Knowledge Mining Research Articles

Characterization of firebricks used in copper smelting furnaces of the 19th century, Atacama, Chile

In the 19th century, mining in northern Chile was of great economic importance. Today, the ruins of the copper smelters and abundant fragments of refractory bricks from their furnaces remain. The refractory materials have not been previously studied. the origin, physical and chemical properties, and compressive strength of 10 different samples of historical refractory bricks collected on-site were investigated. The characterization results showed that the bricks are silico-aluminous, unlike the magnesium-chromium bricks used by current foundries. It was determined that 90% of them were produced in the UK and 10% were manufactured in Chile. The phases present are mullite, quartz, cristobalite, and tridymite, and the chemical composition corresponds mainly to SiO2. The compressive strength of the historical bricks is similart to those currently used,. These results may contribute to the knowledge of the 19th century mining and metallurgical historical heritage of Chile and the UK.

An Interpretable Data-Driven Medical Knowledge Discovery Pipeline Based on Artificial Intelligence.

Difficulty in knowledge validation is a significant hindrance to knowledge discovery via data mining, especially automatic validation without artificial participation. In the field of medical research, medical knowledge discovery from electronic medical records is a common medical data mining method, but it is difficult to validate the discovered medical knowledge without the participation of medical experts. In this article, we propose a data-driven medical knowledge discovery closed-loop pipeline based on interpretable machine learning and deep learning; the components of the pipeline include Data Generator, Medical Knowledge Mining, Medical Knowledge Evaluation, and Medical Knowledge Application. In addition to completing the discovery of medical knowledge, the pipeline can also automatically validate the knowledge. We apply our pipeline's discovered medical knowledge to a traditional prognostic predictive model of heart failure in a real-world study, demonstrating that the incorporation of medical knowledge can effectively improve the performance of the traditional model. We also construct a scale model based on the discovered medical knowledge and demonstrate that it achieves good performance. To guarantee its medical effectiveness, every process of our pipeline involves the participation of medical experts.

A meme-based approach for knowledge mining, organization, and presentation of traditional Chinese settlement culture

Traditional Chinese settlements are successful examples of harmonious coexistence between man and nature. Traditional Chinese settlement culture (TCSC) is the product of its inhabitants' adaptation to the natural and social environment and has national cultural originality and transmission. Core cultural elements of these being chosen to stay can be considered as memes. This study considers traditional Chinese settlement cultural memes (TCSCM) as a special kind of knowledge, and proposes a meme-based approach to mine, organize, and present traditional Chinese settlement cultural knowledge from a knowledge management perspective. This approach is composed of identifying memes, constructing the memes ontology, and the memes knowledge graph, mainly describing and establishing the memes identification index system, the ontology conceptual framework, and knowledge graph construction rules. In addition, the feasibility of the above work is verified by a case study of Jinggang Ancient town in Hunan province. This study provides new practical ideas for the authentic conservation and industrial construction of TCSC.

Large-scale biomedical relation extraction across diverse types: model development, and usability study on COVID-19.

Biomedical relation extraction (RE) is of great importance for researchers to conduct systematic biomedical studies. It not only helps knowledge mining such as knowledge graph (KG) and novel knowledge discovery, but also promotes translational applications such as clinical diagnosis, decision making and precision medicine. However, the relations between biomedical entities are complex and diverse, and the comprehensive biomedical RE is not yet well established. This paper aims to investigate and improve large-scale RE with diverse relation types and conduct usability studies with application scenarios to optimize biomedical text mining. Datasets containing 125 relation types with different entity semantic levels are constructed to evaluate the impact of entity semantic information on RE, and performance analysis was conducted on different model architectures and domain models. This study also proposed a continued pre-training strategy and integrated models with scripts into tools. Furthermore, this study applied RE to the COVID-19 corpus with article topics and application scenarios of clinical interest to assess and demonstrate its biological interpretability and usability. The performance analysis revealed that RE achieves the best performance if the detailed semantic type is provided. For a single model, PubMedBERT with our continued pre-training performed best with an F1 score of 0.8998. The usability studies on COVID-19 demonstrated the interpretability and usability of RE, and a relation graph database is constructed, which is used to reveal existing and novel drug paths with edge explanations. The models (including pre-trained and fine-tuned), integrated tool (Docker), and generated data (including COVID-19 relation graph database and drug paths) are publicly available to the biomedical text mining community and clinical researchers. This study provided a comprehensive analysis of RE with diverse relation types. The optimized RE models and tools for diverse relation types are developed, which can be widely used in biomedical text mining. Our usability studies provided a proof-of-concept demonstration of how large-scale RE can be leveraged to facilitate novel research.

