Search

In this study, we explore the important role of graph embedding methods in extracting valuable insights from graph structures, specifically focusing on node classification tasks. It is important to know the structural and semantic features and connections within nodes in a graph to learn more detailed hidden patterns. Hence, we propose a hybrid architecture, BeComE (Bert-ComplEx Embedding Model)1, a novel framework that employs both semantic and structural features from social network structures extracted through label-aware embedding models to aid in node classification in social graphs. A Support Vector Machine (SVM) classifier receives these vector embeddings as input features for classification tasks on social graphs and networks. The evaluation shows that BeComE gives better accuracy and F1-score results on the 'Cora' and 'Citeseer' datasets. [ABSTRACT FROM AUTHOR]

Knowledge graphs (KGs) are gaining prominence for their efficacy in data integration and knowledge representation within Current Research Information Systems (CRIS) Systems. By employing a semantic data model to represent entities, attributes, and their relationships, they prove versatile across scientific applications. In the era of Science platformization, particularly within CRIS systems, KGs serve to amalgamate diverse data sources and formats, facilitating the creation of interconnected data models. This enables stakeholders to access comprehensive, consistent information pertinent to their endeavors. This paper examines the pivotal roles of KGs in current and future data integration within CRIS systems, emphasizing their contributions to scientific reasoning. While their benefits include flexible knowledge modeling, support for semantic queries, and interoperability with various data sources, they face systemic limitations, particularly in methodological and technological aspects, hindering classical scientific investigations. The paper underscores the necessity for novel approaches to address these limitations, offering insights, use cases, and best practices for implementing KGs in CRIS systems. This paves the way for research institutions and scientific organizations to enhance their data analytics capabilities and support scientific reasoning effectively. [ABSTRACT FROM AUTHOR]

Research.f i is a national service for collecting, integrating and disseminating information on research conducted in Finland. This includes information on publications and related research outputs (including in the arts), datasets, funding decisions and research infrastructures. Many other countries and regions also work on the development of CRISes. Sometimes features are developed before there is enough data to take full advantage of them. Sometimes you just wait for a critical mass of data to be available before starting as feature development. In this case it is a question of information on the various connections between different research objects. Research.f i has been somewhere in between, with 4 years since its initial release waiting for relationship data to come. You build it, and the data will come - has been a long wait. But PIDGraph to the rescue! This paper describes to piloting of PIDGraph scientific knowledge graph data in research.f i - to fill in the missing links. The PIDGraph is a collection of metadata about research objects, authors, publishers, institutions, and other research related items, all identified by persistent identifiers, and a series of systems to expose that metadata and make it available for analysis and integration with other systems. [ABSTRACT FROM AUTHOR]

This research proposes SoVisu+, a new system for managing research data in a decentralized way, building upon the foundation laid by the CRISalid consortium, a community-driven approach to research information management. Traditional, centralized systems have issues like fragmented data, unchecked and outdated information. SoVisu+ tackles these problems by giving researchers more control over their data and fostering collaboration to build a shared knowledge graph. Researchers benefit from features like self-archiving accompaniment, AI driven helpers and real-time profile views, while institutions gain access to more reliable and comprehensive data. SoVisu+ expects to employ cutting-edge technologies like Solid Pods and Linked Data to streamline data management and improve research evaluation practices. Overall, SoVisu+ aims to create a more transparent and collaborative research ecosystem that ultimately accelerates scientific progress. [ABSTRACT FROM AUTHOR]

The BrCris project aims to create a unified platform integrating scientific and technological data across Brazil, promoting Open Science and adhering to FAIR principles (Findable, Accessible, Interoperable, and Reusable). Developed by the Brazilian Institute of Information in Science and Technology (IBICT), BrCris addresses the challenge of disparate data sources by utilizing models, ontologies, and semantic data based on the VIVO ontology to ensure comprehensive integration and accessibility. The platform aggregates data on researchers, projects, institutions, patents, software, and research outputs into a single repository, enhancing interoperability and usability. Key features include a logical level for entity-relationship modeling based on CERIF, and a semantic level for data visualization as a knowledge graph. The initial focus on scientific publications has laid the foundation for broader integration, including technological innovation data through the BrCris-T extension. This project underscores the significance of fostering collaboration, knowledge exchange, and technological advancement within Brazil's research ecosystem, while ensuring data quality and representation through continuous research and validation efforts. Launched in July 2023, BrCris stands as a pivotal tool for research management, decision-making, and the democratization of scientific knowledge, contributing to a more transparent, accessible, and collaborative scientific community. [ABSTRACT FROM AUTHOR]

Knowledge graph based search is becoming increasingly close to users' reading habits, providing a more user-friendly interactive experience. Integrating knowledge graphs as auxiliary information into recommendation algorithms can model and integrate semantic information of learning resources, providing more diverse personalized recommendation results and better alleviating the problem of information overload. In order to solve the problem of readers searching for suitable digital book resources and improve reading efficiency, the first step is to study the construction of knowledge graphs, including core tasks such as knowledge extraction, knowledge representation, and knowledge storage; Then, a mathematical model is constructed, based on the optimized graph convolutional network layer and the quantification algorithm of knowledge point centrality, to construct a universal computational framework for attention mechanism; Finally, simulation experiments were conducted to compare and analyze different models through evaluation indicators, in order to verify the advantages of the model constructed in this paper. The experimental results show that the research method proposed in this paper is superior to conventional methods such as FM, CKE, DNN, NFM, CFKG, KGAT, and KGCN. [ABSTRACT FROM AUTHOR]

