Semantic Relations Research Articles

Biomedical relation extraction method based on ensemble learning and attention mechanism

BackgroundRelation extraction (RE) plays a crucial role in biomedical research as it is essential for uncovering complex semantic relationships between entities in textual data. Given the significance of RE in biomedical informatics and the increasing volume of literature, there is an urgent need for advanced computational models capable of accurately and efficiently extracting these relationships on a large scale.ResultsThis paper proposes a novel approach, SARE, combining ensemble learning Stacking and attention mechanisms to enhance the performance of biomedical relation extraction. By leveraging multiple pre-trained models, SARE demonstrates improved adaptability and robustness across diverse domains. The attention mechanisms enable the model to capture and utilize key information in the text more accurately. SARE achieved performance improvements of 4.8, 8.7, and 0.8 percentage points on the PPI, DDI, and ChemProt datasets, respectively, compared to the original BERT variant and the domain-specific PubMedBERT model.ConclusionsSARE offers a promising solution for improving the accuracy and efficiency of relation extraction tasks in biomedical research, facilitating advancements in biomedical informatics. The results suggest that combining ensemble learning with attention mechanisms is effective for extracting complex relationships from biomedical texts. Our code and data are publicly available at: https://github.com/GS233/Biomedical.

Zero-Shot Relation Classification Through Inference on Category Attributes.

The goal of relationship classification (RC) is to predict the semantic relationship between two entities in a given sentence. With the advent of deep learning and pretrained language models, RC research has progressed by leaps and bounds. However, the current studies are focused mainly on predicting semantic relationships from a predefined set. How to recognize unseen relationships remains a challenge, which is also known as the zero-shot RC (ZSRC) task. Some ZSRC-related methods directly map relationship categories to numerical indices, constraining the model's ability to autonomously infer and understand these relationships, while others rely heavily on manual definitions. To address these issues and inspired by the way of reasoning in which humans perform RC tasks, we propose a new framework to handle the ZSRC task through inference on category attributes (ICAs). The main idea of ICA is to detect the semantic relationship between promises, which are RC sentences, and hypotheses, which are relational sentences of entities created by templates. Specifically, instead of manual design, we introduce two hypothesis templates derived from the label words (LWs) and descriptions (LDs) associated with each relationship. These templates are used to automatically convert the RC data into the textual entailment (TE) format. Furthermore, they are fine-tuned with a pretrained TE model, facilitating the acquisition of relational knowledge and enabling the generalization of semantic reasoning rules learned from seen classes to unseen classes. Moreover, to implement multirelationship semantic inference for all unseen classes, we propose an entailment difference mechanism to enhance the reasoning capability of the model. Besides the current ZSRC test setting, we also examine our method in an even more challenging setting to deal with data scarcity in real-world applications. The outstanding performance of ICA on the FewRel and Wiki-ZSL datasets demonstrates its effectiveness in the ZSRC task.

Dynamics of second-language learners' semantic memory networks: Evidence from a snowball sampling paradigm.

This article provides an analysis of structural changes in second-language (L2)-based semantic memory networks-graphs composed of L2 words as nodes and semantic relations between them as edges, during L2 learning. We used snowball sampling paradigm to create individual semantic networks of participants divided into two groups differing in L2 learning time and then compare their structural characteristics cross-sectionally. The results showed that as L2 learning progresses, semantic memory networks tend to become more connected (by increasing the average node degree), more efficient (by decreasing the average shortest path length), less fragmented (by decreasing the modularity), less centralized (by decreasing the centralization), less dense (by decreasing the density), and no more "small-worlded" (by similar average clustering coefficients and small-world indices). The findings provide quantitative evidence of how the duration of L2 learning shapes the structure of L2-based semantic memory networks generated in the snowball sampling paradigm. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

GeoEntity-type constrained knowledge graph embedding for predicting natural-language spatial relations

