Building Domain Knowledge Research Articles

Time to transfer: Long-term effects of a sustained and spiraled content literacy intervention in the elementary grades.

We investigated the effectiveness of a sustained and spiraled content literacy intervention that emphasizes building domain and topic knowledge schemas and vocabulary for elementary-grade students. The model of reading engagement intervention underscores thematic lessons that provide an intellectual structure for helping students connect new learning to a general schema in Grade 1 (animal survival), Grade 2 (scientific investigation of past events like dinosaur mass extinctions), and Grade 3 (scientific investigation of living systems). A total of 30 elementary schools (N = 2,870 students) were randomized to a treatment or control condition. In the treatment condition (i.e., full spiral curriculum), students participated in content literacy lessons from Grades 1 to 3 during the school year and wide reading of thematically related informational texts in the summer following Grades 1 and 2. In the control condition (i.e., partial spiral curriculum), students participated in lessons in only Grade 3. The Grade 3 lessons for both conditions were implemented online during the COVID-19 pandemic school year. Results reveal that treatment students outperformed control students on science vocabulary knowledge across all three grades. Furthermore, intent-to-treat analyses revealed positive transfer effects on Grade 3 science reading (ES = .14), domain-general reading comprehension (ES = .11), and mathematics achievement (ES = .12). Treatment impacts were sustained at 14-month follow-up on Grade 4 reading comprehension (ES = .12) and mathematics achievement (ES = .16). Findings indicate that a content literacy intervention that spirals topics and vocabulary across grades can improve students' long-term academic achievement outcomes. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Remote supervised relationship extraction method of clustering for knowledge graph in aviation field

In the process of building domain knowledge graph, the result of relationship extraction between entities is an important guarantee of the quality of the graph. Therefore, we propose a clustering method based on reinforcement learning for remote supervised relation extraction. For the relationship extraction of accident information in the aviation domain mapping, a clustering method combining local dense and global dissimilarity is proposed in combination with remote supervision, which can obtain a large amount of low-noise labeled data and reduce part of the wrong labeling and missing labeling due to the strong specialization in the aviation domain; meanwhile, reinforcement learning is introduced to denoise the negative instance noise in the positive sample data; Finally, we propose a two-attention segmentation (DAPCNN) relationship extraction model to mine deep semantic sentences. The experimental results show that in the civil aviation relationship extraction text constructed in this paper, the Micro_R, Micro_P and Micro_F1 values of the proposed relationship extraction method reach 83.41 %, 84.16 % and 83.96 %. In the open relationship extraction dataset DuIE, The Micro_R, Micro_P and Micro_F1 of the proposed method are up to 83.41 %, 93.58 % and 94.02 % respectively. Compared with the current advanced multi-instance and multi-label model, the proposed method can more accurately extract the relationship between aviation accident entities. At the same time, the performance of the open data set is also good, and has a certain universality.

Chinese Clinical Named Entity Recognition From Electronic Medical Records Based on Multisemantic Features by Using Robustly Optimized Bidirectional Encoder Representation From Transformers Pretraining Approach Whole Word Masking and Convolutional Neural Networks: Model Development and Validation