In the recent digitization era, capturing, representing, and understanding knowledge is essential in countless real-world scenarios. Knowledge graphs emerged as a powerful tool for representing information through an adequately interconnected and interpretable structure in such a context. Nevertheless, generating proper knowledge graphs usually requires significant manual effort and domain expertise, resulting in graphs often affected by human subjectivity, limited scalability, or inability to capture implicit knowledge or handle heterogeneity. This paper proposes an innovative zero-shot strategy tailored to uncover reliable knowledge from text leveraging the recent highly effective generative large language models, with a particular focus on the GPT-3.5 model. Our proposal aims to create a suitable knowledge graph or improve existing ones by discovering missing qualitative triples. To assess the effectiveness of our methodology, we performed experiments on domain-specific datasets, confirming its potential for scalable and versatile knowledge discovery. [ABSTRACT FROM AUTHOR]

In recent years, researchers were interested in learning knowledge graph (KG) embeddings because of their advantages in downstream tasks like link prediction, entity disambiguation, and entity Classification in KGs. The state-of-the-art KG embedding models are mainly implemented in a centralized environment, i.e., they are trained on a single machine. However, most of the real-world KGs are huge in size, and a single machine is not enough to store the whole KG and train embedding models in this context. There are a few frameworks for learning KG embeddings in distributed environments. In this paper, we propose a novel distributed framework, namely DKGR, for training geometric embedding models. Our framework uses the Ray platform as a distributed environment. We evaluate our framework for the link prediction task on real-world datasets. Our preliminary results show a good speed-up when training KG embedding models in our framework. They also show that DKGR is scalable to large-scale KGs without affecting the link prediction performance remarkably. [ABSTRACT FROM AUTHOR]

In the challenging context of Knowledge Graph (KG) construction from text, traditional approaches often rely on Open Information Extraction (OpenIE) pipelines. However, they are prone to generating many incorrect triplets. While domain specific Named Entity Recognition (NER) is commonly used to enhance the results, it compromises the domain independence and misses crucial triplets. To address these limitations, we introduce G-T2KG , a novel pipeline for KG construction that aims to preserve the domain independence while reducing incorrect triplets, thus offering a cost-effective solution without the need for domain-specific adaptations. Our pipeline utilizes state-of-the-art OpenIE combined with both a noun phrase-based cleaning and a LLMs based validation. It is evaluated using gold standards in two distinct domains (i.e., computer science and music) that we have constructed in the context of this study. On computer science corpus, the experimental results demonstrate a higher recall as compared to state-of-the-art approaches, and a higher precision notably increased by the integration of LLMs. Experiments on the music corpus show good performance, underscoring the versatility and effectiveness of G-T2KG in domain-independent KG construction. [ABSTRACT FROM AUTHOR]

The coronavirus disease (COVID-19) spread rampantly around the world at the beginning of 2020 before the governments of each country could prevent it by making decisions based on medical data analysis. With proper formalization, the terabytes of new textual data available online every day could have been used for the early description and detection of cases of this virus. Since then, the number of Event-Based Surveillance (EBS) applications has increased exponentially. These applications aim to mine channels of unstructured data to detect signs of possible public health events. However, one problem with such systems is the need for expert intervention to define which event will be captured, which relevant terms should be used in the search, and to analyze the events to modify the search procedure constantly. Another problem is that many of these applications do not consider both spatial and temporal characteristics. Addressing such limitations, this article presents a novel approach. We propose the biomedical domain specialization of the Core Propagation Phenomenon Ontology (PropaPhen) to capture spatiotemporal characteristics of the propagation of health-related phenomena. We also propose the Description-Detection-Framework (DDF), which leverages PropaPhen, UMLS, and OpenStreetMaps to detect new medical events automatically. Finally, we demonstrate a use case with experiments on extracts from online newspapers about COVID-19. The results show that DDF can be useful for detecting clusters of suspicious cases of possible emerging health-related phenomena. [ABSTRACT FROM AUTHOR]

Automatic ontology population involves identifying, extracting and integrating information from various sources to instantiate the classes and properties of an ontology, thereby building a domain Knowledge Graph (KG). In this paper, we compare two text-based ontology population techniques: KOnPoTe, a semantic approach based on textual and domain knowledge analysis, and a generative AI approach utilizing Claude, a Large Language Model (LLM). We present experiments conducted on two French sales advertisement domains: real estate and boats, and discuss the strengths and limitations of both approaches. [ABSTRACT FROM AUTHOR]

Hepatocellular carcinoma (HCC) is the most common form of liver cancer, presenting a significant challenge in diagnosis and treatment. Artificial intelligence (AI) combined with manually curated ontologies promises to improve tumor diagnosis, but further improvements to these ontologies are necessary to enhance accuracy. This study explores the potential of graph-based machine learning methods to improve current AI-based tumor diagnosis and deepen our understanding of liver cancer. We build a comprehensive knowledge graph tailored specifically for liver cancer research. Then, we leverage state-of-the-art link prediction algorithms to uncover hidden relationships within this data network. Our method shows efficiency in accurately identifying potential missing links within the knowledge space of liver cancer. This research proposes the integration of AI-based tumor diagnosis with refined ontologies to develop precision healthcare solutions in liver cancer. [ABSTRACT FROM AUTHOR]

