Semantic Representation Model Research Articles

Perceptual model for semantic representation of design futures practice cases: Towards the development of a design futures repository

Various media and forms of cases or works have promoted the development and dissemination of "Design Futures", making empirical research focusing on practical cases not only valuable but also urgently needed. While heuristic case libraries enhance creativity and efficiency in traditional design, a resource library for Design Futures has not been established. To fill this gap, our preliminary work includes clarifying the perception of cases by different practitioners through semantic representation to establish a perceptual model. In this paper, we propose the typology of Design Futures practice cases. Using desktop research, expert selection, and empirical methods, we constructed a semantic perceptual space and further extracted it into four categories. Combining multivariate statistical techniques and related visualization methods, we examined the distinctions and relationships between semantic words, dimensions, cases, and their types. We have finally established the perceptual model for semantic representation of design futures practice cases, with 4 main perception dimensions (critical, wellbeing, conceptual, and technological) and 13 key semantics, and explored in depth the implications these semantics may contain individually. We further explored the potential of the model in developing a case library. In summary, our research findings provide a framework for representing, retrieving, and evaluating practice cases and support the development of a Design Futures Repository.

PAMR: Persian Abstract Meaning Representation Corpus

One of the most used and well-known semantic representation models is Abstract Meaning Representation (AMR). This representation has had numerous applications in natural language processing tasks in recent years. Currently, for English and Chinese languages, large annotated corpora are available. In addition, in some low-resource languages, related corpora have been generated with less size; although, until now, to the best of our knowledge, there is not any AMR corpus for the Persian/Farsi language. Therefore, the aim of this article is to create a Persian AMR (PAMR) corpus via translating English sentences and adjusting AMR guidelines and to solve the various challenges that are faced in this regard. The result of this research is a corpus, containing 1,020 Persian sentences and their related AMR that can be used in various natural language processing tasks. In this article, to investigate the feasibility of using the corpus, we have applied it to two natural language processing tasks: Sentiment Analysis and Text Summarization.

Webly Supervised Knowledge-Embedded Model for Visual Reasoning.

Visual reasoning between visual images and natural language remains a long-standing challenge in computer vision. Conventional deep supervision methods target at finding answers to the questions relying on the datasets containing only a limited amount of images with textual ground-truth descriptions. Facing learning with limited labels, it is natural to expect to constitute a larger scale dataset consisting of several million visual data annotated with texts, but this approach is extremely time-intensive and laborious. Knowledge-based works usually treat knowledge graphs (KGs) as static flattened tables for searching the answer, but fail to take advantage of the dynamic update of KGs. To overcome these deficiencies, we propose a Webly supervised knowledge-embedded model for the task of visual reasoning. On the one hand, vitalized by the overwhelming successful Webly supervised learning, we make much use readily available images from the Web with their weakly annotated texts for an effective representation. On the other hand, we design a knowledge-embedded model, including the dynamically updated interaction mechanism between semantic representation models and KGs. Experimental results on two benchmark datasets demonstrate that our proposed model significantly achieves the most outstanding performance compared with other state-of-the-art approaches for the task of visual reasoning.

Translation of Metaphorical Information in Japanese Literature Combined with Text Mining Techniques

Abstract Japanese literature is renowned for its conciseness and the profound, multi-layered meanings conveyed through its economical use of language and rich symbolic content. This paper explores the creative impact of these linguistic features in Japanese literature by analyzing the deep meanings and varied classifications of literary metaphors. The construction of a Japanese literature corpus involved the collection, cleaning, and organization of Japanese literary texts, followed by the selection of specific linguistic features. The corpus was then aligned using a dynamic programming approximation algorithm. A semantic representation model of the Japanese literature text corpus was developed by integrating the Attention Word Embedding Model with the Extended Topic Information Model, and metaphoric corpus features were fused using the static fusion strategy of Extended Topic Information and Attention Word Embedding. This study also examines the translation of symbolic information in Japanese literature, focusing on linguistic and stylistic features. The findings reveal that translations into Chinese tend to employ augmentation translation, resulting in 1,706 additional sentences compared to the original Japanese texts. Furthermore, the lexical density in the Chinese translation reference corpus is 2.45 percentage points higher than in the Japanese literature corpus. Analysis of verb modifications shows that among the 90 altered verbs in the Chinese translation corpus, only 24 have corresponding instances in the original Japanese texts. Text mining techniques were employed to conduct a deep analysis of symbolic content, enhancing the understanding of the nuanced connotations inherent in Japanese literature.