Natural-language spatial relations between geographic entities (geoentities) reflect diverse perceptions influenced by factors like location, culture, and linguistic conventions. These relations play a crucial role in supporting geospatial tasks, such as question answering and cognitive reasoning. While prior studies focused on a limited set of human-selected spatial terms and geometric attributes, they often overlooked essential semantic attributes. To overcome this limitation, we developed a Spatial Relation-based Knowledge Graph Embedding framework, SR-KGE, with new KG fusion functions to predict spatial relation terms among distinct geoentities. This method not only considers graph structures and the diversity of natural language expressions in the embedding and learning process, but also incorporates geoentity types as a constraint to capture spatial and semantic relations more accurately. Our experiments on two knowledge graph datasets, one small-scale and one large-scale, have both shown its superior performance in spatial relation inference compared to popular KGE models, including TransE, RotatE, and HAKE. We hope our research will advance the classic study of natural language described spatial relations in a more automated and intelligent way.

Thematic Relations

This study explores the processing of noun-prep-noun combinations in prepositional phrases in Brazilian Portuguese, focusing on the syntactic-semantic relationship marked by prepositions, implies the semantic sense in which the noun is being modified. We chose to contrast particularly two types of thematic relationships like 'made of' and 'made for’, categorized as: material and function (e.g. ‘toalha de crochê’, crocheted cloth vs. agulha de crochê, crochet needle). To do so, we conducted two psycholinguistic experiments: Experiment 1 consisted of a self-paced reading/picture matching task to investigate the impact of semantic relationship on processing time; and Experiment 2 employed a cloze task (e.g. ‘toalha de ______’) and Likert scale likelihood judgments to norm the stimuli and assess cognitive preferences for the types of relation explored (i.e., material or function). Our results revealed slower reading times for the function type relation, suggesting a higher processing demand, possibly due to covert semantic computations required for this relation. Cloze completions and likelihood ratings also indicated a cognitive preference for this type of relationship. We attribute these observed patterns to the robust syntactic-semantic relationship inherent to functional type relations, coupled with a broader range of potential response options. In contrast, material type relations demonstrated a higher level of convergence among participants for cloze responses, likely due to more restricted response options and, consequently, increased probability. Future EEG studies are proposed to identify specific moments in the cognitive processing of the constituents of these structures and assess possible anticipatory effects of prepositional phrase modifiers. These discoveries make a meaningful contribution to the broader literature on language processing, aligning with analogous patterns observed in different languages with a cognitive distinction between thematic relations and feature-based (i.e., telic and constitutive relations), such as in Wisniewsky & Love, 1998; Estes et al., 2003, 2011; Pustejovsky et al., 2016; Flick et al., 2021.

Quantum Approach for Contextual Search, Retrieval, and Ranking of Classical Information.

Quantum-inspired algorithms represent an important direction in modern software information technologies that use heuristic methods and approaches of quantum science. This work presents a quantum approach for document search, retrieval, and ranking based on the Bell-like test, which is well-known in quantum physics. We propose quantum probability theory in the hyperspace analog to language (HAL) framework exploiting a Hilbert space for word and document vector specification. The quantum approach allows for accounting for specific user preferences in different contexts. To verify the algorithm proposed, we use a dataset of synthetic advertising text documents from travel agencies generated by the OpenAI GPT-4 model. We show that the "entanglement" in two-word document search and retrieval can be recognized as the frequent occurrence of two words in incompatible query contexts. We have found that the user preferences and word ordering in the query play a significant role in relatively small sizes of the HAL window. The comparison with the cosine similarity metrics demonstrates the key advantages of our approach based on the user-enforced contextual and semantic relationships between words and not just their superficial occurrence in texts. Our approach to retrieving and ranking documents allows for the creation of new information search engines that require no resource-intensive deep machine learning algorithms.

