Integrated Learning Framework Research Articles

PSATF-6mA: an integrated learning fusion feature-encoded DNA-6 mA methylcytosine modification site recognition model based on attentional mechanisms

DNA methylation is of crucial importance for biological genetic expression, such as biological cell differentiation and cellular tumours. The identification of DNA-6mA sites using traditional biological experimental methods requires more cumbersome steps and a large amount of time. The advent of neural network technology has facilitated the identification of 6 mA sites on cross-species DNA with enhanced efficacy. Nevertheless, the majority of contemporary neural network models for identifying 6 mA sites prioritize the design of the identification model, with comparatively limited research conducted on the statistically significant DNA sequence itself. Consequently, this paper will focus on the statistical strategy of DNA double-stranded features, utilising the multi-head self-attention mechanism in neural networks applied to DNA position probabilistic relationships. Furthermore, a new recognition model, PSATF-6 mA, will be constructed by continually adjusting the attentional tendency of feature fusion through an integrated learning framework. The experimental results, obtained through cross-validation with cross-species data, demonstrate that the PSATF-6 mA model outperforms the baseline model. The in-Matthews correlation coefficient (MCC) for the cross-species dataset of rice and m. musus genomes can reach a score of 0.982. The present model is expected to assist biologists in more accurately identifying 6 mA locus and in formulating new testable biological hypotheses.

Investigating the contributing factors of urban crash levels: A novel stacking integrated learning framework

Crash contributing factors identification plays crucial role in preventing crashes and informing decision-making processes. However, current methods heavily rely on subjective judgments by technical experts, neglecting a comprehensive and scientific analysis. To address this gap, we propose a research framework that utilizes stacking integrated learning to predict crash risk levels and identify crash contributing factors. This framework first utilizes ArcGIS to construct adaptive buffers and obtain multi-source data (i. e., street view data, POI data, crash data). Then we integrate four single models (e.g., RF, SVM, XGBoost, NBC) as the base models, and treat their prediction results on the training data as a new feature set. We then train a meta model with the true labels as the supervision signal, thereby fusing the results of models into the meta model. The most effective model is determined through performance comparison, and the mean relative importance (MRI) of crash contributing factors on a spatial scale is derived from the results. The research results indicate that POIs have the highest MRI of 5.99% among all crash contributing factors ranking, followed by road signage markings (1.98%). These results contribute to enhancing transportation system performance, and intervening in hazardous areas in advance to reduce the occurrence of crashes.

A transformative model for teaching and learning emotional intelligence for aspiring entrepreneurial leaders

ABSTRACT Entrepreneurial leaders, social-science researchers, and small business practitioners call for the early development of emotional intelligence (EI). Although several programs integrate key EI concepts, there are few examples of how to develop an integrated learning framework for teaching and learning EI skills for daily use. This article introduces a practical transformative model of emotional intelligence (TEI) that emphasizes the personally meaningful development of personal, emotional, relationship, and life skills that are so important for school, work, and life success. A 1-day seminar experience embedded in a European-based international summer school course in the School of Management Fribourg, Fribourg, Switzerland, generated new evidence for how a skills-based curriculum in EI can be used to empower future entrepreneurs and leaders. The study reveals new findings for integrating a practical approach for engendering TEI for leaders.

Unveiling the Potential of Migrasomes: A Machine-Learning-Driven Signature for Diagnosing Acute Myocardial Infarction.

Recent studies have demonstrated that the migrasome, a newly functional extracellular vesicle, is potentially significant in the occurrence, progression, and diagnosis of cardiovascular diseases. Nonetheless, its diagnostic significance and biological mechanism in acute myocardial infarction (AMI) have yet to be fully explored. To remedy this gap, we employed an integrative machine learning (ML) framework composed of 113 ML combinations within five independent AMI cohorts to establish a predictive migrasome-related signature (MS). To further elucidate the biological mechanism underlying MS, we implemented single-cell RNA sequencing (scRNA-seq) of cardiac Cd45+ cells from AMI-induced mice. Ultimately, we conducted mendelian randomization (MR) and molecular docking to unveil the therapeutic effectiveness of MS. MS demonstrated robust predictive performance and superior generalization, driven by the optimal combination of Stepglm and Lasso, on the expression of nine migrasome genes (BMP1, ITGB1, NDST1, TSPAN1, TSPAN18, TSPAN2, TSPAN4, TSPAN7, TSPAN9, and WNT8A). Notably, ITGB1 was found to be predominantly expressed in cardiac macrophages in AMI-induced mice, mechanically regulating macrophage transformation between anti-inflammatory and pro-inflammatory. Furthermore, we showed a positive causality between genetic predisposition towards ITGB1 expression and AMI risk, positioning it as a causative gene. Finally, we showed that ginsenoside Rh1, which interacts closely with ITGB1, could represent a novel therapeutic approach for repressing ITGB1. Our MS has implications in forecasting and curving AMI to inform future diagnostic and therapeutic strategies for AMI.

