Domain Knowledge Research Articles

Unravelling uncertainty in trajectory prediction using a non-parametric approach

Predicting the trajectories of road agents is fundamental for self-driving cars. Trajectory prediction contains many sources of uncertainty in data and modelling. A thorough understanding of this uncertainty is crucial in a safety-critical task like auto-piloting a vehicle. In practice, it is necessary to distinguish between the uncertainty caused by partial observability of all factors that may affect a driver’s near-future decisions, the so-called aleatoric uncertainty, and the uncertainty of deploying a model in new scenarios that are possibly not present in the training set, the so-called epistemic uncertainty. They reflect the trade-off between data collection and model improvement In this paper, we propose a new framework to systematically quantify both sources of uncertainty. Specifically, to approximate the spatial distribution of an agent’s future position, we propose a 2D histogram-based deep learning model combined with deep ensemble techniques for measuring aleatoric and epistemic uncertainty by entropy-based quantities. The proposed Uncertainty Quantification Network (UQnet) employs a causal part to enhance its generalizability so rare driving behaviours can be effectively identified. Experiments on the INTERACTION dataset show that UQnet is able to give more robust predictions in generalizability tests compared to the correlation-based models. Further analysis presents that high aleatoric uncertainty cases are mainly caused by heterogeneous driving behaviours and unknown intended directions. Based on this aleatoric uncertainty component, we estimate the lower bounds of mean-square-error and final-displacement-error as indicators for the predictability of trajectories. Furthermore, the analysis of epistemic uncertainty illustrates that domain knowledge of speed-dependent driving behaviour is essential for adapting a model from low-speed to high-speed situations. Our paper contributes to motion forecasting with a new framework, that recasts the problem of accuracy improvement in a way that focuses on differentiating between unpredictable components and rare cases for which more and different data should be collected.

A review of explainable fashion compatibility modeling methods

The paper reviews methods used in the fashion compatibility recommendation domain. We select methods based on reproducibility, explainability, and novelty aspects and then organize them chronologically and thematically. We presented general characteristics of publicly available datasets that are related to the fashion compatibility recommendation task. Finally, we analyzed the representation bias of datasets, fashion-based algorithms’ sustainability, and explainable model assessment. The paper describes practical problem explanations, methodologies, and published datasets that may serve as an inspiration for further research. The proposed structure of the survey organizes knowledge in the fashion recommendation domain and will be beneficial for those who want to learn the topic from scratch, expand their knowledge, or find a new field for research. Furthermore, the information included in this paper could contribute to developing an effective and ethical fashion-based recommendation system.

As Teachers Not Tourists: Creating Knowledge During a Short-Term Study Abroad Program

Here, I address the call for teacher education programs to feature international elements to help pre-service teachers craft powerful classroom instruction and successfully work with students from culturally diverse backgrounds. I share a qualitative investigation of participants’ experiences during a short-term study abroad program in Japan that explored contemporary disaster preparedness as a result of the Great East Japan Earthquake and Tsunami. The program sponsors, the Japanese Ministry of Foreign Affairs, hoped participants would develop an appreciation for the people and culture of Japan while they learned about disaster preparedness. I wondered if collecting curriculum items during a short-term study abroad program could help social studies pre-service teachers create—not just receive—historical domain knowledge? In Japan, participants collected items to use as featured resources around which they designed classroom activities for secondary students. Their activities demonstrated three levels of second-order historical domain knowledge and two elements of civic competence. This study suggests that pre-service teachers with no previous international experience may require explicit modeling and scaffolding to translate their short-term study abroad participation into dynamic learning. Another implication is that participants need strong support to consider the photographs they capture as artifacts of contemporary material culture.

Utilizing Generative AI for Instructional Design: Exploring Strengths, Weaknesses, Opportunities, and Threats

AbstractRecently, generative AI has been at the center of disruptive innovation in various settings, including educational sectors. This article investigates ChatGPT, which is one of the most prominent generative AI in the market, to explore its usefulness and potential for instructional design. Four researchers used a set of prompts to generate a course map for an online course that is aimed to teach the topic of makerspace and conducted SWOT analysis to identify strengths, weaknesses, opportunities, and threats of using generative AI for instructional design. The findings suggest that there is promise in using ChatGPT as an efficient and effective tool for creating course maps, yet it still requires the domain knowledge and instructional design expertise to warrant quality and reliability of the tool.