Challenges in building scholarly knowledge graphs for research assessment in open science

Abstract Open science has revolutionized scholarly communication and research assessment by introducing research data and software as first-class citizens. Scholarly knowledge graphs (SKGs) are expected to play a crucial role in generating research assessment indicators being able to aggregate bibliographic metadata records and semantic relationships describing all research products and their links (e.g., citations, affiliations, funding). However, the rapid advance of open science has led to publication workflows that do not adequately support and guarantee the authenticity of products and metadata quality required for research assessment. Additionally, the heterogeneity of research communities and the multitude of data sources and exchange formats complicate the provision of consistent and stable SKGs. This work builds upon the experience gained from pioneering and addressing these challenges in the OpenAIRE Graph SKG. The aim is twofold and broader. First, we identify obstacles to the creation of SKGs for research assessment caused by the state-of-the-art publishing workflows for publications, software, and data. Second, we describe repurposing SKGs as tools to monitor such workflows to identify and heal their shortcomings, taking advantage of tools, techniques, and practices that support the actors involved, namely research communities, scientists, organizations, data source providers, and SKG providers, to improve the Open Science scholarly publishing ecosystem.

Unsupervised Multi-Scale Hybrid Feature Extraction Network for Semantic Segmentation of High-Resolution Remote Sensing Images

Generating pixel-level annotations for semantic segmentation tasks of high-resolution remote sensing images is both time-consuming and labor-intensive, which has led to increased interest in unsupervised methods. Therefore, in this paper, we propose an unsupervised multi-scale hybrid feature extraction network based on the CNN-Transformer architecture, referred to as MSHFE-Net. The MSHFE-Net consists of three main modules: a Multi-Scale Pixel-Guided CNN Encoder, a Multi-Scale Aggregation Transformer Encoder, and a Parallel Attention Fusion Module. The Multi-Scale Pixel-Guided CNN Encoder is designed for multi-scale, fine-grained feature extraction in unsupervised tasks, efficiently recovering local spatial information in images. Meanwhile, the Multi-Scale Aggregation Transformer Encoder introduces a multi-scale aggregation module, which further enhances the unsupervised acquisition of multi-scale contextual information, obtaining global features with stronger feature representation. The Parallel Attention Fusion Module employs an attention mechanism to fuse global and local features in both channel and spatial dimensions in parallel, enriching the semantic relations extracted during unsupervised training and improving the performance of unsupervised semantic segmentation. K-means clustering is then performed on the fused features to achieve high-precision unsupervised semantic segmentation. Experiments with MSHFE-Net on the Potsdam and Vaihingen datasets demonstrate its effectiveness in significantly improving the accuracy of unsupervised semantic segmentation.

A strategic evaluation of BIM-driven information management in the context of ISO 19650-2 standard

PurposeThis paper presents an in-depth evaluation of BIM-driven information management in the context of ISO 19650-2. The study aimed to understand and evaluate the existing mechanisms and the recommended activities and tasks of collaborative information management, particularly as the Turkish AEC organizations are currently implementing ISO 19650-2.Design/methodology/approachThe study employed a qualitative research design based on information management stages of ISO 19650-2. Semi-structured interviews were conducted with purposefully selected professionals. The framework analyses were made with MAXQDA through conceptual coding to identify current drivers, challenges and incentives for BIM-driven information management. The frequencies of common themes, required tasks and semantic relationships were reported systematically and interpreted to provide a conceptual implementation map, practical insights and actionable recommendations.FindingsThe study results encompass strategies for establishing new tasks and implementation procedures to uphold business value by aligning information management standards with evolving project requirements, providing detailed and responsive EIRs, early use of CDEs, compliant production methods, clarity of roles and responsibilities, increased organizational capacity and capabilities, sustained planning and production quality and leveraging lessons learned.Originality/valueThe study provides evidence-based arguments and strategic insights for BIM-driven information management, focusing on the current implementation compatibilities with ISO 19650-2. Findings and discussions highlight the necessity of industry-wide transformations, reengineered activities/tasks and compliant information flows, which are crucial for adopting ISO 19650-2 and the formulation of national annex documents.