An Ontology-Based Framework for Geospatial Integration and Querying of Raster Data Cube Using Virtual Knowledge Graphs

The integration of the raster data cube alongside another form of geospatial data (e.g., vector data) raises considerable challenges when it comes to managing and representing it using knowledge graphs. Such integration can play an invaluable role in handling the heterogeneity of geospatial data and linking the raster data cube to semantic technology standards. Many recent approaches have been attempted to address this issue, but they often lack robust formal elaboration or solely concentrate on integrating raster data cubes without considering the inclusion of semantic spatial entities along with their spatial relationships. This may constitute a major shortcoming when it comes to performing advanced geospatial queries and semantically enriching geospatial models. In this paper, we propose a framework that can enable such semantic integration and advanced querying of raster data cubes based on the virtual knowledge graph (VKG) paradigm. This framework defines a semantic representation model for raster data cubes that extends the GeoSPARQL ontology. With such a model, we can combine the semantics of raster data cubes with features-based models that involve geometries as well as spatial and topological relationships. This could allow us to formulate spatiotemporal queries using SPARQL in a natural way by using ontological concepts at an appropriate level of abstraction. We propose an implementation of the proposed framework based on a VKG system architecture. In addition, we perform an experimental evaluation to compare our framework with other existing systems in terms of performance and scalability. Finally, we show the potential and the limitations of our implementation and we discuss several possible future works.

Criminal Action Graph: A semantic representation model of judgement documents for legal charge prediction

Semantic information in judgement documents has been an important source in Artificial Intelligence and Law. Sequential representation is the traditional structure for analyzing judgement documents and supporting the legal charge prediction task. The main problem is that it is not effective to represent the criminal semantic information. In this paper, to represent and verify the criminal semantic information such as multi-linked legal features, we propose a novel criminal semantic representation model, which constructs the Criminal Action Graph (CAG) by extracting criminal actions linked in two temporal relationships. Based on the CAG, a Graph Convolutional Network is also adopted as the predictor for legal charge prediction. We evaluate the validity of CAG on the confusing charges which composed of 32,000 judgement documents on five confusing charge sets. The CAG reaches about 88% accuracy averagely, more than 3% over the compared model. The experimental standard deviation also show the stability of our model, which is about 0.0032 on average, nearly 0. The results show the effectiveness of our model for representing and using the semantic information in judgement documents.

Enhanced semantic representation model for multisource point of interest attribute alignment

Multisource point of interest (POI) attribute alignment is the consistent processing of heterogeneous attribute values from different data sources pointing to the same POI data, which is one of the key technologies to achieve geospatial data fusion. However, semantic heterogeneity problems of synonyms and homographs among different POI data sources are encountered, which makes multisource POI data fusion challenging. This paper proposes a multisource POI attribute alignment method based on the Enhanced Semantic Representation Model (ESRM). First, the unlabeled corpus is preprocessed by Chinese word segmentation and attribute expression sequence construction. Then, the ESRM is pre-trained using the relational consistency prediction and replacement language model tasks. Finally, the model is fine-tuned through supervised learning to perform the attribute alignment task for multisource POI data, as per the specific downstream tasks. We used the POI attributes of Baidu Map, Tencent Map, and Gaode Map in Chengdu, China as the experimental data. The findings demonstrate that the proposed model outperforms existing methods for attribute alignment. Specifically, the category attribute consistency achieves a Macro-F1 value of over 90%, and the address attribute standardization achieves a BLEU-4 score of over 95%.

