Relevant Prior Knowledge Research Articles

Intelligent risk identification for drilling lost circulation incidents using data-driven machine learning

In drilling operations, lost circulation (LC) poses a significant challenge, potentially leading to severe accidents such as overflow or blowouts if not accurately identified and controlled. This study introduces a method for identifying LC risk severity using data-driven machine learning and statistical algorithms, specifically demonstrated in the Ying-Qiong Basin of the South China Sea. The method begins with variance filtering to eliminate invalid data, followed by the application of the Pearson correlation coefficient to remove redundant data. This process results in the selection of 12 logging parameters out of the initial 38. Combined with indirect judgments of LC severity based on relevant prior knowledge, the data is labeled to construct the original dataset. A feature extraction and optimization method is proposed, which employs sliding time windows to extract 21 features. These features are then selected and optimized using a random forest algorithm to construct the optimal feature set. To enhance the controllability of the Extreme Learning Machine (ELM) model, the sparrow search algorithm is integrated. Additionally, several fitness functions are introduced for both regression and classification models, resulting in the development of the improved ELM (IELM) model. The dataset is split into training and test sets, and the performance of the IELM model is compared to that of the standard ELM and BP neural network models. The results indicate that the IELM model outperforms the other two, achieving an impressive test F1 score of 97.22 % for accurately identifying LC severity. Moreover, the model trained using the optimal feature set demonstrates higher accuracy and faster convergence. The proposed feature extraction and optimization method effectively increases the amount of information contained in the data, thereby enhancing the model's performance.

Fault Classification in Reciprocating Compressors: A Comparison of Machine Learning and Deep Learning Approaches

This study compares methodologies for fault classification in reciprocating compressors, focusing on traditional Machine Learning (ML) with classical feature extraction processes and one-dimensional Convolutional Neural Networks (1D-CNN) in Deep Learning (DL). Both techniques demonstrated viability by employing a dataset of compressor vibration signals encompassing ten fault classes. While ML achieved a classification accuracy of 86%, DL reached 90.709%, highlighting its superior learning and generalization abilities, although with longer training times. These findings suggest that, despite ML being effective when relevant prior knowledge is available, DL, particularly with 1D-CNN, offers enhanced fault classification performance for this study case at the expense of additional processing resources.

Prior knowledge guided text to image generation

Generating a realistic and semantically consistent image from a given text is a challenging task. Due to the limited information of natural language, it is difficult to generate vivid images with fine details. To address this problem, we propose a Prior Knowledge Guided GAN for text to image generation. Specifically, the proposed method consists of several Knowledge Guided Up-Blocks. We decompose the image into a superposition of several visual regions, each of which requires corresponding prior knowledge to enrich its visual details. Correspondingly, we construct each Up-Block by incorporating relevant prior knowledge as input, aiming to enhance the quality of each visual region. Prior knowledge progressively provides more visual detail through affine transformations. Finally, high-quality images are synthesized by fusing all image regions. Experimental results on the CUB and COCO datasets demonstrate the superior performance of the proposed method.

Infusing 21st-century competencies into scripted foundation phase literacy lessons

Background: Despite large-scale interventions aimed at developing literacy skills, children’s reading competence levels in South Africa continue to remain an area of concern. In addition, the need to prepare learners for the increased demands of a fast-changing world of learning and working is gaining attention in educational policy and practice.Aim: Using a qualitative multi-site case study research design, the authors aimed to explore teachers’ understanding and enactment of scripted literacy lessons that are designed to promote 21st-century competencies.Setting: Five Grade 1 teachers were selected from four schools, three of which are in peri-urban and the other in a township area.Methods: Data were generated in two phases across three teaching cycles. In the first phase, lessons were observed and recorded on video. The second phase consisted of stimulated recall interviews (SRIs) in which teachers commented on their recorded lessons.Results: The findings showed that when teachers had relevant prior knowledge of specific competencies, they were better positioned to leverage these as a basis for their lessons. However, the findings also indicated that teachers’ entrenched methods and ways of thinking were obstacles for change.Conclusion: Training and development opportunities should pay sufficient attention to the practicalities of changing pedagogies and using scripted lesson plans.Contribution: The findings contribute to knowledge on the affordances and challenges when designing training opportunities aimed at integrating 21st-century competencies into scripted literacy lessons.