Mining knowledge: Nineteenth-century Cornish electrical science and the controversies of clay.

Michael Faraday's laboratory experiments have dominated traditional histories of the electrical sciences in 1820s and 1830s Britain. However, as this article demonstrates, in the mining region of Cornwall, Robert Were Fox fashioned a very different approach to the study of electromagnetic phenomena. Here, it was the mine that provided the foremost site of scientific experimentation, with Fox employing these underground locations to measure the Earth's heat and make claims over the existence of subterranean electrical currents. Yet securing philosophical claims cultivated in mines proved challenging for Fox, with metropolitan audiences, including Faraday, loath to give credit to the results of these underground experiments. This article explores how Fox developed a way of modeling his mine experiments, using clay samples, to communicate knowledge from industrial Cornwall to urban centers of elite science. It argues that the mine was an epistemologically complex venue of scientific activity, at once seeming to provide a way of examining nature directly, without recourse to laboratory contrivance, while simultaneously being a place where knowledge claims were hard to verify without access to these physically challenging locations. In exploring Fox's work, this study contributes to a growing literature of spatial investigation that takes the vertical as its unit of analysis.

Generative Artificial Intelligence GPT-4 Accelerates Knowledge Mining and Machine Learning for Synthetic Biology.

Knowledge mining from synthetic biology journal articles for machine learning (ML) applications is a labor-intensive process. The development of natural language processing (NLP) tools, such as GPT-4, can accelerate the extraction of published information related to microbial performance under complex strain engineering and bioreactor conditions. As a proof of concept, we proposed prompt engineering for a GPT-4 workflow pipeline to extract knowledge from 176 publications on two oleaginous yeasts (Yarrowia lipolytica and Rhodosporidium toruloides). After human intervention, the pipeline obtained a total of 2037 data instances. The structured data sets and feature selections enabled ML approaches (e.g., a random forest model) to predict Yarrowia fermentation titers with decent accuracy (R2 of 0.86 for unseen test data). Via transfer learning, the trained model could assess the production potential of the engineered nonconventional yeast, R. toruloides, for which there are fewer published reports. This work demonstrated the potential of generative artificial intelligence to streamline information extraction from research articles, thereby facilitating fermentation predictions and biomanufacturing development.

Autonomous complex knowledge mining and graph representation through natural language processing and transfer learning

Regulatory documents play a significant role in securing engineering project quality, standard process management and long-term sustainable developments. With the digitisation of knowledge in the AEC industry, the demand for automated knowledge mining has emerged when confronted with substantial regulations. However, the current interpretation approaches for regulatory documents are still mostly labour-intensive and flawed in complex knowledge. Based on transfer learning (BERT) and natural language processing (e.g., NLP-Syntactic Parsing), this paper proposes a fully automated knowledge mining framework to convert complex knowledge in textual regulations to graph-based knowledge representations. The framework uses a BERT-based engine to extract clauses from regulation documents through fine-tuning with the self-developed domain dataset. A constituent extractor is developed to process the provisions with complex knowledge and extract constituents. A knowledge modelling engine integrates the extracted constituents into a graph-based regulation knowledge model, which can be queried, visualised, and directly applied to downstream applications. The outcome has demonstrated promising performance in complex knowledge mining and knowledge graph modelling based on ISO 19650 case study. This research can effectively convert textual regulation documents to their counterpart regulatory knowledge base, contributing to automated knowledge acquisition and multi-domain knowledge fusion toward regulation digitalization.