Social semantics: the organization and grounding of abstract concepts.

concepts, like justice and friendship, are central features of our daily lives. Traditionally, abstract concepts are distinguished from other concepts in that they cannot be directly experienced through the senses. As such, they pose a challenge for strongly embodied models of semantic representation that assume a central role for sensorimotor information. There is growing recognition, however, that it is possible for meaning to be 'grounded' via cognitive systems, including those involved in processing language and emotion. In this article, we focus on the specific proposal that social significance is a key feature in the representation of some concepts. We begin by reviewing recent evidence in favour of this proposal from the fields of psycholinguistics and neuroimaging. We then discuss the limited extent to which there is consensus about the definition of 'socialness' and propose essential next steps for research in this domain. Taking one such step, we describe preliminary data from an unprecedented large-scale rating study that can help determine how socialness is distinct from other facets of word meaning. We provide a backdrop of contemporary theories regarding semantic representation and social cognition and highlight important predictions for both brain and behaviour. This article is part of the theme issue 'Concepts in interaction: social engagement and inner experiences'.

A Semantic Mapping Method of Relation Representation Enhancement for Few-Shot Knowledge Graph Completion

Few-shot knowledge graph completion (FKGC) tasks involve determining the authenticity of triple candidates using a small number of reference triples with a given relation. Intuitively, the expression of relation features contributes to the close correlation among the triples with the same relation. Therefore, the relation features are not comprehensive enough, leading to the fact that the triples cannot learn sufficient association information for the FKGC task. In this paper, an enhanced relation semantic representation model is constructed for associative reference triples from both aspects of external structure and internal semantics. On the one hand, as the structure around the triple is helpful to understand the relation semantics implicated in the triple, a graph convolution network with attention and relation features is proposed to obtain the graph structure features. Furthermore, the local structure information of triples can be used to learn the deep relation semantics. On the other hand, entity information can enhance the perception of the relation semantics in the triple. Afterward, in order to associate the triples with the same relation by enhanced relation semantics, a semantic mapping method is proposed, which uses shared merged variables to map the relation, entity, and graph structure features into the same embedding space. Finally, a prototype network based on attention convolution is established to extract the relation prototype representation, and then classify query triples to achieve the purpose of completing the knowledge graph. Through experimental verification, the proposed model achieves excellent performance on two datasets commonly used in the FKGC.

A novel intrinsically explainable model with semantic manifolds established via transformed priors

Because humans instinctively trust and interact with explainable representations instead of latent features, intrinsically interpretable models (IIMs) aimed at representations with semantic meanings have been extensively studied. Previous IIMs relied heavily on data-driven deep neural networks, which resulted in a tremendous demand for data and complex mappings established between instances and their representations. In this study, inspired by how humans build semantic concepts by using meaningful primitives and recognize semantic instances by matching these instances with transformed primitives, we propose the manifold-based semantic representation model (MSRM), which aims to establish a concise mapping with semantic priors for explainable representations. In the MSRM, to reduce reliance on data, we introduced semantic priors into data-driven learning by building a semantic manifold for each different prior with parameterized transformations and then constructed an explainable representation space with these semantic manifolds for given semantic instances. Specifically, the MSRM represents an input semantic instance in three steps: the model extracts the instance’s low-dimensional features as transformation parameters, transforms semantic priors into semantic variations by using these parameters, and calculates the similarity between the input instance and these semantic variations by using inner products as the model’s explainable representation. Hence, the MSRM provides a succinct, prior-guided, and explainable mapping by establishing semantic manifolds by using transformed priors. For application, we propose a manifold-based semantic convolution (MSConv) for visual representation and a simple classification network that includes only one MSConv. The competitive representation power of the MSRM was theoretically analyzed and experimentally verified.

