Enable Domain Experts Research Articles

Information fusion-based Bayesian optimized heterogeneous deep ensemble model based on longitudinal neuroimaging data

The fusion of multimodal longitudinal data is difficult but crucial for enhancing the accuracy of deep learning models for disease identification and helps provide tailored and patient-centric decisions. This study explores the fusion of multimodal data to detect the progression of Alzheimer’s disease (AD) using ensemble learning. We propose a heterogeneous ensemble framework of Bayesian-optimized time-series deep learning models to identify progressive deterioration of brain damage. Experimental results show that the heterogeneous ensemble of three models with patient’s temporal data outperforms all other variants of ensemble models by achieving an average performance of 95% for accuracy. We also propose a novel explainability approach, which enables domain experts and practitioners to better comprehend the model's final decision. The visual explainability of infected brain regions and the model's robustness is evaluated by our two medical domain experts showing its promising use in real medical environment. To evaluate the model’s generalizability and robustness, our optimized model is tested on a dataset with different distribution. The experiments demonstrate that the proposed model, which was trained on ADNI data, exhibits reliable generalization to NACC data with an average precision of 90%, recall of 91%, F1-score of 89%, AUC of 88%, and accuracy of 88%.

Exploring the Impact of an IoT-based Game on the Experience of Visitors at a Natural Science Museum

The use of Internet of Things (IoT) technology to improve access to cultural heritage sites and enhance visitor engagement has become increasingly popular. Nevertheless, there is a lack of studies evaluating the actual benefits of such technology on visitors’ experiences, which is crucial to improve the IoT adoption in this field. To address this issue, we present an exploratory study we performed to understand whether and how an IoT-based game affects the experience of museum visitors. The game is played after a traditional guided tour; it was created following a meta-design approach to enable domain experts, such as museum professional guides, to adapt the game to specific visitors and/or situations. The study involved a total of 18 participants who took a guided tour of the Museum of Natural Sciences of the University of Bari and played the game afterward. Both quantitative and qualitative data were collected. The study results show that playing the IoT-based game positively affects the experience of visitors and, in particular, their emotions. Visitor workload was also found to be greater than that of a guided tour, but this had no negative impact on visitor experience. The study findings have implications for the design and development of future IoT games to enhance engagement and to improve the experience of visitors at cultural heritage sites.

Developing a data analytics toolbox for data-driven product planning: a review and survey methodology

Abstract The application of data analytics to product usage data has the potential to enhance engineering and decision-making in product planning. To achieve this effectively for cyber-physical systems (CPS), it is necessary to possess specialized expertise in technical products, innovation processes, and data analytics. An understanding of the process from domain knowledge to data analysis is of critical importance for the successful completion of projects, even for those without expertise in these areas. In this paper, we set out the foundation for a toolbox for data analytics, which will enable the creation of domain-specific pipelines for product planning. The toolbox includes a morphological box that covers the necessary pipeline components, based on a thorough analysis of literature and practitioner surveys. This comprehensive overview is unique. The toolbox based on it promises to support and enable domain experts and citizen data scientists, enhancing efficiency in product design, speeding up time to market, and shortening innovation cycles.

Simulation Discovery and Semi-Automatic Scenario Generation for Evaluation of Turbulence in Production Systems

Production systems have always faced changes and disruptions, which require dynamic decision-making to adjust existing plans to the unfolding reality. The interdependence of highly interconnected supply chain networks further adds to this volatility. Given this complexity, mainly caused by ambiguity and the systems’ dynamic, achieving transparency to make decisions in the context of production planning and control is challenging. Simulation models can help assess the outcome of different scenarios through experiments. However, building simulation models by hand requires extensive manual effort and expert knowledge of simulation tools. Although often partly automated, simulation experiments still require the exertion of simulation engineers to be conducted on a large scale. Moreover, the created models are often static and require additional resources to be updated in order to reflect changes in the physical system. To reduce this effort, the authors propose a concept combining the automatic discovery of simulation models from execution data with the semi-automatic generation of scenarios. This facilitates logistical risk analysis and prediction by evaluating the consequences of possible disruptive events. The concept aims to enable domain experts to use digital twins in large-scale virtual scenario evaluation, which is fundamental for increasing the agility of manufacturing systems by speeding up decision processes.