Combatting rumors around the French election: the memorability and effectiveness of fact-checking articles

Across four studies, we examined the how design decisions influenced the effectiveness of fact-checking articles created by CrossCheck France during the 2017 French election. We measured both memory for the article and belief in the false rumor. We saw no difference in fact check efficacy based on the type of headline (question vs negation) or the number of newsroom logos present around the article (one, four, or seven). In addition, informative design features such as an icon identifying the type of misinformation were ignored by readers. Participants failed to remember many of the details from the article, but retrieval practice was beneficial in reducing forgetting over a 1-week delay. In both US and French samples, reading the fact check decreased belief in the false information, even 1 week later. However, the articles were much more effective in the US sample, who lacked relevant prior knowledge and political beliefs. Overall, fact-checking articles can be effective at reducing belief in false information, but readers tend to forget the details and ignore peripheral information.

Computer-based feedback matters when relevant prior knowledge is not activated

BackgroundHow feedback is given may influence its utility. AimWe examined the effect of activated prior knowledge on learning from feedback by manipulating whether knowledge of a foundational concept was activated before solving fraction division problems. Sample and methodsUndergraduates (N = 171) were randomly assigned in a 3 (feedback timing: delayed, immediate, or no feedback) x 2 (knowledge activation: relevant or not) between-subjects design. ResultsIf irrelevant knowledge was activated, immediate feedback enhanced learning as compared to no feedback during the learning task, whereas if relevant knowledge was activated, then there was no impact of immediate feedback. On the posttest, any feedback (immediate or delayed) resulted in greater performance, but feedback timing did not matter. Thus, activating prior knowledge moderates the effect of feedback on learning. ConclusionWhen researchers or practitioners are investigating or giving feedback, they must also consider individual differences of the learner such as the prior knowledge they bring to the task.

Perception and willingness of private secondary school students in forestry education and career: A case study from Nigeria

AbstractOne of the critical issues confronting forest sustainability is inadequate education about the need and actions to sustain these vital ecosystems. Notwithstanding, forestry education attracts low levels of interest and enrollment in Nigeria and globally. Unlike in public secondary schools, information on the perception and willingness toward forestry education and career among private school students is limited. Therefore, our study assessed the (un)willingness and perception toward forestry education and career through a survey of 160 students across eight private secondary schools in Akure metropolis, Nigeria. Descriptive and chi‐square statistics were used to analyze the responses. We found a relatively low interest and willingness in forestry education among the students (20%), despite their positive perceptions toward the profession's lucrativeness (73%), their relevant prior knowledge learnt through school's agriculture subject (88%), and their previous tree planting engagement (55%). None of the students’ demographics had significantly influenced their professional interest and choice besides having family relation(s) in forestry (χ2 = 5.731, p = .017). Likewise, only the availability of scholarships, accessible post‐study forestry jobs, and adequate integration of forestry into secondary school education could have significantly attracted the students to pursue forestry education or a career. Therefore, improving these key areas is recommended while also creating public awareness to facilitate more positive perceptions about the profession.

Promoting the construction of intelligent knowledge with the help of various methods of cognitively activating instruction

How can we promote the construction of intelligent knowledge that facilitates the transfer of knowledge to new situations? This article gives an overview of the current state of empirical research on methods of cognitively activating teaching that promote conceptual change and flexible application of knowledge in new situations. “Cognitively activating” means that these methods promote the active construction of knowledge by facilitating a better understanding of the problem at hand, by activating relevant prior knowledge, and by fostering a better awareness of the limits of one’s own knowledge. Six methods that have each proven successful in boosting learners’ conceptual understanding in experimental studies are presented in this article.