Design of Multiple Ontology Based Agro Knowledge Mining Model

Farming is regarded as a major industry in India, accounting for 17% of the country's GDP growth. Agriculture employs 60% of the population hence it is considered an important sector in India. The important factors for agriculture are pest management, disease prevention, irrigation management, soil mineral composition, crop management, location, and the season in which the crop is grown. Hence all this information along with the techniques are well known only by the experienced farmers. Hence it is important to create an agro knowledge management system. As a result, this work makes an attempt to develop a multiple ontology-based agro knowledge management system. The designed system consists of agriculture information related to attributes of soil mineral, moisture, season, location, crop type, and temperature. It consists of multiple ontologies such as soil ontology, crop ontology, location ontology, and crop season ontology to provide agronomy knowledge. Soil ontology is premeditated to classify the soil type in a hierarchical order while crop ontology classifies the crop type, location ontology classifies locations suitable for different crop types and finally, crop season ontology classifies the season that is suitable for different crops. A rule base is built to develop the knowledge base and to validate the truthfulness of the knowledge base. Visualization of a knowledge base is carried out for better understanding and decision-making.

Shape-Guided Dual Consistency Semi-Supervised Learning Framework for 3-D Medical Image Segmentation

Popular semi-supervised 3D medical image segmentation networks commonly suffer from two limitations: First, the geometry shape constraint of targets is frequently disregarded, leading to coarse segmentation results. Second, semi-supervision is only performed on the last layer of the decoder, resulting in the insufficient representation learning of 3D convolution neural network. To address these issues, we propose a shape-guided dual consistency semi-supervised learning (SDC-SSL) framework for 3D medical image segmentation. Indeed, the proposed framework has two dominating advantages. Initially, a geometry-aware shape constraint is presented and used to learn the shape representation, which converts the differences between two networks into an unsupervised loss and lets the framework learn the boundary distance information of targets in unlabeled challenging regions. Additionally, a deep-supervised knowledge transfer strategy is developed and employed by the proposed framework, which can upgrade the generalization ability of our framework without increasing any extra parameters and computation costs in the inference phase. Experimental results demonstrate that the proposed framework outperforms state-of-the-art methods on two challenging 3D medical image segmentation tasks due to effective geometry-aware shape constraint on unlabeled data and the strong ability of knowledge mining on labeled data. The code is available at: https://github.com/SUST-reynole/SDC-SSL.

Refocusing Indian education system in the perspective of ancient Indian education system

Ancient or Traditional Indian Education System has always been a topic of interest among researchers as well as general educators. The New Education Policy rolled out by Government of India in 2020 has also focused upon this ancient Indian educational pedagogy. This article is an attempt to highlight the relevance of reorienting our present educational structure towards the ancient educational system, which was considered to be the best system of imparting value based and morally structured education among common masses. This article elaborates the relevance of knowledge acquisition, knowledge discrimination and knowledge mining from the ancient education system point of view as well as it’s relevance in the present day education system. This article is an attempt to elaborate the philosophy of Ancient Indian education system through the insight of the knowledge and use of this knowledge which helps in restoring and restructuring of human moral values for society and nation building rather than just gaining it without a defined purpose. In this article, it has been emphasized that for rejuvenating and restructuring the present Indian Education System and provide a national moral value to the education system, we need to understand the outstanding characteristics of Indian Ancient philosophy of life, that revolved around the concept of Karma. And as such it is need of the hour to have a holistic approach in the Present Education System based on our Ancient Education System of India, which provides both physical as well as spiritual solitariness. This article has emphasized on the fact that in order to bring back ancient glory that existed in terms of knowledge acquisition, knowledge discrimination and knowledge mining in our ancient education system we need to re-orient the present education system with Ancient Indian Value based Education.

Constrained Density Peak Clustering

Clustering is a commonly used tool for discovering knowledge in data mining. Density peak clustering (DPC) has recently gained attention for its ability to detect clusters with various shapes and noise, using just one parameter. DPC has shown advantages over other methods, such as DBSCAN and K-means, but it struggles with datasets that have both high and low-density clusters. To overcome this limitation, the paper introduces a new semi-supervised DPC method that improves clustering results with a small set of constraints expressed as must-link and cannot-link. The proposed method combines constraints and a k-nearest neighbor graph to filter out peaks and find the center for each cluster. Constraints are also used to support label assignment during the clustering procedure. The efficacy of this method is demonstrated through experiments on well-known data sets from UCI and benchmarked against contemporary semi-supervised clustering techniques.

