Description Of Entities Research Articles

Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation

Injecting textual information into knowledge graph (KG) entity representations has been a worthwhile expedition in terms of improving performance in KG oriented tasks within the NLP community. External knowledge often adopted to enhance KG embeddings ranges from semantically rich lexical dependency parsed features to a set of relevant key words to entire text descriptions supplied from an external corpus such as wikipedia and many more. Despite the gains this innovation (Text-enhanced KG embeddings) has made, the proposal in this work suggests that it can be improved even further. Instead of using a single text description (which would not sufficiently represent an entity because of the inherent lexical ambiguity of text), we propose a multi-task framework that jointly selects a set of text descriptions relevant to KG entities as well as align or augment KG embeddings with text descriptions. Different from prior work that plugs formal entity descriptions declared in knowledge bases, this framework leverages a retriever model to selectively identify richer or highly relevant text descriptions to use in augmenting entities. Furthermore, the framework treats the number of descriptions to use in augmentation process as a parameter, which allows the flexibility of enumerating across several numbers before identifying an appropriate number. Experiment results for Link Prediction demonstrate a 5.5% and 3.5% percentage increase in the Mean Reciprocal Rank (MRR) and Hits@10 scores respectively, in comparison to text-enhanced knowledge graph augmentation methods using traditional CNNs.

Multi-view semantic enhancement model for few-shot knowledge graph completion

In recent years, few-shot knowledge graph completion (FKGC) has gained popularity as a solution to the long-tail distribution problem of real-world knowledge graphs (KGs). The previous knowledge graph completion (KGC) models obtain triple representation merely relying on the structure view, ignoring the valuable semantic knowledge. In this paper, we propose a multi-view framework for few-shot relation learning to address the issue. Specifically, based the structural information of the graph obtained using the structure view, we add the text view and the commonsense view. The text view employs text descriptions of entities and relations to obtain a richer semantic representation. The commonsense view performs high-quality negative sampling based on complex relations. Moreover, commonsense semantic constraints are invoked to suppress the overfitting caused by the complexity of the relation matrix. Extensive experiments show that our model outperforms state-of-the-art FKGC methods on the frequently-used benchmark datasets FB15k237-One and NELL-One.

Diagnostic des sarcomes utérins et tumeurs mésenchymateuses utérines rares à potentiel de malignité. Référentiels du Groupe Sarcome Français et des Tumeurs Rares Gynécologiques

The landscape of uterine sarcomas is becoming more complex with the description of new entities associated with recurrent driver molecular alterations. Uterine sarcomas, in analogy with soft tissue sarcomas, are distinguished into complex genomic and simple genomic sarcomas. Leiomyosarcomas and undifferentiated uterine sarcomas belong to complex genomic sarcomas group. Low-grade and high-grade endometrial stromal sarcomas, other rare tumors associated with fusion transcripts (such as NTRK, PDGFB, ALK, RET ROS1) and SMARCA4-deficient uterine sarcoma are considered simple genomic sarcomas. The most common uterine sarcoma are first leiomyosarcoma and secondly endometrial stromal sarcomas. Three different histological subtypes of leiomyosarcoma (fusiform, myxoid, epithelioid) are identified, myxoid and epithelioid leiomyosarcoma being more aggressive than fusiform leiomyosarcoma. The distinction between low-grade and high-grade endometrial stromal sarcoma is primarily morphological and immunohistochemical and the detection of fusion transcripts can help the diagnosis. Uterine PEComa is a rare tumor, which is distinguished into borderline and malignant, according to a risk assessment algorithm. Embryonal rhabdomyosarcoma of the uterine cervix is more common in children but can also occur in adult women. Embryonal rhabdomyosarcoma of the uterine cervix is almost always DICER1 mutated, unlike that of the vagina which is wild-type DICER1, and adenosarcoma which can be DICER1 mutated but with less frequency. Among the emerging entities, sarcomas associated with fusion transcripts involving the NTRK, ALK, PDGFB genes benefit from targeted therapy. The integration of molecular data with histology and clinical data allows better identification of uterine sarcomas in order to better treat them.