Clinical electronic medical records (EMRs) contain important information on patients' anatomy, symptoms, examinations, diagnoses, and medications. Large-scale mining of rich medical information from EMRs will provide notable reference value for medical research. With the complexity of Chinese grammar and blurred boundaries of Chinese words, Chinese clinical named entity recognition (CNER) remains a notable challenge. Follow-up tasks such as medical entity structuring, medical entity standardization, medical entity relationship extraction, and medical knowledge graph construction largely depend on medical named entity recognition effects. A promising CNER result would provide reliable support for building domain knowledge graphs, knowledge bases, and knowledge retrieval systems. Furthermore, it would provide research ideas for scientists and medical decision-making references for doctors and even guide patients on disease and health management. Therefore, obtaining excellent CNER results is essential. We aimed to propose a Chinese CNER method to learn semantics-enriched representations for comprehensively enhancing machines to understand deep semantic information of EMRs by using multisemantic features, which makes medical information more readable and understandable. First, we used Robustly Optimized Bidirectional Encoder Representation from Transformers Pretraining Approach Whole Word Masking (RoBERTa-wwm) with dynamic fusion and Chinese character features, including 5-stroke code, Zheng code, phonological code, and stroke code, extracted by 1-dimensional convolutional neural networks (CNNs) to obtain fine-grained semantic features of Chinese characters. Subsequently, we converted Chinese characters into square images to obtain Chinese character image features from another modality by using a 2-dimensional CNN. Finally, we input multisemantic features into Bidirectional Long Short-Term Memory with Conditional Random Fields to achieve Chinese CNER. The effectiveness of our model was compared with that of the baseline and existing research models, and the features involved in the model were ablated and analyzed to verify the model's effectiveness. We collected 1379 Yidu-S4K EMRs containing 23,655 entities in 6 categories and 2007 self-annotated EMRs containing 118,643 entities in 7 categories. The experiments showed that our model outperformed the comparison experiments, with F1-scores of 89.28% and 84.61% on the Yidu-S4K and self-annotated data sets, respectively. The results of the ablation analysis demonstrated that each feature and method we used could improve the entity recognition ability. Our proposed CNER method would mine the richer deep semantic information in EMRs by multisemantic embedding using RoBERTa-wwm and CNNs, enhancing the semantic recognition of characters at different granularity levels and improving the generalization capability of the method by achieving information complementarity among different semantic features, thus making the machine semantically understand EMRs and improving the CNER task accuracy.

From semantic weight to legal ontology via classification of concepts in legal texts

ABSTRACT A difficulty with legal vocabulary is that concepts in different legal systems map imperfectly to each other, particularly from common law systems where classification is often unclear or convoluted to codified civil systems. Even within the English language domain there are numerous legal systems where concepts differ, e.g. between Scotland and England. This causes significant problems for students’, especially foreign language students’, reading comprehension and developing understanding of legal lexis where translation dictionaries, while they may be efficient, are imperfect. This article sets out a classification approach to reading and English language legal concept deep understanding rooted in theories from education and cognitive linguistics, in particular Categories and Prototypes, Schema theory, and Legitimation Code Theory (LCT). I have used the approach successfully for some years with international students studying law in Scotland, assisting their reading of textbooks and especially journal articles while building domain knowledge. The application of the strategy is presented in the context of research on the adoption of reading strategies which finds that adoption is influenced by awareness of the complexity of the concepts in the text, complexity of the strategy, and by how much particular strategies are seen as a valid method in legal study.

Orchestrating Learning Together and Development of Team-Trust in Neurologically Typical and Neurologically Atypical Students: A Multicase Study

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Contribution:</i> This multicase study compares computer orchestrated learning together (COLT) and student orchestrated learning together (SOLT) in cooperatively preparing first year university students for collaborative activities. COLT using computer orchestrated group learning environment (COGLE) was perceived efficient, effective, and inclusive for team effectiveness as it helps build domain knowledge and trust between neurologically typical (NT) and/or neurologically atypical (NAT) teammates. This study confirms existing and identifies two new antecedents for trust, namely, resolving cognitive conflicts and real-time updates to domain knowledge. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Background:</i> Industrial and organizational psychology literature links effective teamworking with domain knowledge and trust. Building team trust within a short period can be challenging, in particular in mixed teams of NT and NAT students. Facilitating teamwork can be resource hungry. Educational studies on trust and teamwork are therefore rare. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Research Questions:</i> This study investigates how orchestration affects teamwork by asking: 1) how does computer and student orchestration affect domain knowledge acquisition in NT and NAT students? and 2) how does computer and student orchestration affect the development of trust between NT and/or NAT students? <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Methodology:</i> Both qualitative and quantitative data were captured at multiple points within multiple (literal and theoretical replication) cases. Case summaries and a cross-case analysis provided further data and the methodological triangulation needed for analytical generalization. Findings: COGLE’s scaffolding and nonsocial prompts for cooperation, shared goal orientation, shared monitoring, and shared working not only helped NT and NAT students learn together but also helped develop team-trust quickly. Delayed interactions, low team-trust, and clique formation were seen in SOLT teams.

Effects of a Digital Game-Based Course in Building Adolescents' Knowledge and Social-Emotional Competencies.