The question of informed and voluntary consent emerges as a matter of significance in healthcare. Obtaining informed consent, encounters many obstacles coupled with systemic, clinician-related, and patient-related factors, demanding interventions at different levels. This paper introduces a novel approach to present personalized consent based on Large Language Models (LLMs). The personalization of information is displayed through the combination of the LLM with a knowledge graph. We focus in our approach on how the knowledge graph enhances and personalize content generation, allowing therefore the acquisition of informed consent. The paper focuses as well on aspects related to hyper-parameters of information retrieval that help giving better prompt to the LLM. Experiments have showcased intresting results in terms of personalization and information retrieval using metrics of Rouge, Faithfulness and Relevance. [ABSTRACT FROM AUTHOR]

The substantiation of the causal relationships behind its development is very important in identifying possible interventions and early treatment. Knowledge Graphs (KG) play a crucial role in the medical research domain by organizing data into interconnected structures that represent relationships between entities such as disease, treatments, and progressions. This paper shows a complete workflow that demonstrates the extraction of causal relationships from medical abstracts using a fine-tuned GPT-based model and the integration of these relationships into a KG. [ABSTRACT FROM AUTHOR]

Maintaining up-to-date knowledge graphs (KGs) is essential for enhancing the accuracy and relevance of artificial intelligence (AI) applications, especially with sensitive domains. Yet, major KGs are either manually maintained (e.g., Wikidata) or infrequently rebuilt (e.g., DBpedia & YAGO). Thus, they contain many outdated facts. The rise of Large Language Models (LLMs) reasoning and Augmented Retrieval Generation approaches (RAG) gives KGs an interface to other trusted sources. This paper introduces a methodology utilizing Large Language Models (LLMs) to validate and update KG facts automatically. In particular, we utilize LLM reasoning capabilities to determine potentially outdated facts. After that, we use RAG techniques to generate an accurate fix for the fact. Experimental results on several LLMs and real-world datasets demonstrate the ability of our approach to propose accurate fixes. In addition, our experiments highlight the efficacy of few-shot prompts over zero-shot prompts. [ABSTRACT FROM AUTHOR]

Developing Natural Language Processing (NLP) tools for the Arabic language and its dialects is very challenging. Named Entity Recognition (NER) is one of these challenges, which serves as the core component in many NLP systems such as information extraction, question answering, machine translation and knowledge graph building. This paper sheds light on applying diferent approaches for Arabic NER (Flat and Nested) using a large and rich Arabic NER corpus, Wojood dataset, which consists of about 550K tokens annotated with 21 entity types. First, we apply the Wojood base model, AraBERTv2, along with various other Arabic BERT models such as MARBERTv2, CaMelBert, mBert,..etc. Next, we utilize the Bi-Encoder Contrastive Learning (CL) approach, a framework developed by Microsoft, which maps candidate text spans and entity types into the same vector representation space. The primary challenge in this approach is distinguishing non-entity spans from entity mentions. This approach could achieve F1 score 91.25% for Flat and 91.40% for Nested NER. Additionally, for evaluating the predicted NER, we employ Few-Shot prompting on LLaMA, and GPT-3.5 using refined prompt-based strategy. Our findings reveal that LLaMA outperforms GPT3.5. [ABSTRACT FROM AUTHOR]

This paper primarily addresses the work requirements of a certain hydroelectric dam and explores a method for constructing an intelligent assistance system for maintenance in hydroelectric stations. The dam boasts a diverse array of equipment and is situated in a region characterized by complex terrain and structures, posing significant challenges to maintenance work across various disciplines. This study aims to integrate emerging technologies such as big data computation, data analysis, and knowledge graphs, alongside cutting-edge human-computer interaction theories. It seeks to develop advanced theories of maintenance assistance for hydroelectric stations, focusing on the intelligent decision-making and guidance in maintenance operations using digital technology. By achieving digitization and intelligence in maintenance management and workflow processes, this research endeavors to provide participants at every stage of maintenance operations with safe, efficient, and reliable support capabilities. Ultimately, it aims to elevate the level of digitization and intelligence in maintenance operations at hydroelectric stations to unprecedented heights, thereby enhancing the management efficiency of maintenance operations and further improving quality and efficiency. [ABSTRACT FROM AUTHOR]

Nested named entity recognition has been widely used in natural language processing, information extraction and other fields. However, the multiple boundaries of nested named entities make the recognition of single entity face great challenges. This article used the information in the knowledge graph to extract and annotate entities in the Chinese nested named entity recognition algorithm, and built a new named entity recognition model based on this. Experimental results showed that this article applied the knowledge graph to the Chinese nested named entity recognition task, and its accuracy could reach up to 95.3%, which has obvious advantages in complex structure processing, relationship extraction and entity links. [ABSTRACT FROM AUTHOR]