A hybrid‐driven continuous‐time filter for manoeuvering target tracking

This paper considers the problem of target tracking in complex manoeuvering scenarios with a lack of relevant prior knowledge. This is a challenge for classical model-based manoeuvering target tracking algorithms because they rely heavily on accurate domain and prior knowledge of target motion. To address this problem, we propose a hybrid-driven continuous-time filter algorithm in this paper, which combines the advantages of the model-driven and data-driven. We use the stochastic differential equation (SDE) with the acceleration model as the basic framework of the proposed algorithm. In order to deal with unpredictable manoeuvres and unknown perturbations, we adopt neural networks as data-driven to estimate target accelerations and compensations of composite perturbations in real time using historical and current measurements. And the direction of gradient descent of the neural network is constrained by the motion model, thus the learning efficiency of the network and the interpretability of the proposed algorithm is improved. The proposed algorithm can simultaneously utilise historical trajectory information and domain knowledge as hybrid-driven to achieve complex manoeuvering target tracking with little prior information. Experimental results demonstrate the superiority of our algorithm in tracking accuracy, robustness and environmental adaptability.

Knowledge‐Informed Deep Neural Networks for Solar Flare Forecasting

AbstractRecently, although various deep learning techniques have been applied to building space weather prediction models, a large amount of relevant prior knowledge of solar eruptions and magnetic properties is ignored during the model development. By integrating prior knowledge in flare production into the convolutional neural network (CNN) structures, we have developed a knowledge‐informed deep neural network model aiming at forecasting solar flares. The line‐of‐sight magnetograms of Space‐weather HMI Active Region Patches (SHARP) from May 2010 to December 2018 are selected. We have surveyed the relationships between solar flares and both the active region (AR) area and magnetic type. When integrating prior knowledge into the CNNs, three aspects are considered: (a) keeping the magnetic structure unchanged (data preprocessing) while filling SHARP magnetograms into squares, (b) grouping the data samples into two subsets according to different flare productivities (sample grouping), and (c) adding AR area as an extra input parameter to the CNN (extra input parameter implementation). Pure CNN model, Fusion model 1, and Fusion model 2 are built to forecast M‐class or above flares in the next 48 hr, which involve data preprocessing, data preprocessing and sample grouping, and all the three aspects, respectively. Fusion model 2 that augments the most prior knowledge has the best performance. Our results imply that prior knowledge can play an important role in building deep learning flare forecasting models. In the future, adopting knowledge‐informed deep neural networks will be an effective way to further improve the forecasting performance for other space weather events.

A Better Mechanistic Understanding of Big Data through an Order Search Using Causal Bayesian Networks

Every year, biomedical data is increasing at an alarming rate and is being collected from many different sources, such as hospitals (clinical Big Data), laboratories (genomic and proteomic Big Data), and the internet (online Big Data). This article presents and evaluates a practical causal discovery algorithm that uses modern statistical, machine learning, and informatics approaches that have been used in the learning of causal relationships from biomedical Big Data, which in turn integrates clinical, omics (genomic and proteomic), and environmental aspects. The learning of causal relationships from data using graphical models does not address the hidden (unknown or not measured) mechanisms that are inherent to most measurements and analyses. Also, many algorithms lack a practical usage since they do not incorporate current mechanistic knowledge. This paper proposes a practical causal discovery algorithm using causal Bayesian networks to gain a better understanding of the underlying mechanistic process that generated the data. The algorithm utilizes model averaging techniques such as searching through a relative order (e.g., if gene A is regulating gene B, then we can say that gene A is of a higher order than gene B) and incorporates relevant prior mechanistic knowledge to guide the Markov chain Monte Carlo search through the order. The algorithm was evaluated by testing its performance on datasets generated from the ALARM causal Bayesian network. Out of the 37 variables in the ALARM causal Bayesian network, two sets of nine were chosen and the observations for those variables were provided to the algorithm. The performance of the algorithm was evaluated by comparing its prediction with the generating causal mechanism. The 28 variables that were not in use are referred to as hidden variables and they allowed for the evaluation of the algorithm’s ability to predict hidden confounded causal relationships. The algorithm’s predicted performance was also compared with other causal discovery algorithms. The results show that incorporating order information provides a better mechanistic understanding even when hidden confounded causes are present. The prior mechanistic knowledge incorporated in the Markov chain Monte Carlo search led to the better discovery of causal relationships when hidden variables were involved in generating the simulated data.