Intelligent identification of risks in construction contract clauses based on semantic reasoning

Purpose The number of construction dispute cases has surged in recent years. The effective exploration and management of risks associated with construction contracts helps to directly enhance the overall project performance. The existing approaches to identify the risks associated with construction project contracts have a heavy reliance on manual review techniques, which are inefficient and highly restricted by personnel experience. The existing intelligent approaches only work for the contract query and storage. Hence, it is necessary to improve the intelligence level for contract risk management. This study aims to propose a novel method for the intelligent identification of risks in construction contract clauses based on natural language processing.Design/methodology/approach This proposed method can formalize the linguistic logic and semantic information of contract clauses into multiple triples and transform the structural processing results of general clauses in a construction contract into rights and interests rules for risk review. In addition, the core semantic information of special clauses in a construction contract, rights and interests rules are used for semantic conflict detection. Finally, this study achieves the intelligent risk identification of construction contract clauses.Findings The method is verified by selecting several construction contracts that had been applied in engineering contracting as a corpus. The results showed a high level of accuracy and applicability of the proposed method.Originality/value This novel method can identify the risks in contract clauses with complex syntactic structures and realize rule extension according to the semantic relation network of the ontology. It can support efficient contract review and assist the decision-making process in contract risk management.

Fully Verified Instruction Scheduling

CompCert project, the state-of-the-art compiler that achieves the first end-to-end formally verified C compiler, does not support fully verified instruction scheduling. Instead, existing research that works on such topics only implements translation validation. This means they do not have direct formal proof that the scheduling algorithm is correct, but only a posterior validation to check each compiling case. Using such a method, CompCert accepts a valid C program and compiles correctly only when the untrusted scheduler generates a correct result. However, it does not guarantee the complete correctness of the scheduler. It also causes compile-time validation overhead in the view of runtime performance. In this work, we present the first achievement in developing a mechanized library for fully verified instruction scheduling while keeping the proof workload acceptably lightweight. The idea to reduce the proof length is to exploit a simple property that the topological reordering of a topological sorted list is equal to a sequence of swapping adjacent unordered elements. Together with the transitivity of semantic simulation relation, the only burden will become proving the semantic preservation of a transition that only swaps two adjacent independent instructions inside one block. After successfully proving this result, proving the correctness of any new instruction scheduling algorithm only requires proof that it preserved the syntax-level dependence among instructions, instead of reasoning about semantics details every time. We implemented a mechanized library of such methods in the Coq proof assistant based on CompCert's library as a framework and used the list scheduling algorithm as a case study to show the correctness can be formally proved using our theory. We show that with our method that abstracts away the semantics details, it is flexible to implement any scheduler that reorders instructions with little extra proof burden. Our scheduler in the case study also abstracts away the outside scheduling heuristic as a universal parameter so it is flexible to modify without touching any correctness proof.

Dependency-Aware Code Naturalness

Code naturalness, which captures repetitiveness and predictability in programming languages, has proven valuable for various code-related tasks in software engineering. However, precisely measuring code naturalness remains a fundamental challenge. Existing methods measure code naturalness over individual lines of code while ignoring the deep semantic relations among different lines, e.g., program dependency, which may negatively affect the precision of the measure. Despite the intuitive appeal of extending the code naturalness measure to the code dependency domain (as there are some work that have initiated the utilization of code dependency for diverse code-related tasks), this assumption remains unexplored and warrants direct investigation. In this study, we aim to perform the first empirical study to investigate whether incorporating code dependency, instead of analyzing individual lines, can enhance the precision of measuring code naturalness. To achieve that, we first propose a new method named DAN for measuring code naturalness by incorporating the rich dependency information in the code. Specifically, DAN extracts multiple sequences of code lines by traversing the program dependency graph, where different code lines are connected by dependencies in each sequence, and then the code naturalness will be measured by taking each sequence as a whole. In this way, the dependency information can be well captured. Finally, we have conducted an extensive study to evaluate the influence of code dependency for measuring code naturalness with DAN, and compared it with the state-of-the-art methods under three emerging application scenarios of code naturalness. The results demonstrate that DAN can not only better distinguish natural and unnatural code, but also substantially boost two important downstream applications of code naturalness, i.e., distinguishing buggy and non-buggy code lines and data cleansing for training better code models, reflecting the significance of code dependency in measuring code naturalness.

