Specific User Research Articles

Reinforcement Learning of Beam Codebooks in Millimeter Wave and Terahertz MIMO Systems

Millimeter wave (mmWave) and terahertz MIMO systems rely on pre-defined beamforming codebooks for both initial access and data transmission. These pre-defined codebooks, however, are commonly not optimized for specific environments, user distributions, and/or possible hardware impairments. This leads to large codebook sizes with high beam training overhead which makes it hard for these systems to support highly mobile applications. To overcome these limitations, this paper develops a deep reinforcement learning framework that learns how to optimize the codebook beam patterns relying only on the receive power measurements. The developed model learns how to adapt the beam patterns based on the surrounding environment, user distribution, hardware impairments, and array geometry. Further, this approach does not require any knowledge about the channel, RF hardware, or user positions. To reduce the learning time, the proposed model designs a novel <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Wolpertinger</i> -variant architecture that is capable of efficiently searching the large discrete action space. The proposed learning framework respects the RF hardware constraints such as the constant-modulus and quantized phase shifter constraints. Simulation results confirm the ability of the developed framework to learn near-optimal beam patterns for line-of-sight (LOS), non-LOS (NLOS), mixed LOS/NLOS scenarios and for arrays with hardware impairments without requiring any channel knowledge.

PIP: Pictorial Interpretable Prototype Learning for Time Series Classification.

Time series classifiers are not only challenging to design, but they are also notoriously difficult to deploy for critical applications because end users may not understand or trust black-box models. Despite new efforts, explanations generated by other interpretable time series models are complicated for non-engineers to understand. The goal of PIP is to provide time series explanations that are tailored toward specific end users. To address the challenge, this paper introduces PIP, a novel deep learning architecture that jointly learns classification models and meaningful visual class prototypes. PIP allows users to train the model on their choice of class illustrations. Thus, PIP can create a user-friendly explanation by leaning on end-users definitions. We hypothesize that a pictorial description is an effective way to communicate a learned concept to non-expert users. Based on an end-user experiment with participants from multiple backgrounds, PIP offers an improved combination of accuracy and interpretability over baseline methods for time series classification.

Blasts from the Past: Reimagining Historical Storms with Model Simulations to Modernize Dam Safety and Flood Risk Assessment

Abstract Accurate estimation of the potential “upper limit” for extreme precipitation is critical for dam safety and water resources management, as dam failures pose significant risks to life and property. Methods used to estimate the theoretical upper limit of precipitation are often outdated and in need of updating. The rarity of extreme events means that old storms with limited observational data are often used to define the upper bound of precipitation. Observations of many important old storms are limited in spatial and temporal coverage, and sometimes of dubious quality. This reduces confidence in flood hazard assessments used in dam safety evaluations and leads to unknown or uncertain societal risk. This paper describes a method for generating and applying ensembles of high-resolution, state-of-the-art numerical model simulations of historical past extreme precipitation events to meet contemporary stakeholder needs. The method was designed as part of a research-to-application-focused partnership project to update state dam safety rules in Colorado and New Mexico. The results demonstrated multiple stakeholder and user benefits that were applied directly into storm analyses utilized for extreme rainfall estimation, and diagnostics were developed and ultimately used to update Colorado state dam safety rules, officially passed in January 2020. We discuss how what started as a prototype research foray to meet a specific user need may ultimately inform wider adoption of numerical simulations for water resources risk assessment, and how the historical event downscaling method performed offers near-term, implementable improvements to current dam safety flood risk estimates that can better serve society today.

Digital In Situ Data Collection in Earth Observation, Monitoring and Agriculture—Progress towards Digital Agriculture

Digital solutions in agricultural management promote food security and support the sustainable use of resources. As a result, remote sensing (RS) can be seen as an innovation for the fast generation of reliable information for agricultural management. Near real-time processed RS data can be used as a tool for decision making on multiple scales, from subplot to the global level. This high potential is not yet fully applied, due to often limited access to ground truth information, which is crucial for the development of transferable applications and acceptance. In this study we present a digital workflow for the acquisition, processing and dissemination of agroecological information based on proprietary and open-source software tools with state-of-the-art web-mapping technologies. Data is processed in near real-time and thus can be used as ground truth information to enhance quality and performance of RS-based products. Data is disseminated by easy-to-understand visualizations and download functionalities for specific application levels to serve specific user needs. It thus can increase expert knowledge and can be used for decision support at the same time. The fully digital workflow underpins the great potential to facilitate quality enhancement of future RS products in the context of precision agriculture by safeguarding data quality. The generated FAIR (findable, accessible, interoperable, reusable) datasets can be used to strengthen the relationship between scientists, initiatives and stakeholders.

