Semantic Entities Research Articles

Multi-Object Navigation using Potential Target Position Policy Function.

Visual object navigation is an essential task of embodied AI, which follows the user's demands to let the agent navigate to the goal objects. Previous methods often focus on single object navigation. However, in real life, human demands are generally continuous and multiple, requiring the agent to implement multiple tasks in sequence. These demands can be addressed by repeatedly performing previous single task methods. However, dividing multiple tasks into several independent tasks to perform, without the global optimization between different tasks, the agents' trajectories may overlap, reducing the efficiency of navigation. In this paper, we propose an efficient reinforcement learning framework with a hybrid policy for multi-object navigation, aiming to maximally eliminate the noneffective actions. First, the visual observations are embedded to detect the semantic entities (such as objects). And the detected objects are memorized and projected into semantic maps, which can also be regarded as a long-term memory of the observed environment. Then a hybrid policy consisting of exploration and long-term planning strategies is proposed to predict the potential target position. In particular, when the target is directly oriented, the policy function makes long-term planning to the target based on the semantic map, which is implemented by a sequence of motion actions. In the alternative, when the target is not oriented, the policy function estimates an object potential position towards exploring the most possible objects (positions) that have close relations to the target. The relation between different objects is obtained with prior knowledge, which is used to predict the potential target position by integrating with the memorized semantic map. And then a path to the potential target is planned by the policy function. We evaluate our proposed method on two large-scale 3D realistic environment datasets, Gibson and Matterport3D, and the experimental results demonstrate the effectiveness and generalization of the proposed method.

Rel Topic : A graph-based semantic relatedness measure in topic ontologies and its applicability for topic labeling of old press articles

Graph-based semantic measures have been used to solve problems in several domains. They tend to compare semantic entities in order to estimate their similarity or relatedness. While semantic similarity is applicable to hierarchies or taxonomies, semantic relatedness is adapted to ontologies. In this work, we propose a novel semantic relatedness measure, named Rel Topic , within topic ontologies for topic labeling purposes. In contrast to traditional measures, which are dependent on textual resources, Rel Topic considers semantic properties of entities in ontologies. Thus, correlations of nodes and weights of nodes and edges are assessed. The pertinence of Rel Topic is evaluated for topic labeling of old press articles. For this purpose, a topic ontology representing the articles, named Topic-OPA, is derived from open knowledge graphs by applying a SPARQL-based automatic approach. A use-case is presented in the context of the old French newspaper Le Matin. The generated topics are evaluated using a dual evaluation approach with the help of human annotators. Our approach shows an agreement quite close to that shown by humans. The entire approach’s reuse is demonstrated for labeling a different context of articles, recent (modern) newspapers.

Across-language masculinity of oceans and femininity of guitars: Exploring grammatical gender universalities.

This is the first cross-language study to reveal nouns with invariable masculine or feminine grammatical gender assignments in nine gendered languages from different groups of one linguistic family. It evidences that many cases of gender universality have semantic motivation-an entity's grammatical gender correlates with either traditional masculine/feminine connotations, or cultural and symbolic implications. The study's findings also testify thematic preferences: most masculine grammatical gender universalities are found for the nouns denoting artifacts, whereas most feminine universalities are identified for abstract concepts. The apparent existence of grammatical gender universalities has a cognitive significance. From a psycholinguistic perspective, grammatical gender is viewed as a built-in personification pattern for speakers' mental representations. This research presents cross-linguistic constants in conceptualizing the natural kinds, artifacts, and abstract concepts denoted by the considered nouns, as "male" or "female".

Semantic Search System for Common Cultural Traditions of the Greater Mekong Subregion

This article presents the development of a semantic search system for the common cultural traditions of the Greater Mekong Subregion, which has been studied using common cultural tradition data from Thailand, Laos, and Cambodia. Our system presents the implementation of semantic search to solve the problem of semantic gaps in common cultural characteristic concepts for information access and retrieval. The field of cultural heritage management represents a growing area of research that contains valuable information for users, researchers, and domain experts. Finding information on traditions or cultures of interest among unstructured materials requires a laborious search process. In this paper, we demonstrate how to extract semantic entities and relationships represented in the information using the theory of information science, specifically the Knowledge Organization (KO) approach, which involves determining the concepts and semantic relations of a particular issue and establishing their relations with relevant concepts. It serves as a foundation for knowledge structure development and ontology development, which contribute to effective information searching and knowledge access. The system was developed based on systematic ontology using the Hozo Ontology Editor and Database and ontology mapping using the Ontology Application Management (OAM) framework, which helps distinguish between relevant and irrelevant data, thereby leading to search effectiveness. To cut down on the time wasted in browsing and searching, reduce associated user frustration, and support the research of domain experts, more selective user access is needed.