Challenges of modelling approaches for network meta-analysis of time-to-event outcomes in the presence of non-proportional hazards to aid decision making: Application to a melanoma network.

BackgroundSynthesis of clinical effectiveness from multiple trials is a well-established component of decision-making. Time-to-event outcomes are often synthesised using the Cox proportional hazards model assuming a constant hazard ratio over time. However, with an increasing proportion of trials reporting treatment effects where hazard ratios vary over time and with differing lengths of follow-up across trials, alternative synthesis methods are needed.ObjectivesTo compare and contrast five modelling approaches for synthesis of time-to-event outcomes and provide guidance on key considerations for choosing between the modelling approaches.MethodsThe Cox proportional hazards model and five other methods of estimating treatment effects from time-to-event outcomes, which relax the proportional hazards assumption, were applied to a network of melanoma trials reporting overall survival: restricted mean survival time, generalised gamma, piecewise exponential, fractional polynomial and Royston-Parmar models.ResultsAll models fitted the melanoma network acceptably well. However, there were important differences in extrapolations of the survival curve and interpretability of the modelling constraints demonstrating the potential for different conclusions from different modelling approaches.ConclusionThe restricted mean survival time, generalised gamma, piecewise exponential, fractional polynomial and Royston-Parmar models can accommodate non-proportional hazards and differing lengths of trial follow-up within a network meta-analysis of time-to-event outcomes. We recommend that model choice is informed using available and relevant prior knowledge, model transparency, graphically comparing survival curves alongside observed data to aid consideration of the reliability of the survival estimates, and consideration of how the treatment effect estimates can be incorporated within a decision model.

The negative reminding effect: Reminding impairs memory for contextual information

Encountering events that are meaningfully related to prior episodes can prompt retrieval of those prior events. Reminders are events that prompt retrieval of prior learned information, while reminded materials are the events that are retrieved in response. Reminders are an important component of efficient and effective cognition because they partially automate the process of bringing relevant prior knowledge to bear on novel situations. Across four experiments, we investigated whether reminders boosts memory for the entire prior reminded episode or for only specific aspects of prior experiences that are relevant to the reminder. To do so, we combined a reminding procedure with a paradigm for measuring memory for the incidental context of encoding. Participants studied lists of words in which semantically related pairs (e.g., “volcano” and “erupt”) were presented across brief lags and in different color contexts. Memory for the content and color of the first-presented word in pairs (the reminded information) was measured. Recall of related word pairs was consistently greater than recall of unrelated pairs, in agreement with prior work on the reminding effect. However, when the color of related words differed across the pair, memory for the color of the reminded words was impaired compared to unrelated words. The results support a view of reminding as a leveling and sharpening process in which memory for the focal content of reminded episodes is enhanced, but differences in peripheral details are smoothed over. Such a process can lead to the acquisition and application of prototypical knowledge over experience.

KGGCN: Knowledge-Guided Graph Convolutional Networks for Distantly Supervised Relation Extraction

Distantly supervised relation extraction is the most popular technique for identifying semantic relation between two entities. Most prior models only focus on the supervision information present in training sentences. In addition to training sentences, external lexical resource and knowledge graphs often contain other relevant prior knowledge. However, relation extraction models usually ignore such readily available information. Moreover, previous works only utilize a selective attention mechanism over sentences to alleviate the impact of noise, they lack the consideration of the implicit interaction between sentences with relation facts. In this paper, (1) a knowledge-guided graph convolutional network is proposed based on the word-level attention mechanism to encode the sentences. It can capture the key words and cue phrases to generate expressive sentence-level features by attending to the relation indicators obtained from the external lexical resource. (2) A knowledge-guided sentence selector is proposed, which explores the semantic and structural information of triples from knowledge graph as sentence-level knowledge attention to distinguish the importance of each individual sentence. Experimental results on two widely used datasets, NYT-FB and GDS, show that our approach is able to efficiently use the prior knowledge from the external lexical resource and knowledge graph to enhance the performance of distantly supervised relation extraction.