The Effects of Gender Signals and Performance in Online Product Reviews.

This work quantifies the effects of signaling gender through gender specific user names, on the success of reviews written on the popular amazon.com shopping platform. Highly rated reviews play an important role in e-commerce since they are prominently displayed next to products. Differences in reviews, perceived—consciously or unconsciously—with respect to gender signals, can lead to crucial biases in determining what content and perspectives are represented among top reviews. To investigate this, we extract signals of author gender from user names to select reviews where the author’s likely gender can be inferred. Using reviews authored by these gender-signaling authors, we train a deep learning classifier to quantify the gendered writing style (i.e., gendered performance) of reviews written by authors who do not send clear gender signals via their user name. We contrast the effects of gender signaling and performance on the review helpfulness ratings using matching experiments. This is aimed at understanding if an advantage is to be gained by (not) signaling one’s gender when posting reviews. While we find no general trend that gendered signals or performances influence overall review success, we find strong context-specific effects. For example, reviews in product categories such as Electronics or Computers are perceived as less helpful when authors signal that they are likely woman, but are received as more helpful in categories such as Beauty or Clothing. In addition to these interesting findings, we believe this general chain of tools could be deployed across various social media platforms.

Using ISO 9241-11 To Identify How E-Commerce Companies Applied UX Guidelines

A lot of the company's business activities failed due to not adapting to user needs and technological developments. Previous studies show that there is no way to implement UX guidelines that explain the specific user needs for the UX of e-commerce systems. Therefore, we need a way of implementing UX for e-commerce websites. We used usability parameters in ISO 9241-11, namely effectiveness, efficiency, and satisfaction, to measure the system's usability and then conduct an interview to follow up the result. This research identifies how e-commerce companies implement UX best practices for their systems that can be used for other people who want to design their e-commerce applications.

Enterprise Resource Planning Systems in Iran

Enterprise Resource Planning (ERP) software packages have been widely implemented in the developed world since they first appeared on the market in the late 1970s. In the developing world, however, the lack of the human and financial resources needed for such projects, and the non-availability of sales and support offices for western-based ERP vendors, has limited the uptake of ERP software. However, in Iran, there has been significant recent growth in the development of homegrown ERP software solutions. These products are usually more customizable than western-based ERP products and can thus be adapted more readily to specific user requirements. Using a case study approach, this article examines the origins and development of one of these companies – the Behko software company, and assesses its main product, and how it has been implemented in a typical end-user company. The article concludes that the homegrown ERP sector in Iran is making a significant contribution to the provision of information systems in the country, and is competing successfully with western-based alternatives.

Chronic pain self-management in middle-aged and older adults: A collective intelligence approach to identifying barriers and user needs in eHealth interventions.

ObjectiveseHealth refers to health services and health information delivered or enhanced through the internet and related technologies. The number of eHealth interventions for chronic pain self-management is increasing. However, little evidence has been found for the overall efficacy of these interventions for older adults. The aim of the current study was to use a Collective Intelligence approach to identify the barriers and specific user needs of middle-aged and older adults using eHealth for chronic pain self-management.MethodsA Collective Intelligence workshop was conducted with middle-aged and older adults to generate, clarify, select, and structure ideas in relation to barriers to eHealth use and specific design requirements for the purposes of chronic pain self-management. Prior to attending the workshop, participants received a trigger question requesting the identification of five barriers to eHealth use for chronic pain self-management. These barriers were categorised and presented to the group along with barrier-related scenarios and user need prompts, resulting in the generation of a set of ranked barriers and a set of user needs.ResultsA total of 78 barriers were identified, from which six categories emerged: Content, Support, Technological, Personal, Computer Literacy and Accessibility. Additional idea-writing and group reflection in response to these barriers revealed 97 user needs.ConclusionThis is the first study to use Collective Intelligence methods to investigate barriers to eHealth technology use and the specific user needs of middle-aged and older adults in the context of chronic pain self-management. The results of the current study provide a platform for the design and development of enhanced eHealth interventions for this population.