Modern society has accumulated great amounts of knowledge since human start learning and communicating. With knowledge's continuously thriving and getting refined, knowledge itself has developed into a complex system with a huge number of content and structure features. Together with the development of science and technology, the ways of people learning have been changed greatly. Under the pressure of high-paced life and workplace competition, more and more people are in demand of personalized learning paths, in order to achieve their learning target as fast as possible. A personized and scientific learning path is the key to effective learning grogram. However, studies on algorithms of optimal learning path discovery are not comprehensively developed, especially on its performance and application. Therefore, in this article, we propose a new algorithm, Portfolio ST, which is improved from Steiner Tree, for personalized and optimal learning path mining. At first, this paper constructs the knowledge graph of finance, then builds learner models by profiling learners' owned and aimed knowledge based on the knowledge network. At last, according to the knowledge structure of a person's learner model, the most personized and shortest learning path is calculated by the Portfolio ST. After the experiment compared with two baseline algorithms, the Portfolio ST algorithm is proved to have the best performance and scalability, by changing owned/aimed/overall knowledge respectively, which is also the best algorithm of learning path mining that can achieve the global optimal among these three algorithms. [ABSTRACT FROM AUTHOR]

In the process of constructing a knowledge graph for research and develop (R&D) projects, entity extraction for R&D project abstracts helps to quickly understand project content and promote the application and transformation of research results. This article adopts a scientific entity recognition model based on BERT-IDCNN-CRF (BERT combined with Iterative Dilated Convolutional Neural Network and Conditional Random Field) for the entity extraction task in the construction of knowledge graph for R&D projects. This model uses BERT to extract semantic representations from abstract texts, and then uses IDCNN to deeply capture local and global features in the text. Combined with CRF to learn the context transfer relationship between text labels. Finally it achieved the precision of 69.3%, the recall of 65.8%, and the F1-score of 67.5% in scientific entity recognition. On the basis of this model, this article proposes a knowledge graph construction scheme for R&D projects, and constructs a knowledge graph centered on R&D projects, which includes 12 kinds of entities and 13 kinds of relationships. [ABSTRACT FROM AUTHOR]

With the rapid acceleration of urbanization and the widespread adoption of social networks, A vast amount of spatiotemporal check-in data from social media users has been generated. Traditional relational databases have limitations in handling massive spatiotemporal data, whereas graph databases, with their efficiency in data querying and processing, offer a novel solution for the analysis of spatiotemporal data. The paper introduces a method for constructing a user check-in spatiotemporal knowledge graph based on the Neo4j graph database, utilizing a designed graph model to create a visual knowledge graph. By incorporating temporal and spatial characteristics into entities, we realize efficient management and complex query of spatiotemporal data of social media user check-ins. By adding sentiment attributes to relationships through sentiment analysis of check-in text, the application of check-in text is realized. Ultimately, employing Cypher to achieve knowledge querying, we analyze the hot spots of urban activities and categories. [ABSTRACT FROM AUTHOR]

Commodities play a pivotal role in the global economy, and their dramatic price fluctuations can affect national security and global economic stability. This paper focus on the field of commodities, utilizing the Web of Science and CNKI databases. Based on the CiteSpace knowledge graph visualization analysis tool, it conducts a bibliometric analysis of the Chinese and English literature on commodity risk published from 2004 to 2023, aiming to reveal the keywords, emerging trends, major research institutions and their collaborative networks, trends in the issuance of articles, and major contributing countries, and authors. This analysis provides insights into the academic landscape of commodity risk and helps to identify future research directions. [ABSTRACT FROM AUTHOR]

Based on data sources from Web of Science and CNKI, this study employs CiteSpace software to conduct a bibliometric analysis of literature from 2014 to 2024 related to innovation and industry chains in the salt lake chemical industry. The objective is to gain a deep understanding of the research progress and development trends in the integration of innovation and industry chains both internationally and domestically. Through keyword emergence analysis, keyword clustering, and timeline and evolutionary analyses, this study highlights the integration of innovation and industry chains, showcasing the structure, collaboration, current research status, and trends in the future research areas of the salt lake chemical industry. [ABSTRACT FROM AUTHOR]

Developing support technology for elderly people with dementia is an urgent issue. The purpose of this study was to propose a computational method for designing "micro happiness" by applying the concept of subjective experience to dementia. In this paper, subjective experiences that can positively affect the quality of life are referred to as micro happiness. The proposed method provides experience recommendations by evaluating both the similarity of life functions and the similarity of experiences among elderly people. To verify the effectiveness of the method, the authors created a database consisting of 1124 subjective experiences of 77 people with dementia using a knowledge graph. The authors confirmed that the prototype data service system could recommend plausible social events considering the life functions and past subjective experiences of a target person with dementia. From the evaluations, it was suggested that the proposed method could be used to present candidates for positive experiences to elderly people with dementia. [ABSTRACT FROM AUTHOR]