Integrated learning framework for multistep pick-place-arrange of arbitrarily shaped objects in a narrow crate

To maximize the space occupied by unknown objects in a narrow crate and handle uncertain situations, we propose a novel human-like system-level learning framework: sequential infiltration synthesis of combined learning (SISCL). This framework leverages the synergies between picking by grasping, three-dimensional modeling, preferable placement, and seamless arranging by pushing in a systematic perspective. Within SISCL, a closed-loop system organizes different designing methods including reinforcement learning, knowledge induction, and analytic methods. This integration trains the system to autonomously package unknown objects and accommodate for uncertainties in measurements or executions, combining efficiency and flexibility. Initially trained in simulations, the SISCL framework underwent comprehensive evaluation in the laboratory. Furthermore, an initial validation was conducted on an extraterrestrial test ground. The results demonstrate the adaptability of the proposed framework to different shapes, materials, and stiffness of objects, as well as the capacity for uncertainties, thereby considerably reducing the need for precision in each sub-procedure in a human-like manner, facilitating seamless operations and efficient utilization of limited space.

Single-cell RNA sequencing integrated with bulk RNA sequencing analysis identifies a tumor immune microenvironment-related lncRNA signature in lung adenocarcinoma

BackgroundRecently, long non-coding RNAs (lncRNAs) have been demonstrated as essential roles in tumor immune microenvironments (TIME). Nevertheless, researches on the clinical significance of TIME-related lncRNAs are limited in lung adenocarcinoma (LUAD).MethodsSingle-cell RNA sequencing and bulk RNA sequencing data are integrated to identify TIME-related lncRNAs. A total of 1368 LUAD patients are enrolled from 6 independent datasets. An integrative machine learning framework is introduced to develop a TIME-related lncRNA signature (TRLS).ResultsThis study identified TIME-related lncRNAs from integrated analysis of single‑cell and bulk RNA sequencing data. According to these lncRNAs, a TIME-related lncRNA signature was developed and validated from an integrative procedure in six independent cohorts. TRLS exhibited a robust and reliable performance in predicting overall survival. Superior prediction performance barged TRLS to the forefront from comparison with general clinical features, molecular characters, and published signatures. Moreover, patients with low TRLS displayed abundant immune cell infiltration and active lipid metabolism, while patients with high TRLS harbored significant genomic alterations, high PD-L1 expression, and elevated DNA damage repair (DDR) relevance. Notably, subclass mapping analysis of nine immunotherapeutic cohorts demonstrated that patients with high TRLS were more sensitive to immunotherapy.ConclusionsThis study developed a promising tool based on TIME-related lncRNAs, which might contribute to tailored treatment and prognosis management of LUAD patients.

Lithium-ion battery state of health estimation method based on variational quantum algorithm optimized stacking strategy

Accurate state-of-health (SOH) estimation is critical for the performance and safety of lithium-ion batteries. An innovative method for SOH estimation is proposed by employing a variational quantum algorithm to optimize a stacking integrated learning strategy. The strategy effectively combines multiple model advantages, enhancing the estimation accuracy and generalizability. Using this method, eight sets of health factors are extracted, focusing on the relationship between battery capacity degradation and electrothermal parameters. A stacking integrated learning framework is developed by utilizing diverse primary learners to effectively capture the dynamic changes in health factors. A ridge regression meta-learner is incorporated to address overfitting problems found in primary learners. A significant innovation is the integration of a variational quantum circuit module as the primary learner. This module plays a crucial role in optimizing the hyperparameters for the analysis of complex and high-dimensional battery data. The effectiveness of the method is validated using four different types of batteries, showing a 77.4% improvement in prediction accuracy compared with traditional methods, with the SOH estimation error maintained within a tight margin of 0.67%. The mean absolute error, mean absolute percentage error, and root mean square error with maximum reduction rates are 76.7%, 77.4%, and 62.7%, respectively. The maximum increase in the R-squared coefficient is 5.3%. This study demonstrates the potential of variational quantum algorithms in enhancing the SOH estimation accuracy and opens new possibilities for the advanced health status management of lithium-ion batteries.