Objective: The purpose of this study was to investigate the efficacy of a digital game-based course to build domain knowledge and social emotional competencies of empathy and compassion in adolescents. Materials and Methods: The study used a digital game Bury me, my Love with an accompanying course which was administered to 201 participants between ages 13-18 across United Arab Emirates (UAE) and India. Standardized self-reports were used to score participants on measures of knowledge and attitudes, empathy, and compassion before and after the intervention. Mixed analyses of variance were conducted with 1 between-subjects factor (gender) and 1 within-subjects factor (time) to determine the impact of the intervention, followed by post hoc t-tests. Results: Increased effects of intervention were obtained for both knowledge and social emotional learning in both UAE and India. Specifically, there was a significant increase in awareness of migration and refugees in both India (P < 0.001) and UAE (P < 0.001). Interesting effects of gender were seen in which females in both countries showed increases in compassion from others (P < 0.05). Conclusion: This study opens a new window in game-based learning. The design of a structured course with learning outcomes that are centered around a digital game establishes its potential to create engaging and accessible solutions to simultaneously build domain knowledge and social-emotional competencies in adolescents.

The sound of music: from increased personalization to therapeutic values

Music providers like Spotify leverage music recommendation systems to connect users with relevant music. Based on content-based and collaborative-filtering statistical methods, these machine learning algorithms quantify user-song probabilities and present the highest-ranked songs. However, most music providers do not fully address their users’ music seeking and retrieval needs. Likewise, the fields of Recommender Systems (RecSys), Music Recommendation Systems (MRS) and Music Information Retrieval (MIR) remain disconnected from real-world use cases of music seeking. In this conceptual paper, we review the literature of the RecSys, MRS, MIR and Music Therapy (MT) academic fields. We discuss trends towards greater user control and personalization in the MRS and MIR fields and the connections between MT and positive health outcomes such as reductions in stress, anxiety and heart rate. We argue that greater control and visibility into the characteristics of songs and recommended items can generate positive downstream benefits. We recommend features that empower users to better seek, find, store, retrieve and learn from their musical catalogs.Results. We suggest design enhancements that recognize music’s wider psychological and physiological benefits and create opportunities to build domain knowledge. Unlocking music’s myriad benefits through the enhancements proposed would catalyze positive outcomes for business stakeholders, users and society.

Improving students’ summative knowledge of introductory chemistry through the forward testing effect: examining the role of retrieval practice quizzing

Building domain knowledge is essential to a student's success in any course. Chemistry, similar to other STEM disciplines, has a strong cumulative element (i.e., topic areas continuously build upon prior coursework). We employed the testing effect, in the form of post-exam retrieval quizzes, as a way to improve students’ understanding of chemistry over an entire semester. Students (n = 146) enrolled in Introduction to Chemistry were presented with retrieval quizzes released one week after each during-term exam (that covered that exam's content). We measured students’ level of quiz participation, during-term exam scores (a control variable), and cumulative final exam scores to determine the effectiveness of implementing a post-exam retrieval quiz system. Most critically, students completing more than 50% of the retrieval quizzes performed significantly better (i.e., more than a half letter grade) on the cumulative final exam than those who were below 50% participation as determined by one-way between-subjects ANOVA and planned follow-up analyses. We found no significant differences between the participating groups on during-term exam scores, suggesting that high achieving students were not more likely than struggling students to participate in the practice testing (and thus benefit from it).

Square Kilometre Array Science Data Challenge 1: analysis and results

ABSTRACT As the largest radio telescope in the world, the Square Kilometre Array (SKA) will lead the next generation of radio astronomy. The feats of engineering required to construct the telescope array will be matched only by the techniques developed to exploit the rich scientific value of the data. To drive forward the development of efficient and accurate analysis methods, we are designing a series of data challenges that will provide the scientific community with high-quality data sets for testing and evaluating new techniques. In this paper, we present a description and results from the first such Science Data Challenge 1 (SDC1). Based on SKA MID continuum simulated observations and covering three frequencies (560, 1400, and 9200 MHz) at three depths (8, 100, and 1000 h), SDC1 asked participants to apply source detection, characterization, and classification methods to simulated data. The challenge opened in 2018 November, with nine teams submitting results by the deadline of 2019 April. In this work, we analyse the results for eight of those teams, showcasing the variety of approaches that can be successfully used to find, characterize, and classify sources in a deep, crowded field. The results also demonstrate the importance of building domain knowledge and expertise on this kind of analysis to obtain the best performance. As high-resolution observations begin revealing the true complexity of the sky, one of the outstanding challenges emerging from this analysis is the ability to deal with highly resolved and complex sources as effectively as the unresolved source population.