The Balinese Lontar Manuscript Metadata Model: An Ontology-Based Approach

The digitization of cultural heritage, particularly lontar manuscripts, is a focus of research to aid stakeholders in the management of lontar metadata. Managing the metadata of Balinese Lontar manuscripts through the application of ontology is one method for ensuring the preservation and accessibility of Balinese Lontar manuscripts. This research aims to apply ontology technology to the preservation of the cultural heritage of Balinese lontar manuscripts so that all information about lontar details can be categorized according to their respective properties and relevant information can be presented based on the user's preferences. This research method employs the stages of the Design Science Research Methodology (DSRM) when devising the ontology for the Balinese lontar manuscript. The results demonstrated that the construction of metadata using an ontology-based methodology can provide the information required to describe, categorize, and connect ontology entities. Metadata consists of entity descriptions, hierarchies and classifications, relationships and properties, as well as the necessary semantics for constructing effective ontologies

Contrastive Multi-Modal Knowledge Graph Representation Learning

Representation learning of knowledge graphs (KGs) aims to embed both entities and relations as vectors in a continuous low-dimensional space, which has facilitated various applications such as link prediction and entity retrieval. Most existing KG embedding methods focus on modeling the structured fact triples independently and ignore the multi-type relations among triples as well as the variety of data types (e.g., texts and images) associated with entities in KGs, and thus fail to capture the complex and multi-modal information that is inherently inside the entity-relation triples. In this paper, we propose a novel approach for knowledge graph embedding named Contrastive Multi-modal Graph Neural Network (CMGNN), which can encapsulate comprehensive features from multi-modal content descriptions of entities and high-order connectivity structures. Specifically, CMGNN first learns entity embeddings from multi-modal content and then contrasts encodings from multi-relational local neighbors and high-order connectivities to obtain latent representations of entities and relations simultaneously. Experimental results demonstrate that CMGNN can effectively model the multi-modalities and multi-type structures in KGs, and significantly outperforms existing state-of-the-art methods on benchmark datasets for the tasks of link prediction and entity classification.

Cryptic species Hydatigera kamiyai and other taeniid metacestodes in the populations of small mammals in Serbia

BackgroundHydatigera (Cestoda: Taeniidae) is a recently resurrected genus with the description of a new species, Hydatigera kamiyai, a cryptic entity within the Hydatigera taeniaeformis species complex. Rodents are intermediate hosts and correct taxonomic identification of H. taeniaeformis sensu lato (s.l.) species is difficult without the use of molecular methods. The aim of this study was to identify and explore the genetic diversity of Hydatigera and other taeniid species.MethodsTen different small mammals species (856 individuals) (Rattus rattus, three Apodemus, three Arvicolinae and three Soricidae species) were examined from 2013 to 2023. Captured animals were visually examined for cysts and visible lesions. Two markers were used for amplification and sequencing: cox1 and 12S rDNA.ResultsMolecular analysis of cysts and visible lesions revealed four taeniid species: Hydatigera kamiyai, H. taeniaeformis sensu stricto (s.s.), Taenia martis and T. crassiceps. Hydatigera kamiyai was found in Apodemus flavicollis, A. agrarius, Microtus arvalis and Crocidrua leucodon, while H. taeniaeformis s.s. is registered in R. rattus. Hydatigera kamiyai cox1 sequences clustered with European populations and showed at least 25 nucleotid differences compared to Asian, African, Australian and one of our isolates of H. taeniaeformis s.s acquired from a rat, followed by large sequence distances (9.4% to 12.9%), indicating clear molecular distinction of two species.ConclusionsThis is one of the few mitochondrial gene-based studies performed after the description of cryptic entities within the Hydatigera taeniaeformis s.l. complex and represents a valuable contribution to understanding of genetic diversity, host suitability and geographic distribution of these tapeworm species. Also, our study provides an important basis of molecular data from this part of Europe for further studies. We emphasize the importance of additional studies of intermediate hosts, especially rats from Europe and Apodemus spp. and voles from Asia and Africa.Graphical abstract

Incorporating logic rules with textual representations for interpretable knowledge graph reasoning

Reasoning on knowledge graphs (KGs) is significant for downstream applications, such as question answering and information extraction. On the basis of using factual triples in KGs, learning logic rules has been explored to handle KG reasoning in an interpretable pattern. Some rule learning methods simply apply thresholds to reduce the exponential complexity caused by the vast candidate rules derived from KGs. Others are designed to prune candidates on the basis of embedding facts in the KG. However, these methods only consider the factual triples, which mainly contain structural information, and ignore sufficient semantic information in real-world KGs. This situation weakens the methods in searching for high-quality rules. To this end, we propose a joint method by incorporating logic rules with textual representations for interpretable KG reasoning, named LoTus. Firstly, LoTus obtains the structural embeddings with only factual triples, and the textual embeddings with descriptions of entities, which are later integrated into the representations of entities and relations. Secondly, top candidate rules are selected by a proposed joint pruning strategy incorporating structural and textual representations in KGs. Finally, LoTus captures high-quality rules and removes low-quality ones by a filtering process with metrics of rules. It reduces the time complexity on KG reasoning by learning rules and provides interpretability simultaneously. Extensive experiments on different datasets show that LoTus obtains competitive performance comparing to baselines on KG reasoning metrics Hits@1, Hits@10 and mean reciprocal rank (MRR), and achieves significant superiority among the recent rule learning methods.