Knowledge Fusion Analysis and Research Based on Knowledge Graph

With the rapid popularization of Internet technology, it is a new challenge to mine the hidden knowledge behind the data in an organized and directed way. Knowledge graph not only describes the content of semantic associations and entities existing in the objective world, but also uses graph structure to visually present the structured knowledge system for system users. Therefore, in the development of modern technology, knowledge graph has been widely concerned by industry and academia. In the era of big data, in order to build high-quality knowledge graph, knowledge fusion is very critical, among which entity alignment and entity link are important parts of knowledge fusion task. Starting from the field of film and television, this paper uses the entity alignment technology of NovEA model and the candidate entity ranking structure based on CNN-DSSM to deeply discuss the key technologies of knowledge fusion during the construction of multi-level film and television knowledge graph. The final experimental results prove that, compared with other models of the same task, the NovEA model studied in this paper has higher alignment accuracy, and the candidate entity ranking structure based on CNN-DSSM is better designed.

Knowledge Fusion Analysis and Research Based on Knowledge Graph

With the rapid popularization of Internet technology, it is a new challenge to mine the hidden knowledge behind the data in an organized and directed way. Knowledge graph not only describes the content of semantic associations and entities existing in the objective world, but also uses graph structure to visually present the structured knowledge system for system users. Therefore, in the development of modern technology, knowledge graph has been widely concerned by industry and academia. In the era of big data, in order to build high-quality knowledge graph, knowledge fusion is very critical, among which entity alignment and entity link are important parts of knowledge fusion task. Starting from the field of film and television, this paper uses the entity alignment technology of NovEA model and the candidate entity ranking structure based on CNN-DSSM to deeply discuss the key technologies of knowledge fusion during the construction of multi-level film and television knowledge graph. The final experimental results prove that, compared with other models of the same task, the NovEA model studied in this paper has higher alignment accuracy, and the candidate entity ranking structure based on CNN-DSSM is better designed.

Discovery of topic evolution path and semantic relationship based on patent entity representation

PurposeThis paper aims to propose a method for better discovering topic evolution path and semantic relationship from the perspective of patent entity extraction and semantic representation. On the one hand, this paper identifies entities that have the same semantics but different expressions for accurate topic evolution path discovery. On the other hand, this paper reveals semantic relationships of topic evolution for better understanding what leads to topic evolution.Design/methodology/approachFirstly, a Bi-LSTM-CRF (bidirectional long short-term memory with conditional random field) model is designed for patent entity extraction and a representation learning method is constructed for patent entity representation. Secondly, a method based on knowledge outflow and inflow is proposed for discovering topic evolution path, by identifying and computing semantic common entities among topics. Finally, multiple semantic relationships among patent entities are pre-designed according to a specific domain, and then the semantic relationship among topics is identified through the proportion of different types of semantic relationships belonging to each topic.FindingsIn the field of UAV (unmanned aerial vehicle), this method identifies semantic common entities which have the same semantics but different expressions. In addition, this method better discovers topic evolution paths by comparison with a traditional method. Finally, this method identifies different semantic relationships among topics, which gives a detailed description for understanding and interpretation of topic evolution. These results prove that the proposed method is effective and useful. Simultaneously, this method is a preliminary study and still needs to be further investigated on other datasets using multiple emerging deep learning methods.Originality/valueThis work provides a new perspective for topic evolution analysis by considering semantic representation of patent entities. The authors design a method for discovering topic evolution paths by considering knowledge flow computed by semantic common entities, which can be easily extended to other patent mining-related tasks. This work is the first attempt to reveal semantic relationships among topics for a precise and detailed description of topic evolution.

Parametric Interval Temporal Logic over Infinite Words

Model checking for Halpern and Shoham's interval temporal logic HS has been recently investigated in a systematic way, and it is known to be decidable under three distinct semantics. Here, we focus on the trace-based semantics, where the infinite execution paths (traces) of the given (finite) Kripke structure are the main semantic entities. In this setting, each finite infix of a trace is interpreted as an interval, and a proposition holds over an interval if and only if it holds over each component state (homogeneity assumption). In this paper, we introduce a quantitative extension of HS over traces, called parametric HS (PHS). The novel logic allows to express parametric timing constraints on the duration (length) of the intervals. We show that checking the existence of a parameter valuation for which a Kripke structure satisfies a PHS formula (model checking), or a PHS formula admits a trace as a model under the homogeneity assumption (satisfiability) is decidable. Moreover, we identify a fragment of PHS which subsumes parametric LTL and for which model checking and satisfiability are shown to be EXPSPACE-complete.