Recommendation systems address content delivery system information overload. Recent recommender systems include knowledge-aware deep recommender models. These models train user and item embeddings using Graph Neural Networks and operate on collaborative knowledge graphs that combine user-item interactions with item-entity relationships to deliver more item information. Certain models employ a two-step learning procedure wherein the primary objective is to generate recommendations while simultaneously training knowledge graph embeddings. By co-learning, the models are able to comprehend the structural connections that exist between entities and relations, while also improving user embeddings through the use of previous encounters. The objective of this research endeavour is to assess the efficacy of cutting-edge models through the integration of knowledge graph embedding co-learning and the manipulation of embedding techniques. Co-learning boosts quicker learning and enhances the precision of recommendations, according to empirical findings. When a KG embedding layer was incorporated into KGCN, the recall metric outperformed the initial model by as much as 37%. Compared to translational embedding techniques, RotatE decreased training duration by 18% and demonstrated a noteworthy 15% improvement in the recall metric. The study's outcome offer valuable guidance for enhancing recommendation accuracy and reducing training time for models incorporating Knowledge Graph (KG) embeddings. Furthermore, the study introduces a new application for complex-valued KG embeddings. [ABSTRACT FROM AUTHOR]

In the rapidly evolving landscape of artificial intelligence, there is an escalating demand for enhanced reasoning capabilities and robust representation of higher-order relationships. Despite their prowess in pattern identification and generalization, Neural networks often grapple with challenges linked to interpretability and the absence of explicit relational reasoning. Symbolic logic, on the other hand, excels in formal knowledge representation and inference yet remains limited in deriving knowledge directly from data. We introduce the Neuro-Symbolic Knowledge Hypergraphs (NeSyKHG) framework to bridge this gap. This innovative model synergistically merges the strengths of neural networks and symbolic logic, offering an optimized representation and reasoning system for higher-order relationships using knowledge hypergraphs. Our empirical evaluations on the Chinese Medical High-order Relational (CMHR) dataset revealed that NeSyKHG significantly outperforms several established baseline models. Specifically, when compared to models such as Hypergraph Neural Networks (HGNN), Hypergraph Convolutional Networks (HyperGCN), Logical Hypergraph Link Prediction (LHP), and Universal Graph Convolutional Networks (UniGCN), NeSyKHG achieved an F1-score of 0.845, an accuracy (ACC) of 0.947, and an area under the curve (AUC) of 0.965. These metrics underscore NeSyKHG's exceptional ability in higher-order relational learning and reasoning. [ABSTRACT FROM AUTHOR]

As the influence of Incremental Sheet Forming (ISF) grows in manufacturing sectors, so does the demand for precise and updated knowledge construction in this domain. In this research, we evaluate the capability of Large Language Models (LLMs) to capture domain-specific knowledge, using ISF as a case study. Recognizing common LLMs' limitations such as potential inaccuracies and outdated information reliance, we propose a comprehensive approach involving automated and adaptive knowledge extraction, enrichment, and integration into an ISF-specific dataset. We then fine-tune the LLMs for ISF-related text classification and prompt response tasks. Our results reveal a significant enhancement in LLMs' performance within the ISF domain, with a domain knowledge acquisition rate exceeding that of GPT-3.5 by 10.4%, achieved by the fine-tuned Alpaca-33B model. Additionally, we introduce a novel conversational prototype designed to refine the accuracy and relevance of LLMs in the ISF domain. Our findings will guide future efforts in downstream tasks such as ISF-domain knowledge graph construction and quality prediction. [ABSTRACT FROM AUTHOR]

Large enterprises usually operate with numerous autonomous data sources, each utilizing diverse storage schemes and underlying technological infrastructures. This can lead to the inability of data analysis at a high level and a violation of the integrity and reliability of data within the organization, hindering the adoption of high-quality decisions and further development of the company. To address this issue, we propose a methodology that represents heterogeneous organization data in a high-level space by combining them into single ontology represented by knowledge graph. The graph is constructed autonomously using semi-structured data from various sources. Our method identifies links between sources based on several similar metric between keys sets of each collection. Then document collections are combined into merged entities if collections' key sets are similar. The experiments demonstrate the applicability of the developed approach for constructing a knowledge graph based on partially structured data from various sources, performing with high F1 measure (0.8 on average) when compared to the expert markup. [ABSTRACT FROM AUTHOR]

Reviewing the literature is one of the key elements of scientific research that allows you to identify existing solutions and research niches. However, it can be difficult for researchers to find relevant scientific articles related to the research topic. A limited number of available sources of information, diverse ways of describing them, and a multitude of scientific publications mean that scientists often have to spend a lot of time and effort to find the information they need. For this reason, we propose a solution for text analysis and its presentation in the form of a graph visualization. In our research, we used Natural Language Processing (NLP) methods and word weighting measures such as TF and TF-IDF. In addition, knowledge graphs were used to present the results visually. The conducted research was based on the analysis of the content of scientific articles, which allowed to draw important conclusions related to the presentation of texts in graphic form. Experimental results also identified potential methods and suggestions for literature reviews in specific fields of science. The analysis of the methods used and the results obtained allowed for a better understanding of the potential of natural language processing and graph text representation in the analysis of scientific articles. [ABSTRACT FROM AUTHOR]

Since it is widely conveyed on the Web, tabular data is mostly organized in a table structure. Indeed, this data format is widely used in multiple scenarios on the web and even within data storage entities. In addition, tabular data is a source of information that deserves to be interpreted and exploited. In this framework, efforts to extract meaningful information from tabular data based on semantic approaches, such as an ontology or a knowledge graph, are commonly referred to as Semantic Table Interpretation (STI). In this paper, we present an interactive tool that deploys a mapping approach to overcome possible semantic gaps in tabular data against a Knowledge Graph. Indeed, the ultimate goal of this tool is to provide an application and adaptable framework that combines search and filtering services associated with text preprocessing techniques. The experimental evaluation was conducted under the SemTab challenge and yielded encouraging and promising results regarding its performance and rankings. [ABSTRACT FROM AUTHOR]