Interval Type-2 Fuzzy Envelope of Proportional Hesitant Fuzzy Linguistic Term Set: Application to Large-Scale Group Decision Making

Large-scale group decision-making (LS-GDM) problems are common in the daily life of human beings. Both information fusion and computing with words (CWW) technologies in LS-GDM suffer from challenges. In the current research, a proportional hesitant fuzzy linguistic term set (PHFLTS) will be applied to capture the preferences of sub-groups in LS-GDM, which decreases the information lost in information fusion processes. Novel fuzzy semantic representation models of PHFLTS, such as type-1 fuzzy envelope and interval type-2 fuzzy envelope, are respectively studied. The application of the proposed fuzzy entropies facilitates the CWW process with the PHFLTS under the framework of a fuzzy linguistic approach. In particular, linguistic uncertainties contained in the PHFLTS can be reflected in a comprehensive way when the type-2 fuzzy envelope is applied, which contributes to the decrease in the information lost during the CWW process. A novel LS-GDM method cooperating with the fuzzy semantic models of PHFLTS is proposed, in which weights for the sub-groups are determined by size, cohesion, and degree of reliability among the sub-groups. Finally, the proposed decision method as well as CWW tools are applied to the process of urban renewal plan selection.

NEURAL NETWORK “PERSEPTRON” AND A SEMANTIC MODEL OF KNOWLEDGE REPRESENTATION FOR STUDYING THE DYNAMICS OF THE RF GDP

Theoretical foundations of GDP dynamics under conditions of market uncertainty and the formation of a digital economy are studied. Forecasting GDP is important because it allows to ensure balance, sustainability in the development of various sectors of the national economy, and its economic security, which determines the practical significance of the study.
 Purpose. To put forward and prove the hypothesis that with the help of the neural network model “perceptron” and the semantic model of knowledge representation, it is possible to obtain a forecast of Russia’s GDP for the next year.
 The scientific novelty lies in the fact that forecasts of the RF GDP were generated using a semantic knowledge representation model and the Perceptron AI system.
 Method or methodology of the work. The work used such methods as monographic, analytical, artificial intelligence system - perceptron, Mind histogram, as well as comparative analysis. Calculations were performed in tables XL perceptron was formed on the Deductor platform. We used a semantic knowledge representation model developed in the DOT language using the Graphviz 2.38 program.
 Results. Theoretical foundations of the use of neural networks in economic forecasts have been studied. A semantic model of knowledge representation has been formed regarding forecasting the GDP of the Russian Federation by a neural network. The developed AI-system “perceptron” is presented, designed to forecast Russia’s GDP based on input parameters representing a set of data reflecting the development of the real sector of the economy and the financial sector of Russia for 2011–2021.
 The scope of the results: economics, the financial sector, forecasting and planning of economic and financial activities.

An associative knowledge network model for interpretable semantic representation of noun context

Uninterpretability has become the biggest obstacle to the wider application of deep neural network, especially in most human–machine interaction scenes. Inspired by the powerful associative computing ability of human brain neural system, a novel interpretable semantic representation model of noun context, associative knowledge network model, is proposed. The proposed network structure is composed of only pure associative relationships without relation label and is dynamically generated by analysing neighbour relationships between noun words in text, in which incremental updating and reduction reconstruction strategies can be naturally introduced. Furthermore, a novel interpretable method is designed for the practical problem of checking the semantic coherence of noun context. In proposed method, the associative knowledge network learned from the text corpus is first regarded as a background knowledge network, and then the multilevel contextual associative coupling degree features of noun words in given detection document are computed. Finally, contextual coherence detection and the location of those inconsistent noun words can be realized by using an interpretable classification method such as decision tree. Our sufficient experimental results show that above proposed method can obtain excellent performance and completely reach or even partially exceed the performance obtained by the latest deep neural network methods especially in F1 score metric. In addition, the natural interpretability and incremental learning ability of our proposed method should be extremely valuable than deep neural network methods. So, this study provides a very enlightening idea for developing interpretable machine learning methods, especially for the tasks of text semantic representation and writing error detection.