Integrated Project Development through Combined Theory and Practices of Core Courses focusing on Software Development Skills: Integrated Learning Framework

The National Education Policy promotes moving from the conventional content-heavy and memorization learning practice towards holistic learning/integrated learning. It imparts a creative and multidisciplinary curriculum that focuses equally on curriculum and assessment. All educational establishments assess students using written examinations, quizzes, seminars, term paper writing, and course projects. A semester typically includes 4-5 courses, and students must earn credits for these courses by scoring a good Cumulative Grade Point Average (CGPA). As course projects offer depth knowledge/holistic learning/lifelong learning of a course for the student, many courses include course projects as one of the activities in the course. If all the courses are intended to include course projects as a mandatory pedagogy, it will be difficult for students to acquire in-depth knowledge and required skills while also dealing with stress. So we are proposing an integrated learning framework by applying the theory and practices of two core courses- Software Engineering and Web Technologies to develop a web application. This integrated learning focuses on developing software development and software testing skills in computer science for undergraduate students pursuing a bachelor of engineering degree. This framework alleviated the pressure on students during placement and created job opportunities in software development. The framework consists of three important phases- The first phase includes the identification of the problem as a need for customers, writing requirements and analyzing the same. Students apply modular design principles and break down the codebase into distinct modules. This technique enhances code organization, reusability, and maintainability. The second phase focused on developing the front end by harnessing the power of Angular, a leading web framework, to craft a sleek and interactive user interface. The backend is built using Node.js, which serves as the foundation, enabling the software system to cater to highperformance server environments. These modules communicated seamlessly through well-defined APIs, facilitating the integration of various components within the application, ultimately delivering a seamless and responsive user experience. An industry expert conducted a workshop on Angularsoftware testing workshop was conducted for students by industry experts to expose the students to designing test cases, test plans, and testing strategies. The hands-on experience on testing tools was provided during the workshop. Faculty reviews are conducted on each phase, and rubrics-based assessment is done on each phase. Approximately sixty teams created web-based applications for real-world scenarios. Positive aspects of the framework in feedback indicated that more than 87% of the students agreed that they could apply Software engineering principles and practices such as requirements management. modular design and testing in web applications. Also, more than 85% of students acquire skills from code-to-web design mastery by developing web applications in Angular Node.js and backend implementation. This framework helped to improve teamwork, presentation and communication skills. Confidence in software development improved to a greater extent. The design and implementation of the framework met the stated outcome of the courses. The student's academic performance improved by 10% compared to the previous year when students were not involved in the integrated project development. Keywords— Framework, learning, practices, project, skills, technology, testing, web

An Integrated Learning Framework for Seamless High-Resolution Soil Moisture Estimation

An Integrated Learning Framework for Seamless High-Resolution Soil Moisture Estimation

IHerd: an integrative hierarchical graph representation learning framework to quantify network changes and prioritize risk genes in disease.

Different genes form complex networks within cells to carry out critical cellular functions, while network alterations in this process can potentially introduce downstream transcriptome perturbations and phenotypic variations. Therefore, developing efficient and interpretable methods to quantify network changes and pinpoint driver genes across conditions is crucial. We propose a hierarchical graph representation learning method, called iHerd. Given a set of networks, iHerd first hierarchically generates a series of coarsened sub-graphs in a data-driven manner, representing network modules at different resolutions (e.g., the level of signaling pathways). Then, it sequentially learns low-dimensional node representations at all hierarchical levels via efficient graph embedding. Lastly, iHerd projects separate gene embeddings onto the same latent space in its graph alignment module to calculate a rewiring index for driver gene prioritization. To demonstrate its effectiveness, we applied iHerd on a tumor-to-normal GRN rewiring analysis and cell-type-specific GCN analysis using single-cell multiome data of the brain. We showed that iHerd can effectively pinpoint novel and well-known risk genes in different diseases. Distinct from existing models, iHerd's graph coarsening for hierarchical learning allows us to successfully classify network driver genes into early and late divergent genes (EDGs and LDGs), emphasizing genes with extensive network changes across and within signaling pathway levels. This unique approach for driver gene classification can provide us with deeper molecular insights. The code is freely available at https://github.com/aicb-ZhangLabs/iHerd. All other relevant data are within the manuscript and supporting information files.