Enabling End-Users in Designing and Executing of Complex, Collaborative Robotic Processes

Over the last years, capabilities of robotic systems have quantitatively and qualitatively improved. But going beyond isolated robotic systems, the integration and interoperability of robotic capabilities in complex work processes remains a major challenge. This lack of tools to integrate robots needs to be addressed on technical, semantic and organizational level. In the ROBxTASK research project, we developed an approach to support cooperation between different types of users in order to enable domain experts, with no robotic know-how, to work with robot-assisted workflows. By engineering robotic skills at a useful and usable level of abstraction for experts in different domains, we aim to increase re-usability of these skills on two different levels, (robotic) device level, and on level of application specific workflows. The researched prototype consists of a web platform, which allows (a) engineers to register (robotic) devices and the implemented skills of the devices, (b) domain experts to use a graphical task design environment to create workflows across multiple robotic devices and lastly (c) robot co-workers to download and execute the workflow code in a local environment with digital twins or real robots. Additionally skills and workflows can be shared across organisations. Initial user studies have shown that the visual programming environment is accessible and the defined skill-set is easy to understand even for domain experts that are inexperienced in the field of robotics.

Leveraging Low Code Development of Smart Personal Assistants: An Integrated Design Approach with the SPADE Method

ABSTRACT Smart personal assistants (SPAs), such as Alexa for example, promise individualized user interactions owing to their varying interaction possibilities, knowledgeability, and human-like behaviors. To support the widespread adoption and use of SPAs, organizations such as Google or Amazon provide low code environments that support the development of SPAs (e.g., for Google Home or Amazon’s Alexa). These so-called low code platforms enable domain experts (e.g., business users without programming skills or experience) to develop SPAs for their purposes. However, using these platforms alone does not guarantee a useful and good conversation with novel SPAs due to non-intuitive design choices. Following a design science research approach, we propose the Smart Personal Assistant for Domain Experts (SPADE) method to address the missing link. This method supports domain experts in the development and contextualization of sophisticated SPAs for various application scenarios and focuses especially on conversational and anthropomorphic design steps. Our proof of concept and proof of value results show that SPADE is useful for supporting domain experts to create effective SPAs in different domains beyond private set-ups.

MaD GUI: An Open-Source Python Package for Annotation and Analysis of Time-Series Data.

Developing machine learning algorithms for time-series data often requires manual annotation of the data. To do so, graphical user interfaces (GUIs) are an important component. Existing Python packages for annotation and analysis of time-series data have been developed without addressing adaptability, usability, and user experience. Therefore, we developed a generic open-source Python package focusing on adaptability, usability, and user experience. The developed package, Machine Learning and Data Analytics (MaD) GUI, enables developers to rapidly create a GUI for their specific use case. Furthermore, MaD GUI enables domain experts without programming knowledge to annotate time-series data and apply algorithms to it. We conducted a small-scale study with participants from three international universities to test the adaptability of MaD GUI by developers and to test the user interface by clinicians as representatives of domain experts. MaD GUI saves up to 75% of time in contrast to using a state-of-the-art package. In line with this, subjective ratings regarding usability and user experience show that MaD GUI is preferred over a state-of-the-art package by developers and clinicians. MaD GUI reduces the effort of developers in creating GUIs for time-series analysis and offers similar usability and user experience for clinicians as a state-of-the-art package.