Movie Recommendation Algorithm Based on Ensemble Learning

With the rapid development of personalized services, major websites have launched a recommendation module in recent years. This module will recommend information you are interested in based on your viewing history and other information, thereby improving the economic benefits of the website and increasing the number of users. This paper has introduced content-based recommendation algorithm, K-Nearest Neighbor (KNN)-based collaborative filtering (CF) algorithm and singular value decomposition-based (SVD) collaborative filtering algorithm. However, the mentioned recommendation algorithms all recommend for a certain aspect, and do not realize the recommendation of specific movies input by specific users which will cause the recommended content of the website to deviate from the need of users, and affect the experience of using. Aiming at this problem, this paper combines the above algorithms and proposes three ensemble recommendation algorithms, which are the ensemble recommendation of KNN + text, the recommendation of user KNN + movie KNN, and the recommendation of user KNN + singular value decomposition. Compared with the traditional collaborative filtering algorithm based on matrix factorization, the method we proposed can realize the recommendation of specific movies input by specific users and make more personalized recommendations and can deal with the problem of cold start and sparse matrix processing issues to a certain extent.

Methods and Applications of 3D Patient-Specific Virtual Reconstructions in Surgery.

3D modelling has been highlighted as one of the key digital technologies likely to impact surgical practice in the next decade. 3D virtual models are reconstructed using traditional 2D imaging data through either direct volume or indirect surface rendering. One of the principal benefits of 3D visualisation in surgery relates to improved anatomical understanding-particularly in cases involving highly variable complex structures or where precision is required.Workflows begin with imaging segmentation which is a key step in 3D reconstruction and is defined as the process of identifying and delineating structures of interest. Fully automated segmentation will be essential if 3D visualisation is to be feasibly incorporated into routine clinical workflows; however, most algorithmic solutions remain incomplete. 3D models must undergo a range of processing steps prior to visualisation, which typically include smoothing, decimation and colourization. Models used for illustrative purposes may undergo more advanced processing such as UV unwrapping, retopology and PBR texture mapping.Clinical applications are wide ranging and vary significantly between specialities. Beyond pure anatomical visualisation, 3D modelling offers new methods of interacting with imaging data; enabling patient-specific simulations/rehearsal, Computer-Aided Design (CAD) of custom implants/cutting guides and serves as the substrate for augmented reality (AR) enhanced navigation.3D may enable faster, safer surgery with reduced errors and complications, ultimately resulting in improved patient outcomes. However, the relative effectiveness of 3D visualisation remains poorly understood. Future research is needed to not only define the ideal application, specific user and optimal interface/platform for interacting with models but also identify means by which we can systematically evaluate the efficacy of 3D modelling in surgery.

Integrated multi-material portable 3D-printed platform for electrochemical detection of dopamine and glucose.

3D-printing has become a fundamental part of research in many areas of investigation since it provides rapid and personalized production of parts that meet very specific user needs. Biosensing is not an exception, and production of electrochemical sensors that can detect a variety of redox mediators and biologically relevant molecules has been widely reported. However, most 3D-printed electrochemical sensors detailed in the literature rely on big, individual, single-material electrodes that require large sample volumes to perform effectively. Our work exploits multi-material fused filament fabrication 3D-printing to produce a compact electrochemical sensor able to operate with only 100 μL of sample. We report cyclic voltammetry, differential pulse voltammetry, and chronoamperometry results to assess sensor performance and sensitivity. We investigated the influence of layer print orientation and layer thickness on the electrochemical performance of the sensor, and used the optimal parameters to produce the final device. The integrated 3D-printed platform successfully detects electrochemical activity for hexaammineruthenium(III) chloride and potassium ferricyanide (0.1 mM to 2 mM in 100 mM KCl), dopamine (50 μM to 1 mM in 1×PBS), and glucose via mediated amperometric glucose oxidase enzyme-based sensing (1 mM to 12 mM in 1×PBS), indicating good acceptance of biological modification. These results reveal the exciting potential of multi-material 3D-printing and how it can be used for the rapid development of efficient, small, integrated, personalized electrochemical biosensors.

Defining mobility packages by using city specific parameters and user groups: a case study

By taking city specific parameters into account, this paper explains the definition of mobility packages for users in the Mobility as a Service framework. Therefore, a method that creates package levels for various transportation modes in cities is elaborated. The method involves urban characteristics, the aspects of city structure, environmental awareness, modal split, and financial features, as well. Mobility packages are created for public transport, bike-sharing, car-sharing, and taxi. In order to create more diverse mobility packages, the preferences of user groups are considered, and special packages are prepared for travelers. The method is tested with a short survey in Budapest, Hungary. The results show that the workers’ package is similar to the basic package, while in case of students and pensioners, the service level of car-sharing and taxi is lowered based on the revealed usage requirements.