Retrieval system enhanced by fine‐grained knowledge entities

ABSTRACTAlong with the massive increase of academic full texts, users now have a growing need for fine‐grained academic literature retrieval. However, few studies have attempted to build domain knowledge retrieval system for academic literature in information science. Compared with other disciplines, information science pays more attention to the domain knowledge entities sequence including data resources, models and software. To fill the gap, this paper use academic full texts from JASIST. Using SVM, Text‐CNN, Bi‐LSTM and BERT to extract the knowledge entities sequence including data resource, model, software. The highest Precision, Recall and F1 are 92.45%, 93.86% and 93.14%, respectively. And the knowledge entity sequence indexing system is built based on the extracted knowledge entities sequences. It was found that the MRR score of the knowledge entities sequence retrieval system was higher by 13% as compared with the full text retrieval system.

AN INTEGRATED FRAMEWORK UTILISING SOFTWARE AGENT REASONING AND ONTOLOGY MODELS FOR SENSOR BASED BUILDING MONITORING

Smart building monitoring demands a new software infrastructure that can elaborate building domain knowledge in order to provide advanced and intelligent functionalities. Conventional facility management (FM) software tools lack semantically rich components, and that limits the capability of supporting software for automatic information sharing, resource negotiation and to assist in timely decision making. Recent hardware innovation on compact ZigBee sensor devices, software developments on ontology and intelligent software agent paradigms provide a good opportunity to develop tools that can further improve current FM practices. This paper introduces an integrated framework which includes a ZigBee based sensor network and underlying multi-agent software (MAS) components. Several different types of sensors were integrated with the ZigBee host devices to produce compact multi-functional sensor units. The MAS framework incorporates the belief-desire-intention (BDI) abstraction with ontology support (provided via explicit knowledge bases). The different software agent types have been developed to work with sensor hardware to conduct resource negotiation, to optimize battery utilization, to monitor building space in a non-intrusive way and to reason about its usage through real time ontology model queries. The deployed sensor network shows promising intelligent characteristics, and it has been applied in several on-going research projects as an underlying decision making service. More applications and larger deployments have been planned for future work.

Concern level assessment: Building domain knowledge into a visual system to support network-security situation awareness

Information officers and network administrators require tools to help them achieve situation awareness about potential network threats. We describe a response to mini-challenge 1 of the 2012 IEEE Visual Analytics Science and Technology challenge in which we developed a visual analytic solution to a network-security situation awareness problem. To support conceptual design, we conducted a series of knowledge elicitation sessions with domain experts. These provided an understanding of the information they needed to make situation awareness judgements as well as a characterisation of those judgements in the form of production rules, which define a parameter we called the ‘concern level assessment’. The concern level assessment was used to provide heuristic guidance within a visual analytic system called Middlesex Spatial Interactive Visualisation Environment. An analysis of Visual Analytics Science and Technology challenge assessment sessions using Middlesex Spatial Interactive Visualisation Environment provides some evidence that intelligent heuristics like these can provide useful guidance without unduly dominating interaction and understanding.

Automatically structuring domain knowledge from text: An overview of current research

This paper presents an overview of automatic methods for building domain knowledge structures (domain models) from text collections. Applications of domain models have a long history within knowledge engineering and artificial intelligence. In the last couple of decades they have surfaced noticeably as a useful tool within natural language processing, information retrieval and semantic web technology. Inspired by the ubiquitous propagation of domain model structures that are emerging in several research disciplines, we give an overview of the current research landscape and some techniques and approaches. We will also discuss trade-offs between different approaches and point to some recent trends.

Building of Domain Knowledge Base by Ontology

In this paper, method of building domain knowledge base is proposed through the comparison of ontology and other tools for knowledge expression. In addition, we lay out a model to describe the procedure vividly. Facts show that ontology is the best way to build domain knowledge.