As Data Flows Surge Data Infrastructure Groans

As Data Flows Surge Data Infrastructure Groans

Bootstrapping MDE development from ROS manual code: Part 2\u2014Model generation and leveraging models at runtime

Model-driven engineering (MDE) addresses central aspects of robotics software development. MDE could enable domain experts to leverage the expressiveness of models, while implementation details on different hardware platforms would be handled by automatic code generation. Today, despite strong MDE efforts in the robotics research community, most evidence points to manual code development being the norm. A possible reason for MDE not being accepted by robot software developers could be the wide range of applications and target platforms, which make all-encompassing MDE IDEs hard to develop and maintain. Therefore, we chose to leverage a large corpus of open-source software widely adopted by the robotics community to extract common structures and gain insight on how and where MDE can support the developers to work more efficiently. We pursue modeling as a complement, rather than imposing MDE as separate solution. Our previous work introduced metamodels to describe components, their interactions, and their resulting composition. In this paper, we present two methods based on metamodels for automated generation of models from manually written artifacts: (1) through static code analysis and (2) by monitoring the execution of a running system. For both methods, we present tools that leverage the potentials of our contributions, with a special focus on their application at runtime to observe and diagnose a real system during its execution. A comprehensive example is provided as a walk-through for robotics software practitioners.

An information entropy and latent Dirichlet allocation approach to noise patent filtering

Defining valid patents in a particular technological field is an indispensable step in patent analysis. To minimise the risk of missing valid patents, domain experts manually exclude irrelevant patents, known as noise patents, from an initial patent set derived using a loose retrieval query. However, this task has become time-consuming and labour intensive due to the increasing number of patents and rising complexity of technological knowledge. This study proposes a semi-automated approach to noise patent filtering based on information entropy theory and latent Dirichlet allocation. The proposed approach comprises four discrete steps: (1) structuring patents using a term-weighting method; (2) recommending noise patent seeds based on the information quantity of patents in terms of focal keyword groups; (3) measuring text similarities for patent clustering using latent Dirichlet allocation; and (4) identifying potential noise patent clusters with respect to the noise patent seeds. Our case study confirms that the proposed approach is valuable as a complementary noise patent filtering tool that will enable domain experts to focus more on their own knowledge-intensive tasks such as prior art analysis and research and development (R&D) strategy formulation.

AntVis: A web-based visual analytics tool for exploring ant movement data

We present AntVis, a web-based visual analytics tool for exploring ant movement data collected from the video recording of ants moving on tree branches. Our goal is to enable domain experts to visually explore massive ant movement data and gain valuable insights via effective visualization, filtering, and comparison. This is achieved through a deep learning framework for automatic detection, segmentation, and labeling of ants, ant movement clustering based on their trace similarity, and the design and development of five coordinated views (the movement, similarity, timeline, statistical, and attribute views) for user interaction and exploration. We demonstrate the effectiveness of AntVis with several case studies developed in close collaboration with domain experts. Finally, we report the expert evaluation conducted by an entomologist and point out future directions of this study.

Graphical Flow-based Spark Programming

Increased sensing data in the context of the Internet of Things (IoT) necessitates data analytics. It is challenging to write applications for Big Data systems due to complex, highly parallel software frameworks and systems. The inherent complexity in programming Big Data applications is also due to the presence of a wide range of target frameworks, with different data abstractions and APIs. The paper aims to reduce this complexity and its ensued learning curve by enabling domain experts, that are not necessarily skilled Big Data programmers, to develop data analytics applications via domain-specific graphical tools. The approach follows the flow-based programming paradigm used in IoT mashup tools. The paper contributes to these aspects by (i) providing a thorough analysis and classification of the widely used Spark framework and selecting suitable data abstractions and APIs for use in a graphical flow-based programming paradigm and (ii) devising a novel, generic approach for programming Spark from graphical flows that comprises early-stage validation and code generation of Spark applications. Use cases for Spark have been prototyped and evaluated to demonstrate code-abstraction, automatic data abstraction interconversion and automatic generation of target Spark programs, which are the keys to lower the complexity and its ensued learning curve involved in the development of Big Data applications.