Forecasting international trade flows is crucial for understanding global economic activity and serving as critical economic indicators used by economists and policymakers with significant implications for respective countries' economic policies. However, this is a challenging task due to the complexity of the relationships between entities involved in international trade. While several approaches have been proposed, knowledge graph-based models have emerged as promising solutions for predicting international trade flows. In this paper, we introduce a synthetic triple-generation algorithm for enhancing downstream tasks in knowledge graph embeddings based on the graph complement. The algorithm identifies missing relationships between entities by exploiting the complement graph and generates high-quality synthetic triples that improve the accuracy of predictions. We validate the generated triples using several knowledge graphs embedding methods and computing metrics. To perform an initial result screening, an international trade scenario is explored, demonstrating the effectiveness of the proposed approach in enhancing the performance of knowledge graph-based models for predicting international trade flows. [ABSTRACT FROM AUTHOR]

This work is focused on the sailing domain, for which several innovative technologies are being adopted to improve sailing efficiency, performance, and safety. In this context a knowledge graph could be used, for example, to represent information about different types of boats, sailing techniques, maritime safety, or weather conditions. Although numerous construction methods or ready-to-go knowledge graphs have been proposed in many fields, the sailing domain still needs to be explored. As the most effective methods rely on domain-specific datasets, the absence of suitable and available sailing datasets is one of the main challenges. Although several Open Information Extraction (OpenIE) methods may generate relevant triplets (the elementary units composing a knowledge graph) from arbitrary text without any additional information about its topic, such methods usually generate many incorrect triplets. In this paper, we aim (i) to address the aforementioned problem by proposing an innovative method that combines in an improved and strengthened way different OpenIE tools to generate proper triplets from domain-specific sources and, in particular, (ii) to build and release a suitable dataset for the sailing domain. Results confirm that our proposal can maximize the extracted information and infer unique information irretrievable by the classical OpenIE tools and, furthermore, that the generated dataset is significantly valuable for the sailing scenario. [ABSTRACT FROM AUTHOR]

With increased data in various e-commerce domains or in different online streaming platforms like Amazon Prime, Netflix, etc, it has become more challenging to give a personalized recommendation to a user. In terms of the movie recommendation system, different multimodal information such as audio, video, and text are effectively fused together to tackle this problem. However, in real-world scenarios this information may not always be present, leading to the problem of missing modality. Existing works in Recommender Systems (RSs) may fail to recommend effectively if their model is not susceptible to missing modality. In this work, we propose a meta-learning model to deal with the missing modality problem. Our proposed model tries to reconstruct the missing features with the help of the existing modalities and once reconstructed, we pass them to novel trident architecture for solving the recommendation task. We have performed experiments on the newly created multimodal version of the MovieLens dataset. Additionally, we have incorporated a new modality where we extracted subtitles of the movies to increase the effectiveness of our missing modality model. [ABSTRACT FROM AUTHOR]

Knowledge graph (KG), which is often described by a set of triplets (head, relation, tail), has shown to be very useful for many downstream applications, but suffers from the issue of incomplete connections. Knowledge graph completion (KGC) is to predict the missing connections among entities. Driven by recent advances of BERT in extracting meaningful textual representations, some recent KGC methods have been proposed to leverage the associated texts of triplets to assist the completion. Under the BERT-based framework, we argue that if more meaningful representations can be learned from the texts, better performance can be expected. Following this framework, inspired by recent successes of contrastive learning, we propose to adapt it to the KGC task by introducing a novel dual-encoder architecture to produce semantic-similar positive pairs, which have been widely shown to be pivotal in inducing semantic-rich representations for contrastive learning. To further improve the performance, a hard negative sampling strategy is developed to train the model. Extensive experimental results on three public datasets show that the proposed techniques can improve the performance of existing KGC methods effectively. [ABSTRACT FROM AUTHOR]

Unintentional injuries are common among children below the age of 14 years. Although the concern about the need for injury prevention is increasing among caregivers, preventable injuries are still one of the leading causes of death among children aged 0-14 years. Hence, there is a need for an effective system that can empower caregivers, parents and nursery teachers in making better and faster decisions to protect the lives of children through situational design of safe environments. In this paper, as a new situational design tool, we propose a situation-aware system for unintentional injury prevention with the integration of image processing using an RGB-D camera and spatial knowledge graphs. This paper describes the identification of dangerous situations based on the results obtained from an R-Map analysis of a large amount of data on injury situations. This paper also describes an implementation of the system and the advantages that it can offer in understanding possible dangerous situations in indoor environments like classrooms or homes and providing alternative suggestions on what kind of preventive measures can be taken to minimize such injuries. The feasibility of the proposed system was proved by applying it into nursery injury situational data. [ABSTRACT FROM AUTHOR]