Enhancing text representations separately with entity descriptions

Several studies have focused on incorporating language models with entity descriptions to facilitate the model with a better understanding of knowledge. Existing methods usually either integrate descriptions in the pre-training stage by designing description-related tasks, or in the fine-tuning stage by directly appending description strings to the original input, this paper falls into the latter group. We separate entity descriptions from the original text and process them by another lighter module. Specifically, we use the original large model to encode the original input, while the lighter module processes the entity descriptions. We also propose a layer-wise fusion strategy to deeply couple the representations of the input and descriptions. To further improve the fusion of the two representations, we explore two auxiliary tasks: the entity-description enhancement task and the entity contrastive task. Experiments on (Open Entity, FIGER, FewRel, TACRED, SST) datasets yield respective improvements of (0.9, 1.4, 0.6, 0.5, 0.3). Utilizing ChatGPT as the description embedding method holds the potential for even more promising results.

Entity Linking based on RoFormer-Sim for Chinese Short Texts

Entity linking is an important means to identify named entities in text and a key technology for constructing knowledge graphs, playing an important role in fields such as intelligent question answering and information retrieval. However, existing entity linking methods for short texts have low accuracy due to the lack of rich contextual information, informal expression, and incomplete grammar structures. Therefore, this paper proposes a short-text entity linking model based on the RoFormer-Sim pre-training model. Firstly, entity context features are extracted by the RoFormer-Sim pre-training model, and then text similarity calculation and sorting are performed with candidate entity description texts to obtain the corresponding entity in the knowledge base with the disambiguated entity. The experimental results show that the RoFormer-Sim model can provide prior knowledge for entity linking, and the proposed model in this paper has an F1 value of 0.8851, which is better than other entity linking models based on other pre-training models.

Formal topological description of three-dimensional topological entities based on k-dimensional pseudo-manifold

This paper clarifies the necessity of studying three-dimensional spatial data models, analyzes the research progress and existing problems of topological spatial relationship description methods, and proposes a complete and formal description framework of three-dimensional topological spatial relations based on point set topological theory and dimension expansion method. On this basis, the topological spatial relations existing in three-dimensional spatial objects are classified, and five fundamental topological spatial relations are defined. The mutual exclusivity and completeness of the minimal set of 3D topological space relations are proved. Based on point set topology and k-dimensional pseudo-manifold, this paper proposes the formal description of three-dimensional topological properties and spatial relations between spatial entities. The simplicial homology is applied to manifolds, the topological properties of three-dimensional space entities are revealed, and the formal description method of three-dimensional space entities is given through the geometric structural relationship between simplex and simplicial k-complex.

Proceedings of the North American Society of Head and Neck Pathology Companion Meeting, New Orleans, LA, March 12, 2023: Classification of Salivary Gland Tumors: Remaining Controversial Issues?

The salivary gland section in the 5th edition of the World Health Organization Classification of Head and Neck Tumours includes a description of several new entities. In addition, numerous tumor variants were described and new concepts proposed, most of which have been based on recent molecular discoveries. However, there are still some controversial issues that remain to be resolved, and some of them are discussed in this review.

JKRL: Joint Knowledge Representation Learning of Text Description and Knowledge Graph

The purpose of knowledge representation learning is to learn the vector representation of research objects projected by a matrix in low-dimensional vector space and explore the relationship between embedded objects in low-dimensional space. However, most methods only consider the triple structure in the knowledge graph and ignore the additional information related to the triple, especially the text description information. In this paper, we propose a knowledge graph representation model with a symmetric architecture called Joint Knowledge Representation Learning of Text Description and Knowledge Graph (JKRL), which models the entity description and relationship description of the triple structure for joint representation learning of knowledge and balances the contribution of the triple structure and text description in the process of vector learning. First, we adopt the TransE model to learn the structural vector representations of entities and relations, and then use a CNN model to encode the entity description to obtain the text representation of the entity. To semantically encode the relation descriptions, we designed an Attention-Bi-LSTM text encoder, which introduces an attention mechanism into the Bi-LSTM model to calculate the semantic relevance between each word in the sentence and different relations. In addition, we also introduce position features into word features in order to better encode word order information. Finally, we define a joint evaluation function to learn the joint representation of structural and textual representations. The experiments show that compared with the baseline methods, our model achieves the best performance on both Mean Rank and Hits@10 metrics. The accuracy of the triple classification task on the FB15K dataset reached 93.2%.