Intelligent adaptive lighting algorithm: Integrating reinforcement learning and fuzzy logic for personalized interior lighting

The lighting requirements are subjective and one light setting cannot work for all. However, there is little work on developing smart lighting algorithms that can adapt to user preferences. To address this gap, this paper uses fuzzy logic and reinforcement learning to develop an adaptive lighting algorithm. In particular, we develop a baseline fuzzy inference system (FIS) using the domain knowledge, generating light recommendation based on a set of intuitive rules. These rules, derived from existing literature, are based on environmental conditions i.e. daily glare index, and user information including age, activity, and chronotype. Through a feedback mechanism, the user interacts with the algorithm, correcting the algorithm output to their preferences. We interpret these corrections as rewards to a Q-learning algorithm, which tunes the FIS parameters online to match the user preferences. The Q-learning is a model-free learning algorithm that learns to act optimally by interacting with the user and the rewards it receives. This allows the proposed algorithm to work in a model-free manner, effectively handling the uncertainties that might arise from the individualistic preferences of users. To the authors’ best knowledge, this algorithm is pioneering work in designing intelligent algorithms for personalized lighting control, featuring several elements of novelty, including the number of environmental and user-related inputs, the continuous control of light intensity as opposed to common on/off control, and the ability to learn user preferences. The algorithm is implemented in a real aircraft cabin and is evaluated in an extensive user study. The implementation results demonstrate that the developed algorithm possesses the capability to learn user preferences while successfully adapting to a wide range of environmental conditions and user characteristics. This underscores its viability as a potent solution for intelligent light management, featuring advanced learning capabilities.

Enhancing youth involvement in community development: A pragmatic strategy for local churche

This article reports on the findings of the Christian youth ministry involvement in community development in the Mayibuye community of Tembisa, in the Gauteng province of South Africa. The article employed Richard Osmer’s model of the four tasks of practical theological interpretation. These tasks have been used to understand better what is happening in youth ministry and community development contexts in the Mayibuye community by utilising documentary analysis and in-depth semi-structured interviews. The findings divulged that the Mayibuye community faces several socio-economic challenges. Most importantly, young people are the most affected. Furthermore, the findings exposed that youth involvement is limited in community development processes within local communities. As a result, the article proposes implementing a pragmatic strategy to enhance the youth ministry’s involvement in community development. Various principles of the youth ministry have been presented in the study as the essential principles that need youth leaders’ attention for functional youth ministry, and to achieve developmental outcomes. The study also recommended further empirical studies on youth culture, the importance of young people’s inclusion in leadership roles and the use of young people as catalysts for community transformation.Intradisciplinary and/or interdisciplinary implications: The anticipated outcomes of this research endeavour are poised to make a valuable contribution to the current corpus of knowledge in various academic domains, including Practical Theology, Youth Ministry, Theology and Development, Community Development, Congregational Studies, Missiology, Anthropology, and Sociology. This contribution suggests that adopting a pragmatic strategy is likely to yield positive outcomes for local churches, youth ministry, and community development as a whole.

Current and future applications of artificial intelligence in surgery: implications for clinical practice and research

Surgeons are skilled at making complex decisions over invasive procedures that can save lives and alleviate pain and avoid complications in patients. The knowledge to make these decisions is accumulated over years of schooling and practice. Their experience is in turn shared with others, also via peer-reviewed articles, which get published in larger and larger amounts every year. In this work, we review the literature related to the use of Artificial Intelligence (AI) in surgery. We focus on what is currently available and what is likely to come in the near future in both clinical care and research. We show that AI has the potential to be a key tool to elevate the effectiveness of training and decision-making in surgery and the discovery of relevant and valid scientific knowledge in the surgical domain. We also address concerns about AI technology, including the inability for users to interpret algorithms as well as incorrect predictions. A better understanding of AI will allow surgeons to use new tools wisely for the benefit of their patients.

HME-KG: A method of constructing the human motion encoding knowledge graph based on a hierarchical motion model

The diversity, infinity, and nonuniform description of human motion make it challenging for computers to understand human activities. To explore and reuse captured human motion data, this work defines a more comprehensive hierarchical theoretical model of human motion and proposes a standard human posture encoding scheme. We construct a domain knowledge graph (DKG) named the human motion encoding knowledge graph (HME-KG) based on posture codes and action labels. Community detection, similarity analysis, and centrality analysis are used to explore the potential value of motion data. This paper conducts an evaluation and visualization of HME-KG.