Towards a citizen development andragogy: Low-code platforms, design thinking and knowledge-based dynamic capabilities

Addressing the ongoing deficit in technical skills and talent necessitates an effective management approach that can harness the power of accelerating innovation. In response to the challenge, this paper presents a comprehensive learning model called Citizen Development Andragogy (CDA), specifically tailored for adult learners. CDA is an integrative learning framework, designed for an array of academic programs from micro-credentialing to four-year degrees, which combines leading-edge technology, such as low-code tools and platforms and generative AI, design thinking for creativity, and the institutionally oriented, knowledge-based dynamic capabilities (KBDC) model. Additionally, CDA is epistemologically rooted in the individually focused constructivist learning theory, providing a consistent structure that caters to the learning needs of working adults, enhancing their ability to conceive, design, build, and implement digital solutions. The andragogical learning theory is designed for adult learners and has important implications regarding teaching and program differences for reaching traditional aged college students, emerging learners, versus more seasoned individuals. Emerging strategic components conveyed in this paper—via a discursive style—represent some first steps towards considering and building a citizen development andragogy. An initial CDA framework conceptualization is detailed in this paper, which explores its theoretical constructs, practical considerations, current limitations, and it also shares insights from early implementation efforts, and presents a variety of potential future research avenues. Importantly, the paper emphasizes CDA's role as an accessible gateway for introducing citizen development into adult education and upskilling programs. In a broader societal context, citizen development, facilitated by CDA, can serve as a vital influencer for larger 'citizen' involvement movements like citizen scientists, human rights observers, and environmentalists. As such, an effective, scalable CDA prototype has the potential to yield meaningful benefits, empowering individuals, and organizations to achieve impact on an unprecedented scale, ultimately benefiting society at large.

An Integrated New Deep Learning Framework for Reliable CSI Acquisition in Connected and Autonomous Vehicles

An Integrated New Deep Learning Framework for Reliable CSI Acquisition in Connected and Autonomous Vehicles

I-Code: An Integrative and Composable Multimodal Learning Framework

Human intelligence is multimodal; we integrate visual, linguistic, and acoustic signals to maintain a holistic worldview. Most current pretraining methods, however, are limited to one or two modalities. We present i-Code, a self-supervised pretraining framework where users may flexibly combine the modalities of vision, speech, and language into unified and general-purpose vector representations. In this framework, data from each modality are first given to pretrained single-modality encoders. The encoder outputs are then integrated with a multimodal fusion network, which uses novel merge- and co-attention mechanisms to effectively combine information from the different modalities. The entire system is pretrained end-to-end with new objectives including masked modality unit modeling and cross-modality contrastive learning. Unlike previous research using only video for pretraining, the i-Code framework can dynamically process single, dual, and triple-modality data during training and inference, flexibly projecting different combinations of modalities into a single representation space. Experimental results demonstrate how i-Code can outperform state-of-the-art techniques on five multimodal understanding tasks and single-modality benchmarks, improving by as much as 11% and demonstrating the power of integrative multimodal pretraining.

Detection of Moisture Content of Pinus massoniana Lamb. Seedling Leaf Based on NIR Spectroscopy with a Multi-Learner Model