Las dimensiones ontoepistemológicas de la minería aurífera ancestral en Colombia: evidencias del Resguardo Indígena Cañamomo y Lomaprieta / The Onto-Epistemological Dimensions of Ancestral Gold Mining in Colombia: Evidence from the Cañamomo and Lomaprieta Indigenous Reservation

The emergence of non-dualist ontologies, in combination with a renewed interest in relational worldviews of indigenous peoples, has forged new analytical frameworks that seek to reevaluate worldviews and knowledge systems of marginalized peoples in the Global South. An important project to recover this “pluriverse” is the Epistemologies of the South framework, which is used in this article to analyze the geography of conflicts in the Cañamomo and Lomaprieta indigenous reservation in the department of Caldas, Colombia. Based on an analysis of qualitative interviews that were carried out in this indigenous reservation during the period 2018-2020, we argue that the struggle for ancestral gold mining is not only an ontological struggle for ancestral territory, but also an epistemological struggle in defense of ancestral mining knowledge and practices. These onto-epistemological dimensions serve as inputs for an emerging definition of ancestral gold mining. As such, the contribution made in this article has a dual purpose: on the one hand, it seeks to provide knowledge that stimulates debates from the Global South on ontological struggles and the construction of alternative territorialities in mining areas while, on the other hand, it contributes to current debates on ancestral mining in Colombia.

Self-supervised knowledge mining from unlabeled data for bearing fault diagnosis under limited annotations

Deep learning has become a popular approach for fault diagnosis due to its powerful feature extraction and adaptability. However, its reliance on extensive annotations poses challenges in real-world applications. To confront this issue, this article proposes the CLTrans, a contrastive learning-based knowledge transfer method for semi-supervised fault diagnosis. CLTrans utilizes large-scale unlabeled data to benefit downstream tasks by simply performing unsupervised similarity matching. A feature encoder pre-trained by CLTrans can extract discriminative representations of vibration signals and can efficiently adapt to various tasks, even with data under different distributions. Experimental results of inner-dataset and inter-dataset knowledge transfer demonstrate that CLTrans outperforms conventional deep learning and state-of-the-art semi-supervised fault diagnosis approaches in terms of accuracy and domain adaptability, especially under limited labels. The capability of unsupervised knowledge mining and transfer allows for reducing the burden of data collection and annotation.

Applied Techniques for Twitter Data Retrieval in an Urban Area: Insight for Trip Production Modeling

This paper presents methods of retrieving Twitter data, both streaming and archive data, using Application Programming Interfaces. Twitter data are a kind of Location Based Social Network Data that, nowadays, is emerging in transportation demand modeling. Data regarding the locations of trip makers represent the most crucial step in the modeling. No research article has specifically addressed this topic with an up-to-date method; hence, this paper aims to refresh methods for retrieving Twitter data that can capture relevant data. The method is unique as the data are gathered for trip production modeling in zonal urban areas. Python script programs were built for both data retrieving methods. The programs were run for streaming data from May 2020 to April 2021 and archive data from 2018. The data were collected within Serang City, which is the nearest provincial city to Jakarta, the capital of Indonesia. In order to gather streaming data with no loss, the program has been run with referencing on sub-district office coordinate locations. Retrieving the intended data produces 1,090,623 documents, of which 54,103 are geotagged data from 2495 users. The study concluded that streaming data produce more geolocation data, while historical data capture more Twitter user data with relatively very little geotagged data and greater textual data than the period covered in this research. Thus, both techniques of retrieving Twitter data for urban personal trip modeling are necessary. Obtaining sufficient data collection using data streaming retrieval resulted in the most effective data preprocessing. This research contributes to Location Based Social Network data mining knowledge, both geolocation and text mining, and is useful for insight into developing trip production modeling in passenger transportation demand modeling using Machine Learning. This study also aims to provide useful methods for transportation system researchers and data scientists in utilizing Location Based Social Network data.