Neural stylist: Towards online styling service

Online stylist service enjoys huge economic potentials due to the trend of transformation of the fashion industry to digitalisation. Existing works either predict the fashion compatibility from the overall aspect or evaluate the compatibility with type-conditional representations. The prediction is hard to interpret due to the abstractive forecast. This paper proposes a visual and semantic representation model for explainable evaluation and recommendation. The model considers fashion compatibility from different factors, such as colour, material and style, by leveraging low to high-level features from former to later layers of CNN. The colour correlation and the pairwise relationship of fashion items in the same outfit are considered during the prediction stage. Instead of just predicting an outfit as compatible or incompatible, the model can classify an outfit as three precise evaluation levels: Good, Normal and Bad. The detailed compatible level is more consistent with the fashion sense of our human brain as Good or Bad outfits may have specific characteristics while Normal outfits tend to be ordinary. Additionally, the model can diagnose and recommend substitutions of the problematic fashion items from overall compatibility or colour-specific aspects by tracking the prediction matrices’ backpropagation gradients during the recommendation stage. Experiments in terms of outfit compatibility prediction and fill in the blank are conducted to evaluate the prediction ability of the proposed model. In contrast, fashion substitution recommendation experiments are conducted to assess the compatibility diagnosis and recommendation ability. Quantitative and qualitative results show that the model enables online stylist services with excellent explainability and generalisation on fashion prediction and recommendation.

A semantic element representation model for malicious domain name detection

The existing detection methods of algorithmically generated malicious domain names lack theoretical modelling methods for domain name element composition. To address this problem, a semantic element representation model for domain names is constructed based on the set of semantic elements of domain names and the probabilistic context free grammar model. The model first analyses and categorises the constituent elements of the domain name, and then proposes a syntax tree analysis method for the semantical relationships between the elements, which enables efficient representation of multiple elements in domain names. Based on the proposed model, the malicious domain names are categorised into four categories: random character-based, word-based, predicted character-based, and multi-element hybrid. Experiments are conducted to analyse the anomalies and concealment of domain names, the results denote that there are significant differences between malicious and legitimate domain names, as well as between malicious domain names, and the comparative experimental results denote the proposed model can effectively improve the detection accuracy of malicious domain names.

Automated post scoring: evaluating posts with topics and quoted posts in online forum.

Online forumpost evaluationis an effective way for instructors to assess students’ knowledge understanding and writing mechanics. Manually evaluating massive posts costs a lot of time. Automatically grading online posts could significantly alleviate instructors’ burden. Similar text assessment tasks like Automated Text Scoring evaluate the writing quality of independent texts or relevance between text and prompt. And Automatic Short Answer Grading measures the semantic matching of short answers according to given problems and correct answers. Different from existing tasks, we propose a novel task, Automated Post Scoring (APS), which grades all online discussion posts in each thread of each student with given topics and quoted posts. APS evaluates not only the writing quality of posts automatically but also the relevance to topics. To measure the relevance, we model the semantic consistency between posts and topics. Supporting arguments are also extracted from quoted posts to enhance posts evaluation. Specifically, we propose a mixture model including a hierarchical text model to measure the writing quality, a semantic matching model to model topic relevance, and a semantic representation model to integrate quoted posts. We also construct a new dataset called Online Discussion Dataset containing 2,542 online posts from 694 students of a social science course. The proposed models are evaluated on the dataset with correlation and residual based evaluation metrics. Compared with measuring posts alone, experimental results demonstrate that incorporating topics and quoted posts could improve the performance of APS by a large margin, more than 9 percent on QWK.

DA-GAN: Dual Attention Generative Adversarial Network for Cross-Modal Retrieval

Cross-modal retrieval aims to search samples of one modality via queries of other modalities, which is a hot issue in the community of multimedia. However, two main challenges, i.e., heterogeneity gap and semantic interaction across different modalities, have not been solved efficaciously. Reducing the heterogeneous gap can improve the cross-modal similarity measurement. Meanwhile, modeling cross-modal semantic interaction can capture the semantic correlations more accurately. To this end, this paper presents a novel end-to-end framework, called Dual Attention Generative Adversarial Network (DA-GAN). This technique is an adversarial semantic representation model with a dual attention mechanism, i.e., intra-modal attention and inter-modal attention. Intra-modal attention is used to focus on the important semantic feature within a modality, while inter-modal attention is to explore the semantic interaction between different modalities and then represent the high-level semantic correlation more precisely. A dual adversarial learning strategy is designed to generate modality-invariant representations, which can reduce the cross-modal heterogeneity efficiently. The experiments on three commonly used benchmarks show the better performance of DA-GAN than these competitors.