RETRACTED: Transformer-induced graph reasoning for multimodal semantic segmentation in remote sensing

As a large amount of earth observation data is available on a global scale, it becomes possible to apply multimodal semantic segmentation technology to remote sensing scene analysis. However, the diversity of objects in large-scale scenes and the cross-modal gap between different images are still challenging in practical applications. To address these problems, we propose a Transformer-Induced Hierarchical Graph Network (GraFNet) for multimodal semantic segmentation in remote sensing scenes, which promotes the exploration of potential intra- and inter-modal relations by introducing a new modeling paradigm. Different from existing methods, GraFNet parses multimodal remote sensing images into semantic topological graphs, and exploits the structural information of land cover categories to learn joint representations. Specifically, an attentive heterogeneous information aggregation mechanism is presented to parse diverse objects in remote sensing scenes into semantic entities, and capture modality-specific object–object interaction patterns in a topology-aware environment. In addition, modality hierarchical dependency modeling is introduced to encode the interactive representation of cross-modal objects, and distinguish the modality-specific contribution to improve cross-modal compatibility. Extensive experiments on several multimodal remote sensing datasets demonstrate that the proposed GraFNet outperforms the state-of-the-art approaches, achieving F 1 /mIoU accuracy 91.1%/82.4% on the ISPRS Vaihingen dataset, 93.4%/88.4% on ISPRS Potsdam dataset, and 91.8%/84.0% on the MSAW dataset.

Digital Philological Practices: the Project “Chekhov Digital”

The article examines the role of “digital practices” in the formation of interdisciplinary humanitarian knowledge, the peculiarities of the development of digital humanitarian projects in the field of philology. The model of development of a digital project in the field of preparation of semantic markup of literary publications “Chekhov Digital”, which is a digital publication of the academic Сomplete works and letters of A. Chekhov, is considered. The goal of the project is to develop machine-readable (semantic) markup of the writer’s texts based on the standards of digital publication Text Encoding Initiative (TEI). Within the framework of the project, standards for the preparation of digital Russian-language publications are being clarified, conceptual and technical conditions for implementation are being formulated, and infrastructure and new research methods are being developed. The structure of machine-readable annotation of documents has been developed, which enables marking up semantic entities in Chekhov’s texts, notes and comments for building semantic search within the corpus of the writer’s texts. To clarify the markup of semantic entities in the works of A. Chekhov the methods of automatic text processing were used, including topic modeling and vector semantic models to analyze the most important author’s concepts in the texts; corpus methods for studying the contexts of the use of verbal representations of concepts. The conceptual analysis made it possible to reconstruct the author’s concepts in the context of the markup of semantic entities. To mark up the names of real people and objects, a special database, based on pointers to letters, has been created. The project implements the principle of Open data, one of the goals of which is to create scientific communities around data. The work on the project has led to the development of scientific cooperation between the Centers for Digital Humanities of the HSE and the SFedU.

Contextual Semantic-Guided Entity-Centric GCN for Relation Extraction

Relation extraction tasks aim to predict potential relations between entities in a target sentence. As entity mentions have ambiguity in sentences, some important contextual information can guide the semantic representation of entity mentions to improve the accuracy of relation extraction. However, most existing relation extraction models ignore the semantic guidance of contextual information to entity mentions and treat entity mentions in and the textual context of a sentence equally. This results in low-accuracy relation extractions. To address this problem, we propose a contextual semantic-guided entity-centric graph convolutional network (CEGCN) model that enables entity mentions to obtain semantic-guided contextual information for more accurate relational representations. This model develops a self-attention enhanced neural network to concentrate on the importance and relevance of different words to obtain semantic-guided contextual information. Then, we employ a dependency tree with entities as global nodes and add virtual edges to construct an entity-centric logical adjacency matrix (ELAM). This matrix can enable entities to aggregate the semantic-guided contextual information with a one-layer GCN calculation. The experimental results on the TACRED and SemEval-2010 Task 8 datasets show that our model can efficiently use semantic-guided contextual information to enrich semantic entity representations and outperform previous models.

Why was this cited? Explainable machine learning applied to COVID-19 research literature.

