this paper, we address these challenges by proposing a software engineering process that provides a methodological framework for developing various types of socially-aware applications from requirements elicitation through to concrete implementation. We provide context models and software infrastructure to assist developers in rapid prototyping. We also present two case studies to demonstrate the feasibility and applicability of our software engineering process by presenting how this process can be used to develop two different types of socially-aware applications utilizing our model and infrastructure. Finally, we evaluate our software engineering approach with respect to a set of software quality metrics.

With the advances of Internet technologies and an explosive growth in the popularity of social media, an increasingly large part of human life is getting digitized and becoming available on the web. This phenomenon brings opportunities and motivates us to infer users’ situations by exploiting their interaction events in various social media such as online social networks, blogs and email. One of the key requirements of inferring situations from interaction events is to consider both the semantic and temporal aspects of events in the situation inference process. In this paper, we address this issue and propose a novel approach to exploiting users’ interaction events in social media to infer their situations. We present an ontology-based interaction event model that captures the properties of users’ interaction activities in social media. We further provide a rule-based situation specification technique that integrates the interaction event ontology (for semantically matching interaction events) with temporal event relationships (for correlating historical interaction events). We also provide a platform to realize the situation reasoning/inference process, which combines semantic matching and complex event processing. We conduct a performance evaluation of the platform to quantify its efficacy. The feasibility and applicability of our approach is demonstrated by developing a socially aware phone call application as a case study.

We present a social context as a service (SCaaS) platform for managing adaptations in collaborative pervasive applications that support interactions among a dynamic group of actors such as users, stakeholders, infrastructure services, businesses and so on. Such interactions are based on predefined agreements and constraints that characterize the relationships between the actors and are modeled with the notion of social context. In complex and changing environments, such interaction relationships, and thus social contexts, are also subject to change. In existing approaches, the relationships among actors are not modeled explicitly, and instead are often hard-coded into the application. Furthermore, these approaches do not provide adequate adaptation support for such relationships as the changes occur in user requirements and environments. In our approach, inter-actor relationships in an application are modeled explicitly using social contexts, and their execution environment is generated and adaptations are managed by the SCaaS platform. The key features of our approach include externalization of the interaction relationships from the applications, representation and modeling of such relationships from the domain and actor perspectives, their implementation using a service oriented paradigm, and support for their runtime adaptation. We quantify the platform's adaptation overhead and demonstrate its feasibility and applicability by developing a telematics application that supports cooperative convoy.

The convergence of Internet and mobile devices has radically changed the way people communicate and interact with each other, and demand for applications that are "social" enough to assist their daily interactions. To support such device mediated interactions, the social relationships between actors need to be systematically modeled and represented. In addition, an application facilitating such interactions should be able to deal with the task conflicts that occur when an actor is involved in multiple interactions simultaneously. To address these issues, in this paper we present an approach to modeling and coordinating social interactions with the notion of social context. It supports social interaction modeling from both the domain- and player-centric perspectives. In particular, the player-centric model provides the basis to coordinate multiple interactions in which an actor is involved. We further introduce a fuzzy logic based reasoning technique to infer the overall importance of each interaction, assisting the actor to resolve conflicts and make decisions. Finally, we validate our approach through a prototype implementation and test cases analysis.

In RFID systems, middleware is used to filter enormous streaming data gathered continuously from readers to process application requests. The high volume of data makes middleware often in a highly overloaded situation. Nowadays, readers are becoming smart and provide filtering functionality. The reader filtering capability can be used to reduce data volume as well as middleware work-load. However, if middleware dispatches query conditions to reader without any adjustment, it may generate huge amount of duplicate data which imposes considerable load on the middleware. So, the appropriate schema of data volume reduction is required. In this paper, we propose a query decomposition technique to divide queries into sub-queries for middleware and reader level execution. This new approach of query execution resolves the problem of duplicate data generation. Our experiments show that the proposed approach considerably improves the performance of middleware by reducing the query processing time and the network traffic between reader and middleware.