Improving Visual Reasoning Through Semantic Representation

In visual reasoning, the achievement of deep learning significantly improved the accuracy of results. Image features are primarily used as input to get answers. However, the image features are too redundant to learn accurate characterizations within a limited complexity and time. While in the process of human reasoning, abstract description of an image is usually to avoid irrelevant details. Inspired by this, a higher-level representation named semantic representation is introduced. In this paper, a detailed visual reasoning model is proposed. This new model contains an image understanding model based on semantic representation, feature extraction and process model refined with watershed and u-distance method, a feature vector learning model using pyramidal pooling and residual network, and a question understanding model combining problem embedding coding method and machine translation decoding method. The feature vector could better represent the whole image instead of overly focused on specific characteristics. The model using semantic representation as input verifies that more accurate results can be obtained by introducing a high-level semantic representation. The result also shows that it is feasible and effective to introduce high-level and abstract forms of knowledge representation into deep learning tasks. This study lays a theoretical and experimental foundation for introducing different levels of knowledge representation into deep learning in the future.

Multi-granularity semantic representation model for relation extraction

In natural language, a group of words constitute a phrase and several phrases constitute a sentence. However, existing transformer-based models for sentence-level tasks abstract sentence-level semantics from word-level semantics directly, which override phrase-level semantics so that they may be not favorable for capturing more precise semantics. In order to resolve this problem, we propose a novel multi-granularity semantic representation (MGSR) model for relation extraction. This model can bridge the semantic gap between low-level semantic abstraction and high-level semantic abstraction by learning word-level, phrase-level, and sentence-level multi-granularity semantic representations successively. We segment a sentence into entity chunks and context chunks according to an entity pair. Thus, the sentence is represented as a non-empty segmentation set. The entity chunks are noun phrases, and the context chunks contain the key phrases expressing semantic relations. Then, the MGSR model utilizes inter-word, inner-chunk and inter-chunk three kinds of different self-attention mechanisms, respectively, to learn the multi-granularity semantic representations. The experiments on two standard datasets demonstrate our model outperforms the previous models.

Multi-Ontology Refined Embeddings (MORE): A hybrid multi-ontology and corpus-based semantic representation model for biomedical concepts.

Currently, a major limitation for natural language processing (NLP) analyses in clinical applications is that concepts are not effectively referenced in various forms across different texts. This paper introduces Multi-Ontology Refined Embeddings (MORE), a novel hybrid framework that incorporates domain knowledge from multiple ontologies into a distributional semantic model, learned from a corpus of clinical text. We use the RadCore and MIMIC-III free-text datasets for the corpus-based component of MORE. For the ontology-based part, we use the Medical Subject Headings (MeSH) ontology and three state-of-the-art ontology-based similarity measures. In our approach, we propose a new learning objective, modified from the sigmoid cross-entropy objective function. We used two established datasets of semantic similarities among biomedical concept pairs to evaluate the quality of the generated word embeddings. On the first dataset with 29 concept pairs, with similarity scores established by physicians and medical coders, MORE's similarity scores have the highest combined correlation (0.633), which is 5.0% higher than that of the baseline model, and 12.4% higher than that of the best ontology-based similarity measure. On the second dataset with 449 concept pairs, MORE's similarity scores have a correlation of 0.481, based on the average of four medical residents' similarity ratings, and that outperforms the skip-gram model by 8.1%, and the best ontology measure by 6.9%. Furthermore, MORE outperforms three pre-trained transformer-based word embedding models (i.e., BERT, ClinicalBERT, and BioBERT) on both datasets. MORE incorporates knowledge from several biomedical ontologies into an existing corpus-based distributional semantics model, improving both the accuracy of the learned word embeddings and the extensibility of the model to a broader range of biomedical concepts. MORE allows for more accurate clustering of concepts across a wide range of applications, such as analyzing patient health records to identify subjects with similar pathologies, or integrating heterogeneous clinical data to improve interoperability between hospitals.