Hierarchical Stream Clustering Based NEWS Summarization System

News feed is one of the potential information providing sources which give updates on various topics of different domains. These updates on various topics need to be collected since the domain specific interested users are in need of important updates in their domains with organized data from various sources. In this paper, the news summarization system is proposed for the news data streams from RSS feeds and Google news. Since news stream analysis requires live content, the news data are continuously collected for our experimentation. The major contributions of this work involve domain corpus based news collection, news content extraction, hierarchical clustering of the news and summarization of news. Many of the existing news summarization systems lack in providing dynamic content with domain wise representation. This is alleviated in our proposed system by tagging the news feed with domain corpuses and organizing the news streams with the hierarchical structure with topic wise representation. Further, the news streams are summarized for the users with a novel summarization algorithm. The proposed summarization system generates topic wise summaries effectively for the user and no system in the literature has handled the news summarization by collecting the data dynamically and organizing the content hierarchically. The proposed system is compared with existing systems and achieves better results in generating news summaries. The Online news content editors are highly benefitted by this system for instantly getting the news summaries of their domain interest.

Mobile Devices Supporting People with Special Needs

Over the years, various devices designed for people with special needs have been used for a while and then replaced with modern devices to make everyday life easier. The development of mobile devices is especially improved today and through them different everyday activities are facilitated, not only by people with special needs. The purpose of this paper is to present some of the modern mobile devices with an analysis of their operating systems, functionalities, applications and design. Based on the research, their usability for both sighted and visually and hearing-impaired users is described. Attention is paid to the preferences formed among users when using specialized applications developed for mobile devices. Based on a survey of specific target user groups, the paper provides summary results to support the thesis on the importance of the facilities, offered by modern mobile devices.

A Novel Dictionary Generation Methodology for Contextual-Based Password Cracking

It has been over 50 years since the concept of passwords was introduced and adopted in our society as a digital authentication method. Despite alternative authentication methods being developed since, it is reasonable to assume that this prevailing method of authentication will not be toppled anytime soon. Naturally, each password is tightly connected to its creator. This connection has given rise to advanced techniques aimed at exploiting user habits for password cracking. Such techniques are often generic approaches leveraging large datasets of human created passwords. A 2021 study showed that the online identity of almost one in three Americans was stolen in the last year alone, with a further 13% not being sure if their credentials were also breached [1]. Recent research has underlined the influence that context can have during password selection for a user. Such information could be of significant added value when digital investigators need to target a specific user or group of users during a criminal investigation. Besides manual techniques, there are no automated approaches that can extract and utilize contextual information during password cracking processes. In this paper, a methodology and framework for creating custom dictionary lists for dictionary attacks are introduced, with a specific focus on leveraging the contextual information encountered during an investigation. Furthermore, a detailed explanation the framework&#x2019;s implementation is provided and the benefits of the approach are demonstrated with the use of test cases. This demonstrates the benefits of context in password cracking.

Real-Time Focused Extraction of Social Media Users

In this paper, we explore a real-time automation challenge: the problem of focused extraction of Social Media users. This challenge can be seen as a special form of focused crawling where the main target is to detect users with certain patterns. Given a specific user profile, the task consists of rapidly ingesting Social Media data and early detecting target users. This is a real-time intelligent automation task that has numerous applications in domains such as safety, health or marketing. The volume and dynamics of Social Media contents demand efficient real-time solutions able to predict which users are worth to explore. To meet this aim, we propose and evaluate several methods that effectively allow us to harvest relevant users. Even with little contextual information (e.g., a single user submission), our methods quickly focus on the most promising users. We also developed a distributed microservice architecture that supports real-time parallel extraction of Social Media users. This modular architecture scales up in clusters of computers and it can be easily adapted for user extraction in multiple domains and Social Media sources. Our experiments suggest that some of the proposed prioritisation methods, which work with minimal user context, are effective at rapidly focusing on the most relevant users. These methods perform satisfactorily with huge volumes of users and interactions and lead to harvest ratios 2 to 9 times higher than those achieved by random prioritisation.