Querying Knowledge via Multi-Hop English Questions

AbstractThe inherent difficulty of knowledge specification and the lack of trained specialists are some of the key obstacles on the way to making intelligent systems based on the knowledge representation and reasoning (KRR) paradigm commonplace.Knowledge and query authoringusing natural language, especiallycontrollednatural language (CNL), is one of the promising approaches that could enable domain experts, who are not trained logicians, to both create formal knowledge and query it. In previous work, we introduced theKALMsystem (Knowledge Authoring Logic Machine) that supports knowledge authoring (and simple querying) with very high accuracy that at present is unachievable via machine learning approaches. The present paper expands on the question answering aspect of KALM and introducesKALM-QA(KALM for Question Answering) that is capable of answering much more complex English questions. We show that KALM-QA achieves 100% accuracy on an extensive suite of movie-related questions, calledMetaQA, which contains almost 29,000 test questions and over 260,000 training questions. We contrast this with a published machine learning approach, which falls far short of this high mark.

Harnessing Data Science Through Healthcare IT Interoperability

Harnessing Data Science Through Healthcare IT Interoperability

Rapid software prototyping for heterogeneous and distributed platforms

The software needs of scientists and engineers are growing and their programs are becoming more compute-heavy and problem-specific. This has led to an influx of non-expert programmers, who need to use and program high-performance computing platforms.With the continued stagnation of single-threaded performance, using hardware accelerators such as GPUs or FPGAs is necessary. Adapting software to these compute platforms is a difficult task, especially for non-expert programmers, leading to applications being unable to take advantage of new hardware or requiring extensive rewrites.We propose a programming model that allows non-experts to benefit from high-performance computing, while enabling expert programmers to take full advantage of the underlying hardware. In this model, programs are generically typed, the location of the data is encoded in the type system, and multiple dispatch is used to select functionality based on the type of the data. This enables rapid prototyping, retargeting and reuse of existing software, while allowing for hardware specific optimization if required.Our approach allows development to happen in one source language enabling domain experts and performance engineers to jointly develop a program, without the overhead, friction, and challenges associated with developing in multiple programming languages for the same project.We demonstrate the viability and the core principles of this programming model in Julia using realistic examples, showing the potential of this approach for rapid prototyping, and its applicability for real-life engineering. We focus on usability for non-expert programmers and demonstrate that the potential of the underlying hardware can be fully exploited.

An openEHR Approach to Detailed Clinical Model Development: Tobacco Smoking Summary Archetype as a Case Study.

Data modeling for electronic health records (EHRs) is complex, requiring technological and cognitive sophistication. The openEHR approach leverages the tacit knowledge of domain experts made explicit in a model development process aiming at interoperability and data reuse. The purpose of our research was to explore the process that enabled the aggregation of the tacit knowledge of domain experts in an explicit form using the Clinical Knowledge Manager (CKM) platform and associated assets. The Tobacco Smoking Summary archetype is used to illustrate this. Three methods were used to triangulate findings: (1) observation of CKM discussions by crowdsourced domain experts in two reviews, (2) observation of editor discussions and decision-making, and (3) interviews with eight domain experts. CKM discussions were analyzed for content and editor discussions for decision-making, and interviews were thematically analyzed to explore in depth the explication of tacit knowledge. The Detailed Clinical Model (DCM) process consists of a set of reviews by domain experts, with each review followed by editorial discussions and decision-making until an agreement is reached among reviewers and editors that the DCM is publishable. Interviews revealed three themes: (1) data interoperability and reusability, (2) accurate capture of patient data, and (3) challenges of sharing tacit knowledge. The openEHR approach to developing an open standard revealed a complex set of conditions for a successful interoperable archetype, such as leadership, maximal dataset, crowdsourced domain expertise and tacit knowledge made explicit, editorial vision, and model-driven software. Aggregated tacit knowledge that is explicated into a DCM enables the ability to collect accurate data and plan for the future. The process based on the CKM platform enables domain experts and stakeholders to be heard and to contribute to mutually designed standards that align local protocols and agendas to international interoperability requirements.