Multiple studies have investigated bibliometric factors predictive of the citation count a research article will receive. In this article, we go beyond bibliometric data by using a range of machine learning techniques to find patterns predictive of citation count using both article content and available metadata. As the input collection, we use the CORD-19 corpus containing research articles—mostly from biology and medicine—applicable to the COVID-19 crisis. Our study employs a combination of state-of-the-art machine learning techniques for text understanding, including embeddings-based language model BERT, several systems for detection and semantic expansion of entities: ConceptNet, Pubtator and ScispaCy. To interpret the resulting models, we use several explanation algorithms: random forest feature importance, LIME, and Shapley values. We compare the performance and comprehensibility of models obtained by “black-box” machine learning algorithms (neural networks and random forests) with models built with rule learning (CORELS, CBA), which are intrinsically explainable. Multiple rules were discovered, which referred to biomedical entities of potential interest. Of the rules with the highest lift measure, several rules pointed to dipeptidyl peptidase4 (DPP4), a known MERS-CoV receptor and a critical determinant of camel to human transmission of the camel coronavirus (MERS-CoV). Some other interesting patterns related to the type of animal investigated were found. Articles referring to bats and camels tend to draw citations, while articles referring to most other animal species related to coronavirus are lowly cited. Bat coronavirus is the only other virus from a non-human species in the betaB clade along with the SARS-CoV and SARS-CoV-2 viruses. MERS-CoV is in a sister betaC clade, also close to human SARS coronaviruses. Thus both species linked to high citation counts harbor coronaviruses which are more phylogenetically similar to human SARS viruses. On the other hand, feline (FIPV, FCOV) and canine coronaviruses (CCOV) are in the alpha coronavirus clade and more distant from the betaB clade with human SARS viruses. Other results include detection of apparent citation bias favouring authors with western sounding names. Equal performance of TF-IDF weights and binary word incidence matrix was observed, with the latter resulting in better interpretability. The best predictive performance was obtained with a “black-box” method—neural network. The rule-based models led to most insights, especially when coupled with text representation using semantic entity detection methods. Follow-up work should focus on the analysis of citation patterns in the context of phylogenetic trees, as well on patterns referring to DPP4, which is currently considered as a SARS-Cov-2 therapeutic target.

Retracted: QGMS: A query growth model for personalization and diversification of semantic search based on differential ontology semantics using artificial intelligence

AbstractThe inclusion of collective intelligence through a semantic focused affective computing can incorporate intelligence to web search and ensure its compliance with the Web 3.0. In this article, a query growth model with inclusive and exclusive ontology semantics has been proposed for diversification of query recommendation in semantic search. The ontology semantics include query augmented ontology generation, agent‐driven attractor‐distractor generation to yield a merged ontology, and endowment of merged ontology by using hybridization of a series of knowledge bases. The strategy further includes the formulation of a semantic network and entity leveraging based on description logics (DLs) to improve the quality of query recommendation. A novel hierarchical entropy cognitive similarity covariance model has been proposed for yielding the most appropriate recommendable query words. The strategy also encompasses the user‐click information for capturing the current user intents to improve the quality queries recommended in semantic search, and thereby incorporate personalization. Experimentations are conducted for the CHiC dataset and the Spring 2006 Query Log dataset and an average accuracy of 96.27% and 92.01%, respectively, with a very low false discovery rate of 0.06 and 0.1 for the respective datasets.

A novel semantic representation approach for web documents using deep entity linking and multidocument support

SummaryRepresenting semantic web information in the form of existing formats like extensible mark‐up language (XML), resource description framework (RDF) or web ontology language (OWL) is limited by the inability to maintain intradocument relationships between different semantic entities. Moreover, these standards are built for generic data representation and thus do not have inherent support for the semantic description of the document and, consequently, do not support features like multidocument linkages and context‐sensitivity. In order to remove these drawbacks, a semantic web‐specific document representation approach, namely, deep semantic XML (DS‐XML) using deep linking and machine learning, is presented in this paper. The DS‐XML approach facilitates the use of semantics with the help of a novel document structure that focuses on defining any web document in a semantic form. It includes ontologies and linkage between different vocabularies, which assists in achieving semantic interoperability. In order to make this approach future proof, a novel deep linking approach is integrated at the document level, which allows DS‐XML to understand more about the contents of the document and represent information in an aggregated manner. The deep linking layer is combined with the machine learning‐based classification layer that allows the document information to be represented in an interactive manner. The proposed DS‐XML approach is tested on different document retrieval applications, including a few Internet of Things (IoT) applications, and the representation performance is found to be superior to the existing semantic approach (XML, RDF and OWL).