The growth quality of Pinus massoniana (Lamb.) seedlings is closely related to the survival rate of afforestation. Moisture content detection is an important indicator in the cultivation of forest seedlings because it can directly reflect the adaptability and growth potential of the seedlings to the soil environment. To improve the accuracy of quantitative analysis of moisture content in P. massoniana seedlings using near-infrared spectroscopy, a total of 100 P. massoniana seedlings were collected, and their near-infrared diffuse reflectance spectra were measured in the range of 2500 to 800 nm (12,000 to 4000 cm−1). An integrated learning framework was introduced, and a quantitative detection model for moisture content in P. massoniana seedlings was established by combining preprocessing and feature wavelength selection methods in chemometrics. Our results showed that the information carried by the spectra after multiple scattering correction (MSC) preprocessing had a good response to the target attribute. The stacking learning model based on the full-band spectrum had a prediction coefficient of determination R2 of 0.8819, and the prediction accuracy of moisture content in P. massoniana seedlings could be significantly improved compared to the single model. After variable selection, the spectrum processed by MSC and feature selection with uninformative variable elimination (UVE) showed good prediction effects in all models. Additionally, the prediction coefficient of determination R2 of the support vector regression (SVR)—adaptive boosting (AdaBoost)—partial least squares regression (PLSR) + AdaBoost model reached 0.9430. This indicates that the quantitative analysis model of moisture content in P. massoniana seedlings established through preprocessing, feature selection, and stacking learning models can achieve high accuracy in predicting moisture content in P. massoniana seedlings. This model can provide a feasible technical reference for the precision cultivation of P. massoniana seedlings.

A novel framework combining production evaluation and quantification of development parameters for shale gas wells

Accurate prediction of estimated ultimate recovery (EUR) and quantification of important engineering and geological parameters play a decisive role in production parameter optimization and investment decision of shale gas wells. This study proposes an integrated learning framework for EUR prediction and production parameter quantification in shale gas wells using CatBoost algorithm. CatBoost is designed to identify complex relationships efficiently and accurately between geological and engineering parameters and target values EUR. Furthermore, the EUR prediction model based on CatBoost was used to identify important features. Various machine learning interpretation methods are used to visualize the marginal effects of important and interesting features in an attempt to optimize production parameters by quantifying the features. Experiments are conducted to evaluate the performance of the proposed the integrated learning framework in the deep shale gas well data set of Luzhou block in Sichuan Basin, China. Results demonstrate that compared with other mainstream ensemble learning algorithms, the proposed EUR prediction model based on CatBoost algorithm performs well on small data sets, with a prediction error of only 12.31%. The post-interpretation tool of the model was used to find out four important characteristics affecting EUR in Luzhou block, namely, contents of brittle minerals, fracturing length, porosity and fracturing section. The optimal combination of geological and engineering parameters for the block to maximize EUR is identified by using the established efficient and accurate EUR prediction model as input and the visualization of the marginal effects of the features of interest as output. This work can effectively provide guidance for EUR prediction and production parameter optimization of shale gas wells.

An integrative machine learning framework for classifying SEER breast cancer

Breast cancer is the commonest type of cancer in women worldwide and the leading cause of mortality for females. The aim of this research is to classify the alive and death status of breast cancer patients using the Surveillance, Epidemiology, and End Results dataset. Due to its capacity to handle enormous data sets systematically, machine learning and deep learning has been widely employed in biomedical research to answer diverse classification difficulties. Pre-processing the data enables its visualization and analysis for use in making important decisions. This research presents a feasible machine learning-based approach for categorizing SEER breast cancer dataset. Moreover, a two-step feature selection method based on Variance Threshold and Principal Component Analysis was employed to select the features from the SEER breast cancer dataset. After selecting the features, the classification of the breast cancer dataset is carried out using Supervised and Ensemble learning techniques such as Ada Boosting, XG Boosting, Gradient Boosting, Naive Bayes and Decision Tree. Utilizing the train-test split and k-fold cross-validation approaches, the performance of various machine learning algorithms is examined. The accuracy of Decision Tree for both train-test split and cross validation achieved as 98%. In this study, it is observed that the Decision Tree algorithm outperforms other supervised and ensemble learning approaches for the SEER Breast Cancer dataset.

Deep learning-assisted prediction of protein-protein interactions in Arabidopsis thaliana.