Ensemble Visualization of Bottlenecks in Planar Flow Networks

The problem of finding bottlenecks in flow networks often appears in real world applicationslike production planning, factory layout, flow related physical approaches and even cyber security. This work introduces intuitive visual mechanisms to enable domain experts and users to visually analyzestable regions of a network and identify critical transitions. Those transitions form a varyingbottleneck front for different configurations of network restraints. To tackle this problem, this workenhances the comparability of different network configurations by utilizing ensemble visualizationtechniques. The effectiveness of this approach is demonstrated by showing how this enables users toevaluate the progress of different bottlenecks and individual regions in a flow network.

Towards High-Quality Visualization of Superfluid Vortices

Superfluidity is a special state of matter exhibiting macroscopic quantum phenomena and acting like a fluid with zero viscosity. In such a state, superfluid vortices exist as phase singularities of the model equation with unique distributions. This paper presents novel techniques to aid the visual understanding of superfluid vortices based on the state-of-the-art non-linear Klein-Gordon equation, which evolves a complex scalar field, giving rise to special vortex lattice/ring structures with dynamic vortex formation, reconnection, and Kelvin waves, etc. By formulating a numerical model with theoretical physicists in superfluid research, we obtain high-quality superfluid flow data sets without noise-like waves, suitable for vortex visualization. By further exploring superfluid vortex properties, we develop a new vortex identification and visualization method: a novel mechanism with velocity circulation to overcome phase singularity and an orthogonal-plane strategy to avoid ambiguity. Hence, our visualizations can help reveal various superfluid vortex structures and enable domain experts for related visual analysis, such as the steady vortex lattice/ring structures, dynamic vortex string interactions with reconnections and energy radiations, where the famous Kelvin waves and decaying vortex tangle were clearly observed. These visualizations have assisted physicists to verify the superfluid model, and further explore its dynamic behavior more intuitively.

DeepDive

The dark data extraction or knowledge base construction (KBC) problem is to populate a SQL database with information from unstructured data sources including emails, webpages, and pdf reports. KBC is a long-standing problem in industry and research that encompasses problems of data extraction, cleaning, and integration. We describe DeepDive, a system that combines database and machine learning ideas to help develop KBC systems. The key idea in DeepDive is that statistical inference and machine learning are key tools to attack classical data problems in extraction, cleaning, and integration in a unified and more effective manner. DeepDive programs are declarative in that one cannot write probabilistic inference algorithms; instead, one interacts by defining features or rules about the domain. A key reason for this design choice is to enable domain experts to build their own KBC systems. We present the applications, abstractions, and techniques of DeepDive employed to accelerate construction of KBC systems.

Bridging Theory with Practice: An Exploratory Study of Visualization Use and Design for Climate Model Comparison.

Evaluation methodologies in visualization have mostly focused on how well the tools and techniques cater to the analytical needs of the user. While this is important in determining the effectiveness of the tools and advancing the state-of-the-art in visualization research, a key area that has mostly been overlooked is how well established visualization theories and principles are instantiated in practice. This is especially relevant when domain experts, and not visualization researchers, design visualizations for analysis of their data or for broader dissemination of scientific knowledge. There is very little research on exploring the synergistic capabilities of cross-domain collaboration between domain experts and visualization researchers. To fill this gap, in this paper we describe the results of an exploratory study of climate data visualizations conducted in tight collaboration with a pool of climate scientists. The study analyzes a large set of static climate data visualizations for identifying their shortcomings in terms of visualization design. The outcome of the study is a classification scheme that categorizes the design problems in the form of a descriptive taxonomy. The taxonomy is a first attempt for systematically categorizing the types, causes, and consequences of design problems in visualizations created by domain experts. We demonstrate the use of the taxonomy for a number of purposes, such as, improving the existing climate data visualizations, reflecting on the impact of the problems for enabling domain experts in designing better visualizations, and also learning about the gaps and opportunities for future visualization research. We demonstrate the applicability of our taxonomy through a number of examples and discuss the lessons learnt and implications of our findings.