Decomposing and Recomposing Event Structure

AbstractWe present an event structure classification empirically derived from inferential properties annotated on sentence- and document-level Universal Decompositional Semantics (UDS) graphs. We induce this classification jointly with semantic role, entity, and event-event relation classifications using a document-level generative model structured by these graphs. To support this induction, we augment existing annotations found in the UDS1.0 dataset, which covers the entirety of the English Web Treebank, with an array of inferential properties capturing fine-grained aspects of the temporal and aspectual structure of events. The resulting dataset (available at decomp.io) is the largest annotation of event structure and (partial) event coreference to date.

A knowledge model-based BIM framework for automatic code-compliant quantity take-off

The results of quantity take-off (QTO) based on building information modeling (BIM) technology rely heavily on the geometry and semantics of 3D objects that may vary among BIM model creation methods. Furthermore, conventional BIM models do not contain all the required information for automatic QTO and the results do not follow the descriptive rules in the standard method of measurement (SMM). This paper presents a new knowledge model-based framework that incorporates the semantic information and SMM rules in BIM for automatic code-compliant QTO. It begins with domain knowledge modeling, taking into consideration QTO-related information, semantic QTO entities and relationships, and SMM logic formulation. Subsequently, linguistic-based approaches are developed to automatically audit the BIM model integrity for QTO purposes, with QTO algorithms developed and used in a case study for demonstration. The results indicate that the proposed new framework automatically identifies the semantic errors in BIM models and obtains code-compliant quantities.

Semantic meaning modulates object importance in human fixation prediction

Humans tend to allocate attention to semantic entities. Objects are important in fixation selection, but not all the objects are equally attractive. In this paper, we introduce the concept of attribute bias to characterize the influence of semantic attributes compared with low-level saliency on fixation distribution. Two different ways are adopted to get two sets of semantic attributes. In both cases, most semantic attributes have a positive influence on drawing attention and contribute more than low-level saliency in object areas. We also find that attribute bias is robust to low-level saliency and can consistently reflect the relative attractiveness of objects with different semantic attributes. It is demonstrated that such bias helps make better fixation predictions by distinguishing the importance of objects, although low-level saliency models with better performance are less dramatically improved by attribute bias. These findings indicate the role of conceptual meaning as opposed to features in visual attention.

A Literature Review on Centering Theory

Centering, a model of the conversants’ center of attention in discourse, centers around the relationship between attentional state, inferential complexity and the form of referring expression. Concerned with the local coherence and semantic entity salience, Centering Theory is framed by two strands of early work: (i) research by Joshi, Kuhn and Weinstein (Joshi and Kuhn, 1979; Joshi, Weinstein, 1981); and (ii) research by Grosz and Sidner (Grosz, 1977; Sidner, 1979; Grosz, Sidner, 1986). A synthesis of the two trains of thoughts helps to make up a theoretical background for Centering Theory and a motivation for the future work, such as empirical studies and application studies. In the present article, in order to help the reader to understand the epochal significance of various fruits of recent couple of decades, an attempt to undertake a systematic review of Centering Theory will be made.

Semantic-Based and Entity-Resolution Fusion to Enhance Quality of Big RDF Data

Within an organisation, the quality in big data is a cornerstone to operational, transactional processes and to the reliability of business analytics for decision making. In fact, as organizations are harnessing multi-sources data to rise the benefits of their business, the quality of data becomes important and crucial. This paper presents a new approach to query big data sources using Resource Description Framework (RDF) representation to ensure data quality by harvesting more relevant and complete query results. Our approach handles two important types of heterogeneity over multiple data sources: semantic heterogeneity and URI-based entity identification. It proposes (1) a semantic entity resolution method based on inference mechanism using rules to manage the misunderstanding of data, in real world entities (2) Data Quality enhancement using MapReduce-based query rewriting approach includes the entity resolution results to infer and adds implicit data into query results (3) a parallel combination of MapReduce jobs of saturation and query rewriting inferences to handle transitive and cyclic rules for a richer rules' expression language (4) experiments to assess the efficiency of the proposed approach over real big RDF data originating from insurance and synthetic data sets.

A Semantic Parsing and Reasoning-Based Approach to Knowledge Base Question Answering

Knowledge Base Question Answering (KBQA) is a task where existing techniques have faced significant challenges, such as the need for complex question understanding, reasoning, and large training datasets. In this work, we demonstrate Deep Thinking Question Answering (DTQA), a semantic parsing and reasoning-based KBQA system. DTQA (1) integrates multiple, reusable modules that are trained specifically for their individual tasks (e.g. semantic parsing, entity linking, and relationship linking), eliminating the need for end-to-end KBQA training data; (2) leverages semantic parsing and a reasoner for improved question understanding. DTQA is a system of systems that achieves state-of-the-art performance on two popular KBQA datasets.