Im2Graph: A Weakly Supervised Approach for Generating Holistic Scene Graphs from Regional Dependencies

Conceptual representations of images involving descriptions of entities and their relations are often represented using scene graphs. Such scene graphs can express relational concepts by using sets of triplets ⟨subject—predicate—object⟩. Instead of building dedicated models for scene graph generation, our model tends to extract the latent relational information implicitly encoded in image captioning models. We explored dependency parsing to build grammatically sound parse trees from captions. We used detection algorithms for the region propositions to generate dense region-based concept graphs. These were optimally combined using the approximate sub-graph isomorphism to create holistic concept graphs for images. The major advantages of this approach are threefold. Firstly, the proposed graph generation module is completely rule-based and, hence, adheres to the principles of explainable artificial intelligence. Secondly, graph generation can be used as plug-and-play along with any region proposition and caption generation framework. Finally, our results showed that we could generate rich concept graphs without explicit graph-based supervision.

Types and kinds of etymological exercises in Russian vocabulary

This paper considers educational and school etymological analysis (etymological parsing) as a type of Russian language exercises. The objective that was set and achieved involved developing a typology (a classification) of etymological exercises corresponding to the main subject competences. Based on the ‘presence/absence of reference supports’ ctiterion, we classify etymological exercises into those performed a) with and b) without reference supports, i. e. from memory and intuition. Known and new etymological exercises are divided into three types, namely speech, language, and complex. Speech (textual) tasks split into two types: perception and reproduction of ready-made etymological messages and preparation of new etymological messages using one or several reference supports. Language exercises include classical and game-based vocabulary exercises with partial etymological analysis. Complex exercises are treated as a complete structural analysis (morphemic, derivational) with an etymological follow-up. The essence of the etymological tasks is revealed through typical didactic material, assignments, and sample works. The taxonomic and entity description of etymological exercises creates prerequisites for their continuous and sys- tematic use in teaching Russian in all comprehensive school forms.

Anatomy of Janu Sandhi w.s.r. to Knee joint: An Ayurveda Overview

The knowledge of anatomy in Ayurveda has been poorly understood in today’s era. The lack of sufficient literatures regarding the subject has hindered its development. Despite the hurdles, still the concept of anatomy in Ayurveda is in depth and seeking for scientific exploration. The increasing incidence of joint diseases leading towards chronicity are best being treated through Ayurveda, but the concepts behind the results are not well understood due to lack of proper knowledge in anatomy. The description of various anatomical entities has been clearly mentioned in Samhitas in Ayurveda. Sushruta Samhita, the pioneer text of Ayurveda surgery, in its Shareer Sthana primarily has described about the human body anatomy. Apart from this, Charaka Samhita, the text of medicine has also lightened regarding the anatomy. In this article, the anatomical concepts of Janu Sandhi (knee joint) are extracted from Sushruta Samhita and Charaka Samhita in order to organize the knowledge systematically. The concepts are correlated with the view of contemporary medical science and illustrated in best possible way.

Cluneal neuropathy: Background, diagnosis, and treatment.

Cluneal neuropathy is encompassed by three distinct clinical entities. Superior, middle, and inferior cluneal neuralgia make up the constellation of symptoms associated with cluneal neuropathy. Each has its own variable anatomy. We compiled a narrative review including a review of available literature. We conducted searches on PubMed/MEDLINE, Embase and Google Scholar on the topics of cluneal neuropathy and treatment therein. We collected source articles regarding original descriptions of the disease entities in addition to articles focused on treatment. Adjusted incidence rates of superior cluneal neuropathy are 1.6%-11.7%. Accurate diagnosis remains challenging due to the lack of standardized criteria and the aforementioned variability. Treatment may include therapeutic nerve blocks, ablative techniques, neuromodulation, and surgical decompression. Gaps including those related to true incidence and work up exist. Outcomes from interventional studies are limited and mixed due to significant population heterogeneity and non-standardized treatment approaches coupled with very small sample sizes.