Research on Image Deblurring Processing Technology Based on Genetic Algorithm

Image deblurring, which is an important branch of digital image processing, is one of the difficulties in digital image processing. The main goal of image deblurring is to improve the quality of the given image, and to reconstruct or restore the original image using the relevant prior knowledge. Genetic algorithm (GA) is a global optimization search algorithm, which can quickly and effectively calculate complex multidimensional data area and indirectly. Now,genetic algorithm is slowly showing its excellent performance in the field of image processing. Therefore, based on the fuzzy features of the image, this paper uses the image enhancement algorithm to realize the image deblurring by capturing the gray value of the pixels in the image, and the improved genetic algorithm is used to better select the image threshold. The experimental results show that the new algorithm obtains higher definition than the traditional image deblurring method, and the improved genetic operator algorithm improves the degree of noise reduction.

Children’s learning from implicit analogies during instruction: Evidence from fraction division

This study examined children’s learning from implicit and explicit analogies during instruction. In all conditions, 6th grade children learned about fraction division from a videotaped lesson. In the implicit analogy, the teacher presented whole number division examples, thus activating relevant prior knowledge, and then introduced fraction division without drawing attention to the analogy. In the explicit analogy, the teacher introduced the same examples, drew children’s attention to the analogy, and guided children’s mapping across the analogical similarities. Children in the no analogy condition only viewed examples of fraction division. Children who viewed the implicit analogical instruction had greater conceptual understanding of fraction division at posttest than children in the explicit or no analogy conditions. Children in the explicit condition performed worse on procedural posttest items than children in the implicit or no analogy conditions. Activating children’s prior knowledge of conceptually similar problems was more effective at supporting children’s learning than explicitly guiding that analogy.

Deep Unsupervised Embedding for Remotely Sensed Images Based on Spatially Augmented Momentum Contrast

Convolutional neural networks (CNNs) have achieved great success when characterizing remote sensing (RS) images. However, the lack of sufficient annotated data (together with the high complexity of the RS image domain) often makes supervised and transfer learning schemes limited from an operational perspective. Despite the fact that unsupervised methods can potentially relieve these limitations, they are frequently unable to effectively exploit relevant prior knowledge about the RS domain, which may eventually constrain their final performance. In order to address these challenges, this article presents a new unsupervised deep metric learning model, called spatially augmented momentum contrast (SauMoCo), which has been specially designed to characterize unlabeled RS scenes. Based on the first law of geography, the proposed approach defines spatial augmentation criteria to uncover semantic relationships among land cover tiles. Then, a queue of deep embeddings is constructed to enhance the semantic variety of RS tiles within the considered contrastive learning process, where an auxiliary CNN model serves as an updating mechanism. Our experimental comparison, including different state-of-the-art techniques and benchmark RS image archives, reveals that the proposed approach obtains remarkable performance gains when characterizing unlabeled scenes since it is able to substantially enhance the discrimination ability among complex land cover categories. The source codes of this article will be made available to the RS community for reproducible research.