NEAR: Non-supervised Explainability Architecture for Accurate Review-Based Collaborative Filtering

There is a critical issue in explainable recommender systems that compounds the challenges of explainability yet is rarely tackled: the lack of ground-truth explanation texts for training. It is unrealistic to expect every user-item pair in a dataset to have a corresponding target explanation. Hence, we pioneer the first non-supervised explainability architecture for review-based collaborative filtering (called NEAR) as our novel contribution to the theory of explanation construction in recommender systems. While maintaining excellent recommendation performance, our approach reformulates explainability as a non-supervised (i.e., unsupervised and self-supervised) explanation generation task. We formally define two explanation types, both of which NEAR can produce. An invariant explanation, fixed for all users, is based on the unsupervised extractive summary of an item's reviews via embedding clustering. Meanwhile, a variant explanation, personalized for a specific user, is a sentence-level text generated by our customized Transformer conditioned on every user-item-rating tuple and artificial ground-truth (self-supervised label) from one of the invariant explanation's sentences. Our empirical evaluation illustrates that NEAR's rating prediction accuracy is better than the other state-of-the-art baselines. Moreover, experiments and assessments show that NEAR-generated variant explanations are more personalized and distinct than those from other Transformer-based models, and our invariant explanations are preferred over those from other contemporary models in real life.

A Prototype for Covid Risk Assessment: Designing a Chatbot for Users to Determine Personalized Risk of Severe Outcome

This paper overviews the creation of a COVID Risk Assessment (CRA) Chatbot prototype designed to support recommendations for individuals based on the assessment of their COVID-related risk profile. Among rapidly emerging digital technologies, AI-enabled chatbots have the potential to offer real-time interaction with users via immediate responses, notifications, and reminders; interestingly, these intelligent agents can also be designed to offer personalized advice based on specific user characteristics and may even be scaled to interface with many users concurrently. Drawing from prior research, we designed a CRA-chatbot prototype to incorporate a variety of parameters affecting the severity of COVID-19 systems. These factors include the number of vaccine doses, the date of last vaccination, previous COVID-19 infection history, age, and risk factors such as disease history. Broadly, the prototype is envisioned to take the users’ input with their personalized risk assessment of severe symptoms to further educate them; in so doing, suggestions for treatments, additional serological testing, and supplementary vaccine doses can then be made. We conclude that, within the current learning health system environments, AI-enabled chatbots are digital health tools well-suited for empowering personalized COVID-19 education and related knowledge dissemination.

Scoping Review of the Literature on Smart Healthcare for Older Adults.

Smart healthcare systems are being designed to provide medical services to and improve the daily lives of older adults. However, most research has been focused on technical issues, despite a need to conduct in-depth studies on related ethical issues. Therefore, this study aimed to examine ethical issues in smart healthcare for older adults. We reviewed published literature using PubMed. In total, 292 documents were analyzed by applying the scoping review method. Finally, 29 articles were selected from the 292 articles. Ethical issues in smart healthcare for older adults were analyzed in terms of the themes of responsibility/autonomy (n=10), privacy (n=9), and digital divide (n=10). Technical help provided by smart healthcare may infringe on the autonomy of tacit choice for older adults. This pose a potential ethical issue as the subject of responsibility here is unclear. Privacy is a concern as smart technology may intrude the personal life of the user. The digital divide is a challenge because of low responsiveness from older adults to technological changes. The future development and application of smart healthcare systems must take these ethical aspects into account to enable their efficient and effective use in supplementing healthcare for older adults. Critical discussions to identify ethical issues and customize ethical requirements for specific user needs are necessary among smart healthcare providers.

Community Detection in Social Networks Considering Social Behaviors

The study of community detection in networks has drawn great attention in recent years. To find communities and to understand community semantics, both network topology and network content are utilized. Unfortunately, none of them can explain the driving factors of generating community structure with semantics, which is significant for understanding the mechanisms of community generation. Our observations on a large number of networks show that specific user social behaviors are underlying factors for the generation of community structure. We exploit four types of social behaviors that widely exist in networks, i.e., reciprocity of interactions, posting preference, multitopic preference, and temporal variation of topics. We investigate their impacts on the formation process of links and content in networks, during which communities with topics form. Our analysis shows that they are highly related to community structure. Consequently, a generative community detection model SBCD (social behavior-based community detection) is proposed by combining network topology and content, in which the above social behaviors play a core role. The model is evaluated on two real datasets. The experimental results show that SBCD outperforms state-of-the-art baselines. Finally, a case study illustrates several significant observations with respect to the proposed social behaviors.