Bayesian networks for Risk Assessment and postoperative deficit prediction in intraoperative neurophysiology for brain surgery

Bayesian networks for Risk Assessment and postoperative deficit prediction in intraoperative neurophysiology for brain surgery

Advanced Bayesian study on inland navigational risk of remotely controlled autonomous ship

The arise of autonomous ships has necessitated the development of new risk assessment techniques and methods. This study proposes a new framework for navigational risk assessment of remotely controlled Maritime Autonomous Surface Ships (MASS). This framework establishes a set of risk influencing factors affecting safety of navigation of a remotely-controlled MASS. Next, model parameters are defined based on the risk factors, and the model structure is developed using Bayesian Networks. To this end, an extensive literature survey is conducted, enhanced with the domain knowledge elicited from the experts and improved by the experimental data obtained during representative MASS model trials carried out in an inland river. Conditional Probability Tables are generated using a new function employing expert feedback regarding Interval Type 2 Fuzzy Sets. The developed Bayesian model yields the expected utilities results representing an accident’s probability and consequence, with the results visualized on a dedicated diagram. Finally, the developed risk assessment model is validated by conducting three axiom tests, extreme scenarios analysis, and sensitivity analysis. Navigational environment, natural environment, traffic complexity, and shore-ship collaboration performance are critical from the probability and consequence perspective for inland navigational accidents to a remotely controlled MASS. Lastly, important nodes to Shore-Ship collaboration performance include autonomy of target ships, cyber risk, and transition from other remote control centers.

Recommendations for the Model-Based Systems Engineering Modeling Process Based on the SysML Model and Domain Knowledge

Model-based systems engineering (MBSE) is a modeling approach used in industry to support the formalization, analysis, design, checking and verification of systems. In MBSE modeling, domain knowledge is the basis of the modeling. However, modeling does not happen overnight; it requires systematic training and a significant investment of resources. Unfortunately, many domain experts lack the expertise required for modeling, even though they know the domain well. The question arises about how to provide system modelers with domain knowledge at the right time to support the efficient completion of modeling. Since MBSE research that integrates AI is just beginning to take off, no public dataset is available. In this paper, aerospace SysML models are constructed based on spacecraft-related domain knowledge to form SysML model data. The validation rules are studied to validate the SysML model data, and combined with the concept of the recommended system, a recommendation method for the MBSE modeling process based on the knowledge and SysML model is proposed. A GLOVE language model is pre-trained by using domain knowledge and general knowledge; the model data are also used to fine-tune the GLOVE language model combined with the pre-training to recommend some domain development processes. The recommendation list is manually quality-verified and fed into the pre-training phase, while new requirement texts are continuously added in the fine-tuning phase, resulting in a more relevant and accurate recommendation list. Experiments show that the incremental recommender system can not only effectively recommend SysML models, but also improve the quality and efficiency of MBSE development.

Knowledge graph-guided data-driven design of ultra-high-performance concrete (UHPC) with interpretability and physicochemical reaction discovery capability

Traditional methods for designing concrete materials typically rely on labor-intensive laboratory experiments, resulting in time and cost inefficiencies. Recently, designing concrete using artificial intelligence (AI) methods has shown high efficiency, but existing AI methods often rely solely on data, which can lead to violation with scientific principles and result in models lacking reasoning abilities. To overcome these challenges, this paper presents an interpretable knowledge graph-guided data-driven design approach. By integrating advanced computing techniques with domain knowledge via knowledge graphs, this approach enables the interpretation of data-driven models and uncovers the underlying mechanisms behind predictions. This approach is applied to ultra-high-performance concrete (UHPC) involving complex physicochemical reactions. The domain knowledge about UHPC is imparted using a knowledge graph, and UHPC properties are predicted using a machine learning model considering mixing proportions, processing methods, and physiochemical properties of materials via natural language processing. The results show that the knowledge graph displays crucial design variables and their effects on UHPC properties, aiding in selecting variables for machine learning models and interpreting their results. The prediction accuracy of the machine learning model reached 0.95. The research paves the way for more transparent and scientific AI models for material design and AI-enabled discovery of scientific knowledge.

In-situ particle analysis with heterogeneous background: a machine learning approach