The virtual field trip: Investigating how to optimize immersive virtual learning in climate change education

AbstractImmersive Virtual Reality (IVR) is being used for educational virtual field trips (VFTs) involving scenarios that may be too difficult, dangerous or expensive to experience in real life. We implemented an immersive VFT within the investigation phase of an inquiry‐based learning (IBL) climate change intervention. Students investigated the consequences of climate change by virtually traveling to Greenland and exploring albedo and greenhouse effects first hand. A total of 102 seventh and eighth grade students were randomly assigned to one of two instructional conditions: (1) narrated pretraining followed by IVR exploration or (2) the same narrated training material integrated within the IVR exploration. Students in both conditions showed significant increases in declarative knowledge, self‐efficacy, interest, STEM intentions, outcome expectations and intentions to change behavior from the pre‐ to post‐assessment. However, there was a significant difference between conditions favoring the pretraining group on a transfer test consisting of an oral presentation to a fictitious UN panel. The findings suggest that educators can choose to present important prerequisite learning content before or during a VFT. However, adding pretraining may lead to better transfer test performance, presumably because it helps reduce cognitive load while learning in IVR. Practitioner NotesWhat is already known about this topic? Immersive virtual reality (IVR) simulations lead to higher presence but may lead to less learning when the content is not designed based on the affordances of the technology. One explanation for this finding is that cognitive load may be higher in IVR. The pretraining principle (ie, individuals learn more deeply from interactive multimodal learning environments when they receive pretraining on relevant prior knowledge) can be particularly effective in IVR‐based learning compared to learning through a video. Evidence shows that instructional design principles such as segmentation and generative learning strategies such as summarization can improve learning in IVR simulations. What this paper adds An investigation of the value of two different approaches to designing immersive virtual field trips (VFTs) within a real educational middle school context. Evidence that VFTs featuring IVR climate change simulations, in the context of inquiry‐based learning (IBL), can increase important variables such as declarative knowledge, interest in science and intentions to take climate action in seventh and eighth grade students. Evidence that presenting important learning content before a VFT leads to higher transfer scores. Implications for practice and/or policy Implementing an immersive VFT within the context of an IBL intervention provides students with relevant and engaging learning experiences and results in increased knowledge and interest in science. In the design of instruction using VFTs, educators can choose to either present prerequisite learning content prior to a VFT or during a VFT. However, adding pretraining has an advantage in terms of higher transfer scores.

Look it up: Online search reduces the problematic effects of exposures to inaccuracies.

People often reproduce information they read, which is beneficial when that information is accurate. Unfortunately, people are also often exposed to inaccurate information, with subsequent reproductions allowing for problematic decisions and behaviors. One empirically validated consequence of exposures to inaccuracies is that after reading falsehoods, participants are more likely to make errors answering related questions than if they previously read accurate statements, particularly for unfamiliar information. Interventions designed to attenuate these reproductions are often ineffective, at least as studied in tasks that restrict participants to generating answers based on text content and relevant prior knowledge. In the real world, however, people have access to outside resources to evaluate information. In three experiments, we tested whether affording the option to search for relevant online information following exposure to inaccurate statements would reduce reproductions of those inaccuracies on a post-reading task. Participants given the opportunity to search for information were less likely to reproduce inaccurate information and more likely to produce correct responses, in comparison to the performance of participants who were not allowed to search. We also tested whether warnings about potentially inaccurate information would encourage searches and inform responses. While warnings increased searching, additional reductions in inaccurate reproductions were not observed. Given the contingencies of many lab tasks, reproductions of inaccurate information might be overestimated. Resources available in the real world can offer useful supports for reducing the influence of and uncertainty associated with inaccurate exposures, consistent with contemporary accounts of memory and comprehension.

The impact of an adaptive e-course on students’ achievements based on the students’ prior knowledge

Researchers emphasize that prior knowledge is one of the important factors that influence learning. This study discusses the design and implementation of adapting an e-course to students’ prior knowledge using the Learning Management System (LMS). It aims to investigate the impact of such adaptive e-course to the learner’s prior knowledge through his or her achievement. To fulfill the study’s aim, a quasi-experimental approach was utilized. The study involved 130 students assigned randomly to two groups. An experimental group learned using an adaptive e-course while the other group was taught using traditional teaching methods. Findings of the study show that the experimental group outperformed the control group in academic achievement. The study also finds that activating relevant prior knowledge offered to the beginner learners helped in minimizing the performance gap between them and their advanced peers.