Knowledge Graph Completion Based on Entity Descriptions in Hyperbolic Space

Hyperbolic space has received extensive attention because it can accurately and concisely represent hierarchical data. Currently, for knowledge graph completion tasks, the introduction of exogenous information of entities can enrich the knowledge representation of entities, but there is a problem that entities have different levels under different relations, and the embeddings of different entities in Euclidean space often requires high dimensional space to distinguish. Therefore, in order to solve the above problem, we propose a method that use entity descriptions to complete the knowledge graph in the Poincaré ball model of hyperbolic space. In this method, the text representation of entity descriptions is in Euclidean space and mapped to hyperbolic space through exponential map. Next, the triple embeddings are initialized in hyperbolic space, and the structured representation of the triple is trained by the hyperbolic model. Finally, the text representation and the structured representation of the entity are cross-fused in hyperbolic space, and then the balance factors are used to adjust the unbalanced energy function. Experimental results show that, compared with baseline models, the proposed method can improve the performance of knowledge graphs completion.

Text-based automatic personality prediction using KGrAt-Net: a knowledge graph attention network classifier

Nowadays, a tremendous amount of human communications occur on Internet-based communication infrastructures, like social networks, email, forums, organizational communication platforms, etc. Indeed, the automatic prediction or assessment of individuals’ personalities through their written or exchanged text would be advantageous to ameliorate their relationships. To this end, this paper aims to propose KGrAt-Net, which is a Knowledge Graph Attention Network text classifier. For the first time, it applies the knowledge graph attention network to perform Automatic Personality Prediction (APP), according to the Big Five personality traits. After performing some preprocessing activities, it first tries to acquire a knowing-full representation of the knowledge behind the concepts in the input text by building its equivalent knowledge graph. A knowledge graph collects interlinked descriptions of concepts, entities, and relationships in a machine-readable form. Practically, it provides a machine-readable cognitive understanding of concepts and semantic relationships among them. Then, applying the attention mechanism, it attempts to pay attention to the most relevant parts of the graph to predict the personality traits of the input text. We used 2467 essays from the Essays Dataset. The results demonstrated that KGrAt-Net considerably improved personality prediction accuracies (up to 70.26% on average). Furthermore, KGrAt-Net also uses knowledge graph embedding to enrich the classification, which makes it even more accurate (on average, 72.41%) in APP.

Usage of formalized knowledge about source code for refactoring actions in Swift

The object of research in the paper is the source code of a software product written in the Swift programming language. Most programs as a product of a certain project during the life cycle require changes and modifications, which is costly or impossible to implement in the case of poor code quality. One of the options for solving the problem of poor code quality is the timely application of refactoring principles. The existing problem is that implementation of high-level refactoring must be done manually by the developer without the use of automated tools as built-in solutions cannot fulfill the need due to the architectural complexity of the product. To reduce the number of errors made during refactoring, to simplify the process of performing routine actions, it is suggested to use a new software product for refactoring. It works with high-level user commands based on a formalized description of the source code together with a knowledge base containing a description of code entities and their properties (what specific actions can be performed with them). In the work, the refactoring of the source code was carried out using the example of the Swift programming language. The proposed approach of component architecture (knowledge base, software engine) further allows to expand the functionality of the software product to other programming languages. The work was directed to the development of a prototype of a software product using the proposed approach to check and compare the results with other refactoring tools. A command line utility has been developed that accepts a verbal command as an input and outputs the results of processing and analysis of the source code (search for complex structures in the code) or applies the proposed change. As a result of the conducted testing, it was established that the use of the proposed approach allows performing complex refactoring tasks with the help of a simple verbal formalized command. Accomplishing the same task using only the built-in refactoring tools requires significantly more time and effort or is impossible at all.

Knowledge graph completion based on parsing graph embedding and a weighted graph convolutional network

Knowledge graph completion is an important research issue in knowledge graph construction, knowledge engineering, and natural language processing. A knowledge graph is a knowledge support for realizing accurate knowledge services in general and professional fields. It is also an important breakthrough foundation in information retrieval, question-and-answer interactions, and information recommendation. The low quality and small scale of the knowledge graph are the main bottlenecks that hinder its wide applications. The purpose of knowledge graph completion is to build a large-scale and high-quality knowledge graph for continuously updating and expanding the knowledge graph. Aiming at the difficulty of knowledge graph completion methods to extract deep semantic features from auxiliary information, such as unstructured texts, this study proposes a knowledge graph completion method based on parsing graph embedding and weighted graph convolutional network. This method uses the weighted graph convolutional network to model the semantic dependency parsing of the entity description and construct the semantic dependency parsing graph embedding. Furthermore, it introduces a multi-grained sentence-embedding generation method of the entity description, which is intended to build entity representation learning that can capture multi-grained semantics and deep-level semantic features. The experimental results on two public datasets show that the proposed knowledge graph completion approach outperforms the existing methods, thereby demonstrating its effectiveness and superiority.