We propose a novel framework that combines state-of-the-art deep learning approaches with pre- and post-processing algorithms for particle detection in complex/heterogeneous backgrounds common in the manufacturing domain. Traditional methods, like size analyzers and those based on dilution, image processing, or deep learning, typically excel with homogeneous backgrounds. Yet, they often fall short in accurately detecting particles against the intricate and varied backgrounds characteristic of heterogeneous particle–substrate (HPS) interfaces in manufacturing. To address this, we've developed a flexible framework designed to detect particles in diverse environments and input types. Our modular framework hinges on model selection and AI-guided particle detection as its core, with preprocessing and postprocessing as integral components, creating a four-step process. This system is versatile, allowing for various preprocessing, AI model selections, and post-processing strategies. We demonstrate this with an entrainment-based particle delivery method, transferring various particles onto substrates that mimic the HPS interface. By altering particle and substrate properties (e.g., material type, size, roughness, shape) and process parameters (e.g., capillary number) during particle entrainment, we capture images under different ambient lighting conditions, introducing a range of HPS background complexities. In the preprocessing phase, we apply image enhancement and sharpening techniques to improve detection accuracy. Specifically, image enhancement adjusts the dynamic range and histogram, while sharpening increases contrast by combining the high pass filter output with the base image. We introduce an image classifier model (based on the type of heterogeneity), employing Transfer Learning with MobileNet as a Model Selector, to identify the most appropriate AI model (i.e., YOLO model) for analyzing each specific image, thereby enhancing detection accuracy across particle–substrate variations. Following image classification based on heterogeneity, the relevant YOLO model is employed for particle identification, with a distinct YOLO model generated for each heterogeneity type, improving overall classification performance. In the post-processing phase, domain knowledge is used to minimize false positives. Our analysis indicates that the AI-guided framework maintains consistent precision and recall across various HPS conditions, with the harmonic mean of these metrics comparable to those of individual AI model outcomes. This tool shows potential for advancing in-situ process monitoring across multiple manufacturing operations, including high-density powder-based 3D printing, powder metallurgy, extreme environment coatings, particle categorization, and semiconductor manufacturing.

Range-Spread Target Detection Networks Using HRRPs

Range-spread target (RST) detection is an important issue for high-resolution radar (HRR). Traditional detectors relying on manually designed detection statistics have their performance limitations. Therefore, in this work, two deep learning-based detectors are proposed for RST detection using HRRPs, i.e., an NLS detector and DFCW detector. The NLS detector leverages domain knowledge from the traditional detector, treating the input HRRP as a low-level feature vector for target detection. An interpretable NLS module is designed to perform noise reduction for the input HRRP. The DFCW detector takes advantage of the extracted high-level feature map of the input HRRP to improve detection performance. It incorporates a feature cross-weighting module for element-wise feature weighting within the feature map, considering the channel and spatial information jointly. Additionally, a nonlinear accumulation module is proposed to replace the conventional noncoherent accumulation operation in the double-HRRP detection scenario. Considering the influence of the target spread characteristic on detector performance, signal sparseness is introduced as a measure and used to assist in generating two datasets, i.e., a simulated dataset and measured dataset incorporating real target echoes. Experiments based on the two datasets are conducted to confirm the contribution of the designed modules to detector performance. The effectiveness of the two proposed detectors is verified through performance comparison with traditional and deep learning-based detectors.

ChatGPT for Tinnitus Information and Support: Response Accuracy and Retest after Three and Six Months

Testing of ChatGPT has recently been performed over a diverse range of topics. However, most of these assessments have been based on broad domains of knowledge. Here, we test ChatGPT’s knowledge of tinnitus, an important but specialized aspect of audiology and otolaryngology. Testing involved evaluating ChatGPT’s answers to a defined set of 10 questions on tinnitus. Furthermore, given the technology is advancing quickly, we re-evaluated the responses to the same 10 questions 3 and 6 months later. The accuracy of the responses was rated by 6 experts (the authors) using a Likert scale ranging from 1 to 5. Most of ChatGPT’s responses were rated as satisfactory or better. However, we did detect a few instances where the responses were not accurate and might be considered somewhat misleading. Over the first 3 months, the ratings generally improved, but there was no more significant improvement at 6 months. In our judgment, ChatGPT provided unexpectedly good responses, given that the questions were quite specific. Although no potentially harmful errors were identified, some mistakes could be seen as somewhat misleading. ChatGPT shows great potential if further developed by experts in specific areas, but for now, it is not yet ready for serious application.

Med-ImageTools: An open-source Python package for robust data processing pipelines and curating medical imaging data

Background Machine learning and AI promise to revolutionize the way we leverage medical imaging data for improving care but require large datasets to train computational models that can be implemented in clinical practice. However, processing large and complex medical imaging datasets remains an open challenge. Methods To address this issue, we developed Med-ImageTools, a new Python open-source software package to automate data curation and processing while allowing researchers to share their data processing configurations more easily, lowering the barrier for other researchers to reproduce published works. Use cases We have demonstrated the efficiency of Med-ImageTools across three different datasets, resulting in significantly reduced processing times. Conclusions The AutoPipeline feature will improve the accessibility of raw clinical datasets on public archives, such as the Cancer Imaging Archive (TCIA), the largest public repository of cancer imaging, allowing machine learning researchers to process analysis-ready formats without requiring deep domain knowledge.