FedRAD: Heterogeneous Federated Learning via Relational Adaptive Distillation.

As the development of the Internet of Things (IoT) continues, Federated Learning (FL) is gaining popularity as a distributed machine learning framework that does not compromise the data privacy of each participant. However, the data held by enterprises and factories in the IoT often have different distribution properties (Non-IID), leading to poor results in their federated learning. This problem causes clients to forget about global knowledge during their local training phase and then tends to slow convergence and degrades accuracy. In this work, we propose a method named FedRAD, which is based on relational knowledge distillation that further enhances the mining of high-quality global knowledge by local models from a higher-dimensional perspective during their local training phase to better retain global knowledge and avoid forgetting. At the same time, we devise an entropy-wise adaptive weights module (EWAW) to better regulate the proportion of loss in single-sample knowledge distillation versus relational knowledge distillation so that students can weigh losses based on predicted entropy and learn global knowledge more effectively. A series of experiments on CIFAR10 and CIFAR100 show that FedRAD has better performance in terms of convergence speed and classification accuracy compared to other advanced FL methods.

The epigenetic modifier lysine methyltransferase 2C is frequently mutated in gastric remnant carcinoma.

Gastric remnant carcinoma (GRC), which occurs in the stomach after partial gastrectomy, is a rare and aggressive form of gastric adenocarcinoma (GAC). Comprehensive profiling of genomic mutations in GRC could provide the basis for elucidating the origin and characteristics of this cancer. Herein, whole-exome sequencing (WES) was performed on 36 matched tumor-normal samples from patients with GRC and identified recurrent mutations in epigenetic modifiers, notably KMT2C, ARID1A, NSD1, and KMT2D, in 61.11% of cases. Mutational signature analysis revealed a low frequency of microsatellite instability (MSI) in GRC, which was further identified by MSIsensor, MSI-polymerase chain reaction, and immunohistochemistry analysis. Comparative analysis demonstrated that GRC had a distinct mutation spectrum compared to that of GAC in The Cancer Genome Atlas samples, with a significantly higher mutation rate of KMT2C. Targeted deep sequencing (Target-seq) of an additional 25 paired tumor-normal samples verified the high mutation frequency (48%) of KMT2C in GRC. KMT2C mutations correlated with poor overall survival in both WES and Target-seq cohorts and were independent prognosticators in GRC. In addition, KMT2C mutations were positively correlated with favorable outcomes in immune checkpoint inhibitor-treated pan-cancer patients and associated with higher intratumoral CD3+ , CD8+ tumor-infiltrating lymphocyte counts, and PD-L1expression in GRC samples (p = 0.018, 0.092, 0.047, 0.010, and 0.034, respectively). Our dataset provides a platform for information and knowledge mining of the genomic characteristics of GRC and helps to frame new therapeutic approaches for this disease.

Semantic automatic annotation method based on artificial intelligence for electric power internet of things

AbstractThe development of the power Internet of Things is currently underway, and a proposal for a semantic Internet of Things based on artificial intelligence algorithms is made to address the challenges in obtaining prior knowledge for heterogeneous data fusion, improving the real‐time performance of the ontology library, and enhancing the efficiency of manual labeling of instance object data in the power field. This proposal introduces an Automatic Semantic Annotation Method to provide an effective knowledge organization model for sensor systems. Data mining knowledge is utilized to drive ontology update and improvement, resulting in more accurate semantic annotation and enhanced machine understanding. Experimental results show that artificial intelligence algorithms can automatically extract concepts from sensory data and achieve automatic semantic annotation during ontology instantiation.