Currently, the experimentally identified interactome of Arabidopsis (Arabidopsis thaliana) is still far from complete, suggesting that computational prediction methods can complement experimental techniques. Motivated by the prosperity and success of deep learning algorithms and natural language processing techniques, we introduce an integrative deep learning framework, DeepAraPPI, allowing us to predict protein-protein interactions (PPIs) of Arabidopsis utilizing sequence, domain and Gene Ontology (GO) information. Our current DeepAraPPI comprises (i) a word2vec encoding-based Siamese recurrent convolutional neural network (RCNN) model, (ii) a Domain2vec encoding-based multiple layer perceptron (MLP) model, and (iii) a GO2vec encoding-based MLP model. Finally, DeepAraPPI combines prediction results of the three individual predictors through a logistic regression model. Compiling high-quality positive and negative training and test samples by applying strict filtering strategies, DeepAraPPI shows superior performance compared to existing state-of-the-art Arabidopsis PPI prediction methods. DeepAraPPI also provides better cross-species predictive ability in rice (Oryza sativa) than traditional machine learning methods, although the overall performance in cross-species prediction remains to be improved. DeepAraPPI is freely accessible at http://zzdlab.com/deeparappi/. In the meantime, we have also made the source code and datasets of DeepAraPPI available at https://github.com/zjy1125/DeepAraPPI.

An integrated learning framework of corporate training system: a grounded theory approach

PurposeRooting in the literature on training and laying on Kirkpatrick model, this paper aims to explore key drivers of corporate training to identify how they can be combined into an integrated framework of learning for human capital development.Design/methodology/approachBy adopting the constructivist grounded theory, this contribution analyzes the experience carried out in the last ten years by Virvelle, an Italian corporate training firm.FindingsResults show the rise of five core categories, g1iving rise to an integrated model of Kirkpatrick. Their dynamic interplay led to a new orientation of Kirkpatrick model giving rise to a metalearning ecosystem.Research limitations/implicationsManagerial implications have identified key factors on which building and implementing appropriate corporate training programmes capable of triggering co-generative processes of value creation. Particularly, the essential role of learning quality culture, digital technology and personalization are detected in integrating not only hard but furthermore soft shades of learning. Concerning theoretical implications, the emergence of key structural and systems enabling dimensions for learning, and contextual mechanisms involved in reshaping training effectiveness and achieving integrated learning outcomes are detected. The main limitation of this study lies in the need to generalize results: the conceptualized framework needs to be empirically tested.Originality/valueThe value of this research is built along three main points. The first is the integration among the core categories that an integrated learning system can be built on, promoting learning quality culture through positive feedback loops. The second is represented by the chance to enhance an integrated mutual knowledge development among engaged actors, thereby shaping a more holistic and multidimensional learning model. The third is related to the transversal role that digital technology plays in all phases of the training process as it integrates and enriches them.

Representation learning for knowledge fusion and reasoning in Cyber–Physical–Social Systems: Survey and perspectives

The digital deep integration of cyber space, physical space and social space facilitates the formation of Cyber–Physical–Social Systems (CPSS). Knowledge empowers CPSS to be capable of solving complex tasks and providing intelligent services. Knowledge representation, fusion, and reasoning are complementary in the knowledge graph and worthy of continuous exploration and in-depth research. Since the concept of knowledge representation learning has been proposed, various models have been developed and demonstrated their superior ability in knowledge fusion and reasoning. In this paper, we investigate the progress and function of representation learning models adopted in knowledge fusion and reasoning, and extract their commonalities and characteristics, which is a perspective that has never been developed before, to provide new ideas for scholars and realize the mutual promotion of knowledge fusion and reasoning. This paper comprehensively reviews classic methods and investigates advanced and emerging works, which are published in authoritative journals or conferences. Furthermore, to overcome the limitations of existing methods, we propose an integrated knowledge representation learning and application framework for CPSS and a series of tensor-based knowledge fusion and reasoning models. Finally, four future directions are discussed in this paper, we analyze the challenges faced by them and suggest promising solutions.

STEM Implementation Issues in Indonesia: Identifying the Problems Source and Its Implications

The last few years, STEM education received more attention in the world, including Indonesia. There are many teachers who have participated in socialization and applied STEM in their schools. Therefore, this study aims to describe the various obstacles faced by teachers implementing STEM learning. This qualitative study uses a phenomenological methodology. Data is collected from in-depth interviews with nine teachers who have information and knowledge related to STEM as an integrative learning framework and have participated in socialization, workshops, or seminars. The stages of data analysis are doing data reduction, determining themes, making links between themes, and concluding findings. The conclusions of the study are three broad categories of things that challenge schools in implementing STEM learning in Indonesia, namely policy support and limited learning facilities, complex learning management, and the ability of teachers to manage students that are very diverse.