Multi-head CNN-based activity recognition and its application on chest-mounted sensor-belt

In recent years, a great deal of research has been done on the identification, monitoring, and classification of human activities. Human activity recognition (HAR) is a term commonly used to describe the automatic identification of physical activities. For activity recognition, there are primarily vision-based and sensor-based methods available. The computer vision-based method is generally effective in lab settings, but because of clutter, fluctuating light levels, and contrast, it may not perform well in real-world scenarios. Continuous monitoring and analysis of physiological signals obtained from heterogeneous sensors attached to an individual’s body is required to realise sensor-based HAR systems. Most of the previous research in human activity recognition (HAR) is biased along with feature engineering and pre-processing which requires a good amount of domain knowledge. Application-specific modelling and time-taking methods are involved in these approaches. In this work, the multi-head convolutional neural network-based human activity recognition framework is proposed where automatic feature extraction and classification are involved in the form of an end-to-end classification approach. Experiments of this approach are performed by taking raw wearable sensor data with few pre-processing steps and without the involvement of a handcrafted feature extraction technique. 99.23% and 93.55% accuracy are obtained on the WISDM and UCI-HAR datasets which denoted the much improvement in the assessment of HAR over other similar approaches. The model is also tested on locally collected data from a chest mounted belt with fabric sensors and an accuracy of 87.14% has been achieved on that data.

Deriving Input Variables through Applied Machine Learning for Short-Term Electric Load Forecasting in Eskilstuna, Sweden

As the demand for electricity, electrification, and renewable energy rises, accurate forecasting and flexible energy management become imperative. Distribution network operators face capacity limits set by regional grids, risking economic penalties if exceeded. This study examined data-driven approaches of load forecasting to address these challenges on a city scale through a use case study of Eskilstuna, Sweden. Multiple Linear Regression was used to model electric load data, identifying key calendar and meteorological variables through a rolling origin validation process, using three years of historical data. Despite its low cost, Multiple Linear Regression outperforms the more expensive non-linear Light Gradient Boosting Machine, and both outperform the “weekly Naïve” benchmark with a relative Root Mean Square Errors of 32–34% and 39–40%, respectively. Best-practice hyperparameter settings were derived, and they emphasize frequent re-training, maximizing the training data size, and setting a lag size larger than or equal to the forecast horizon for improved accuracy. Combining both models into an ensemble could the enhance accuracy. This paper demonstrates that robust load forecasts can be achieved by leveraging domain knowledge and statistical analysis, utilizing readily available machine learning libraries. The methodology for achieving this is presented within the paper. These models have the potential for economic optimization and load-shifting strategies, offering valuable insights into sustainable energy management.

A systematic review of multidimensional relevance estimation in information retrieval

AbstractIn information retrieval, relevance is perceived as a multidimensional and dynamic concept influenced by user, task, and domain factors. Relying on this perspective, researchers have introduced multidimensional relevance models addressing diverse search tasks across numerous knowledge domains. Through our systematic review of 72 studies, we categorize research based on domain specificity and the distinct relevance aspects employed for estimating multidimensional relevance. Moreover, we highlight the approaches used to aggregate scores related to these factors and rank information items. Our insights underline the importance of concise definitions and unified methods for estimating relevance factors within and across domains. Finally, we identify benchmark collections for evaluations based on multiple relevance aspects while underscoring the necessity for new ones. Our findings suggest that large language models hold considerable promise for shaping future research in this field, mainly due to their relevance labeling abilities.This article is categorized under: Application Areas > Science and Technology Technologies > Computational Intelligence

Objective numeracy exacerbates framing effects from decision-making under political risk

While Prospect Theory helps to explain decision-making under risk, studies often base frames on hypothetical events and fail to acknowledge that many individuals lack the ability and motivation to engage in complex thinking. We use an original survey of US adults (N = 2813) to test Prospect Theory in the context of the May 2023 debt ceiling negotiations in the US Congress and assess whether objective numeracy moderates framing effects. We hypothesize and find evidence to suggest that most respondents are risk-averse to potential gains and risk-accepting to potential losses; however, high numerates are more risk-averse and risk-accepting to gains and losses, respectively, than low numerates. We also find that need for cognition interacts with numeracy to moderate framing effects for prospective losses, such that higher need for cognition attenuates risk-acceptance among low numerates and exacerbates risk-acceptance among high numerates. Our results are robust to a range of other covariates and in models accounting for the interaction between political knowledge and need for cognition, indicating joint moderating effects from two knowledge domains similarly conditioned by the desire to engage in effortful thinking. Our findings demonstrate that those who can understand and use objective information may remain subjectively persuaded by certain policy frames.