Constant Wideband Compressive Spectrum Sensing for Sparse and Nonsparse Signals With Low-Rank Matrix Recovering and Prior Knowledge Mining

In cognitive radios (CR), compressive spectrum sensing (CSS) is a promising technique to detect wideband spectrum holes with reduced sampling rate requirement of analog-to-digital converter (ADC). It is implied in compressive sensing (CS) that the minimum sampling rate is determined by the sparsity order of the observed wideband signal. Unfortunately, the sparsity order is previously unknown and dynamically varies. Meanwhile, CS technology fails to work efficiently for the nonsparse spectrum case. For example, the CSS performance degenerates significantly for low signal-to-noise ratio (SNR) scenarios since the observed signal becomes nonsparse in frequency domain. To overcome the above-mentioned challenges, we propose a novel CSS algorithm with temporally-correlated prior knowledge mining (TPKM-CSS) to detect spectrum holes without a prior sparsity order information. The proposed scheme can handle both the sparse and nonsparse spectrum cases. The proposed solution has three contributions: firstly, CR implements multiple parallel sampling simultaneously to acquire wireless spectrum signals, which have a low-rank spectrum matrix. Secondly, the temporal correlation between continuous CS data is further exploited to improve the estimation accuracy of the current spectrum. Finally, the theoretical performance of the proposed TPKM-CSS is evaluated, wherein the cumulative distribution function (CDF) of the spectrum reconstruction error is derived in the closed form. The simulation results show that the proposed scheme works well for both sparse and nonsparse spectrum scenarios. Compared with the CSS schemes that are based on low-rank matrix recovery (LMR-CSS), maximizing correlation (MC), iterative compressive filtering (ICF), or Bayesian CS (BCS), the proposed scheme yields the superior performance in terms of the probability of detection and the probability of false alarm, especially for nonsparse signals. Furthermore, the simulated statistical performance of the spectrum reconstruction error matches well the theoretical CDF curve.

Mapping Research Trends of Medications for Multidrug-Resistant Pulmonary Tuberculosis Based on the Co-Occurrence of Specific Semantic Types in the MeSH Tree: A Bibliometric and Visualization-Based Analysis of PubMed Literature (1966-2020).

Before the COVID-19 pandemic, tuberculosis is the leading cause of death from a single infectious agent worldwide for the past 30 years. Progress in the control of tuberculosis has been undermined by the emergence of multidrug-resistant tuberculosis. The aim of the study is to reveal the trends of research on medications for multidrug-resistant pulmonary tuberculosis (MDR-PTB) through a novel method of bibliometrics that co-occurs specific semantic Medical Subject Headings (MeSH). PubMed was used to identify the original publications related to medications for MDR-PTB. An R package for text mining of PubMed, pubMR, was adopted to extract data and construct the co-occurrence matrix-specific semantic types. Biclustering analysis of high-frequency MeSH term co-occurrence matrix was performed by gCLUTO. Scientific knowledge maps were constructed by VOSviewer to create overlay visualization and density visualization. Burst detection was performed by CiteSpace to identify the future research hotspots. Two hundred and eight substances (chemical, drug, protein) and 147 diseases related to MDR-PTB were extracted to form a specific semantic co-occurrence matrix. MeSH terms with frequency greater than or equal to six were selected to construct high-frequency co-occurrence matrix (42 × 20) of specific semantic types contains 42 substances and 20 diseases. Biclustering analysis divided the medications for MDR-PTB into five clusters and reflected the characteristics of drug composition. The overlay map indicated the average age gradients of 42 high-frequency drugs. Fifteen top keywords and 37 top terms with the strongest citation bursts were detected. This study evaluated the literatures related to MDR-PTB drug therapy, providing a co-occurrence matrix model based on the specific semantic types and a new attempt for text knowledge mining. Compared with the macro knowledge structure or hot spot analysis, this method may have a wider scope of application and a more in-depth degree of analysis.

Applications of cutting-edge artificial intelligence technologies in biomedical literature and document mining.

The biomedical literature is a vast and invaluable resource for biomedical research. Integrating knowledge from the literature with biomedical data can help biological studies and the clinical decision-making process. Efforts have been made to gather information from the biomedical literature and create biomedical knowledge bases, such as KEGG and Reactome. However, manual curation remains the primary method to retrieve accurate biomedical entities and relationships. Manual curation becomes increasingly challenging and costly as the volume of biomedical publications quickly grows. Fortunately, recent advancements in Artificial Intelligence (AI) technologies offer the potential to automate the process of curating, updating, and integrating knowledge from the literature. Herein, we highlight the AI capabilities to aid in mining knowledge and building the knowledge base from the biomedical literature.