In this paper, we introduce the novel concept of "social features for process-oriented business objects" that extends existing research on the Social Network of Business Objects (SoNBO). In SoNBO, data is extracted from heterogenous source systems in the form of business objects and a user-specific knowledge graph is generated and used as a navigation structure in a web application (the so-called SoNBO Explorer). Up to now, the SoNBO Explorer has been limited to information retrieval and access, which is now enriched with interactive functionality, the so called "social features", that allow the user to add content and thus interact with the information objects. The solution builds on the idea of "social profiles" that are used to represent people in Enterprise Social Networks and extends this idea to business objects. People and objects become (equal) nodes in a knowledge graph on which the user can perform (social) functionality (e.g. following, liking, commenting, tagging, etc.). This way the idiosyncratic functionalities of human-centred collaboration software is applied to process-aware ERP systems. We present the formalisation of our concept in the Social Business Object Ontology (SoBOOnt). [ABSTRACT FROM AUTHOR]

An automated question-answering system allows students to learn as an integral part of digitized learning. This system responds to queries using text. We also include a knowledge graph, which significantly enhances the model's intrigue and improves learners' understanding. The features of knowledge entity extraction, information point evaluation and analysis, knowledge graph construction from unstructured text, and knowledge entity integration are all explored. The question-answering paradigm we suggest in this study uses knowledge graphs and BERT (Bidirectional Encoder Representations from Transformers) to provide diverse learners with quick feedback on the subject. In order to facilitate non-native learners' understanding, we also include an English to Hindi translation. As a result, access to and continued learning can be very beneficial for educators. [ABSTRACT FROM AUTHOR]

With the deepening of aging in China, the demand of intelligent care system for the elderly is increasing day by day. The rapid development of information technology and artificial intelligence technology have gradually promoted the transformation of traditional care services from artificial to intelligent. Based on knowledge graph technology, this study built a knowledge graph model for elderly chronic disease smart care. The specific process includes ontology design, knowledge extraction, knowledge graph construction and question answering system design. The uniqueness of the knowledge graph in this paper is that it uses specific information about the elderly and data from professional care books. It is well guaranteed in practicality and reliability.With the help of the knowledge map constructed in this paper, primary care staff can effectively query high-quality and customized chronic disease care knowledge. [ABSTRACT FROM AUTHOR]

This paper is based on the needs of nuclear power business, combined with big data processing, deep learning, natural language processing and other related technologies and algorithms to design an intelligent architecture. This architecture includes 1 system, 3 centers, 5 smart applications and N core algorithms. We can use it to meet the future business development needs and management needs of nuclear power generation. [ABSTRACT FROM AUTHOR]

Operation management, which has always been a hot topic in management research for seeking to increase the efficiency of the system using scientific methods. We use the Web of Science database to select the pertinent literature data from 2012 to 2022 with a focus on 10 significant international journals to perform the scoped literature on operation management in this paper. It then conducts a bibliometric analysis of operation management from two perspectives—main direction and development trend—using CiteSpace and VOSviewer. The analysis discovered that between 2012 and 2022, 22952 papers on operation management were published in significant international journals. Of these, 4398 were written by Chinese academics, making up 19.16% of the total. International Journal of Production Research, European Journal of Operational Research, and International Journal of Production Economics were the main venues for Chinese academics' publications. The National Natural Science Foundation of China supported 3446 papers in these journals, or 15.01% of all papers published. A variety of research hotspots are revealed by keyword analysis. The research's findings provide an excellent framework for describing the traits and developments in operation management. [ABSTRACT FROM AUTHOR]

To improve the self-learning ability and information extension ability of existing artificial intelligence, we need a lot of new intelligent theory and methods to support it urgently. This paper combines artificial intelligence technology with Extenics, puts forward a new operator of Extension intelligence, analyzes its realization path and steps based on basic-element base built from multi-source information and knowledge graph. From the Internet and other open domain information we can extract approximately complete set of data, information and knowledge, to establish a Basic-element base in order to support the intelligent extension of information through mapping, extension, transformation and other algorithms. The development and application of the prototype system preliminary verified the effectiveness of the extension intelligent base model and finally we prospect the research directions for the further implementation of the operator. [ABSTRACT FROM AUTHOR]

Current medical artificial intelligence applications and products are confronted the dilemma of lacking standardized practical evaluation guidelines and management methods. This research adopts the Donabedian medical quality management classic model and the DeLone & Mclean benefit evaluation model to develop a basic framework of the medical artificial intelligence application evaluation system for auxiliary diagnosis and treatment and to design the corresponding evaluation procedures. This paper illustrates the overall project framework, followed by a detailed structure and construction process of the model proposed for fast and accurate medical evaluations, NHCKG. NHCKG provides a holistic view of diseases, medical regulations and evaluations. The proposed NHCKG can provide partial solutions to significant challenges faced by knowledge graphs and natural language processing (NLP). Based on the knowledge graph, the structure of a related medical question answering system is elaborated along with its advantages and disadvantages. The evaluation process can take this procedure as a blueprint for exploring standardized and practical evaluation of medical artificial intelligence applications. The process aims to build a concrete measurement basis for medical artificial intelligence products and to promote the healthy and stable development of the medical artificial intelligence industry. [ABSTRACT FROM AUTHOR]

Modelling customer behaviour is extremely important in areas such as banking, allowing for improvement in the quality of customer service. Embedding methods have proven adequate for predicting subsequent customer interactions in several areas. However, there is little research on banking data, and it is also unclear which factors most influence the model's decisions. At the same time, knowledge graph embeddings are gaining more and more popularity, explicitly taking into account various types of relations in the data. In this study, we consider the problem of predicting the store of a customer's next purchase based on historical transactional data. The results show that knowledge graph-based embeddings better account for repetitive customer behaviour, while traditional matrix factorization-based methods better predict previously unobserved triplets. The paper also presents exciting discoveries regarding the interpretation of representations. [ABSTRACT FROM AUTHOR]

The BrCris project has as its main goal to provide a unified view of the Brazilian research ecosystem. The LA Referencia platform is the main resource supporting the project, allowing us to treat heterogeneous data gathered from distinct sources and export the resulting datasets to different tools for data visualization, search, and analysis. We also use the VIVO platform for visualizing the data as a knowledge graph, which builds upon a semantic model based on the VIVO ontology. To make the data flow from the LA Referencia relational database to VIVO, we developed an Extraction, Transformation, and Load tool, which integrates both platforms ensuring data synchronization and making the resulting graph available for easy navigation and search [ABSTRACT FROM AUTHOR]

Automation systems are increasingly being used in dynamic and various operating conditions. With higher flexibility demands, they need to promptly respond to surrounding dynamic changes by adapting their operation. Context information collected during runtime can be useful to enhance the system's adaptability. Context-aware systems represent a design paradigm for modeling and applying context in various applications such as decision-making. In order to address context for automation systems, a state-of-the-art assessment of existing approaches is necessary. Thus, the objective of this work is to provide an overview on the design and applications of context and context models for automation systems. A systematic literature review has been conducted, the results of which are represented as a knowledge graph. [ABSTRACT FROM AUTHOR]

The number of spatial datasets available online has increased exponentially in recent years. Therefore, the search for spatial datasets is becoming a fourishing research field. The use of knowledge graphs has become rampant in search engines and in information retrieval. In this article, we identify the main resources needed and those missing to allow a knowledge graph to support spatial dataset search. We then apply our approach to the water domain in France by building a dedicated knowledge graph and describe an evaluation method to measure its effectiveness. [ABSTRACT FROM AUTHOR]

Systematic Literature Reviews aim at investigating current approaches to conclude a research gap or determine a futuristic approach. They represent a significant part of a research activity, from which new concepts stem. However, with the massive availability of publications at a rapid growing rate, especially digitally, it becomes challenging to efficiently screen and assess relevant publications. Another challenge is the continuous assessment of related work over a long period of time and the consequent need for a continuous update, which can be a time-consuming task. Knowledge graphs model entities in a connected manner and enable new insights using different reasoning and analysis methods. The objective of this work is to present an approach to partially automate the conduction of a Systematic Literature Review as well as classify and visualize the results as a knowledge graph. The designed software prototype was used for the conduction of a review on context-awareness in automation systems with considerably accurate results compared to a manual conduction. [ABSTRACT FROM AUTHOR]

In this article, we present Kepler- aSI, a matching approach to overcome possible semantic gaps in tabular data by referring to a Knowledge Graph. The task proves difficult for the machines, which requires extra effort to deploy the cognitive ability in the matching methods. Indeed, the ultimate goal of our new method is to implement a fast and efficient approach to annotate tabular data with features from a Knowledge Graph. The approach combines search and filter services combined with text pre-processing techniques. The experimental evaluation was conducted in the context of the SemTab 2021 challenge and yielded encouraging and promising results referring to its performance and ranks held. [ABSTRACT FROM AUTHOR]

The COVID-19 (SARS-CoV-2) spread around the globe could have been halted if we had had a better understanding of the situation and applied more restrictive measures for travel adapted to each country. This is due to a lack of efficient tools to visualize, analyze and control the virus dissemination. In the context of virus proliferation, analyzing flight connections between countries and COVID-19 data seems helpful to understand spatial and temporal information about the virus and its possible spread. To manage these complex, massive, and heterogeneous data, we propose a methodology based on knowledge graphs models. Several analyses and visualization tools can be applied, and our results show that these knowledge graph models may be a promising way to study the dissemination of any virus. These graphs can also be easily enriched with additional information that could be useful in the future to analyze or predict other interesting indicators. [ABSTRACT FROM AUTHOR]

In this work, we evaluated research projects of French companies using Natural Language Processing. To this end, we designed a system able to estimate the probability of obtaining a research tax credit (CIR) for a project based on its technical description. This system is designed around two modules whose outputs are concatenated and fed to a fully-connected neural network that predicts the probability of success for the project. The first module uses the FastText algorithm and a Convolutional Neural Network to extract a Text Embedding vector. The second module uses an unsupervised knowledge graph extraction method and a Graph Neural Network to extract a Graph Embedding vector. The texts used as data in this study describe the research projects of companies and are written in French. Due to their high confidentiality, no similar examples exist in the literature. This data is provided by a partner consulting firm whose work consists in helping companies raise funds for their research projects. Since the methods in the literature were not effective in extracting the knowledge graphs used in the second module for our data, we present a new Knowledge Graph extraction using an unsupervised Named Entities Recognition, NER, as our contribution. [ABSTRACT FROM AUTHOR]