Label-Guided relation prototype generation for Continual Relation Extraction

Continual relation extraction (CRE) aims to extract relations towards the continuous and iterative arrival of new data. To address the problem of catastrophic forgetting, some existing research endeavors have focused on exploring memory replay methods by storing typical historical learned instances or embedding all observed relations as prototypes by averaging the hidden representation of samples and replaying them in the subsequent training process. However, this prototype generation method overlooks the rich semantic information within the label namespace and are also constrained by memory size, resulting in inadequate descriptions of relation semantics by relation prototypes. To this end, we introduce an approach termed Label-Guided Relation Prototype Generation. Initially, we enhance the representations of label embeddings through a technique named label knowledge infusion. Following that, we utilize the multi-head attention mechanism to form relation prototypes, allowing them to capture diverse aspects of typical instances. The embeddings of relation labels are utilized at this stage, leveraging their contained semantics. Additionally, we propose a feature-based distillation loss function called multi-similarity distillation, to ensure the model retains prior knowledge after learning new tasks. The experimental results indicate that our method has achieved competitive performance compared to the state-of-the-art baseline models in CRE.

A Hierarchically Discriminative Loss with Group Regularization for Fine-Grained Image Classification

Fine-grained visual classification targets the discrimination of subordinate categories within broader classes, such as avian species, aerial crafts, and notably, in medical diagnostics like breast cancer. In the domain of breast cancer classification, there has been an emergence of numerous fine-grained models leveraging pathological images. Elevating the interpretability of the model’s discriminative processes among these nuanced categories holds promise in enhancing the transparency of AI decision-making. However, the effective utilization of hierarchical label structures within medical data remains a crucial consideration. In this work, we explore how the label hierarchy can be used to better learn subtle feature embeddings. We observe that the semantic relationships between fine-grained categories can help to analyze the misclassified samples. To this end, with the label hierarchy, we introduce two novel losses to cultivate subtle feature representations, coordinate with feature learning, and align with the principles of Explainable AI (XAI). First, we propose hierarchically discriminative loss to enhance the interactions between fine- and coarse-level features, which can help reinforce the discriminability of fine-grained features for reducing false predictions belonging to the out-group relation. Second, we introduce the in-group regularization loss to establish the interactions between the target class and those confusing classes belonging to the in-group relation. Treating another confusing class as a distraction can regularize feature learning of the target class. Thus, it allows the network to discover more discriminative features and reduces in-group false predictions. Five commonly used datasets for fine-grained classification are extensively evaluated. Our experimental results validate the effectiveness of our proposed novel losses when compared to state-of-the-art methods that utilize hierarchical multi-granularity labels.

NBCR-ac4C: A Deep Learning Framework Based on Multivariate BERT for Human mRNA N4-Acetylcytidine Sites Prediction.

N4-acetylcytidine (ac4C) plays a crucial role in regulating cellular biological processes, particularly in gene expression regulation and disease development. However, experiments to identify ac4C in a wet lab are time-consuming and costly, and the learning-based methods struggle to capture the underlying semantic knowledge and relations within sequences. To address this, we propose a deep learning approach called NBCR-ac4C based on pretrained models. Specifically, we employ Nucleotide Transformer and DNABERT2 to construct contextual embedding of nucleotide sequences, which effectively mine and express context relations between different features in the sequence. Convolutional neural network (CNN) and ResNet18 are then applied to further extract shallow and deep knowledge from context embedding. Depending on extensive experiments for the prediction of ac4C sites in nucleotide sequences, we observe that NBCR-ac4C outperforms general learning-based models. It achieves the highest accuracy (ACC) of 83.51% and an Area Under the Receiver Operating Characteristic Curve (AUROC) of 89.58% on an independent test set. Moreover, the proposed model, compared to the current state-of-the-art (SOTA) model LSA-ac4C, demonstrates higher ACC and AUROC by 0.81-3.7% and 0.05-1.58%, respectively. The data set and code are available on https://github.com/2103374200/NBCR to facilitate further discussion on NBCR-ac4C.

Discovering API usage specifications for security detection using two-stage code mining

An application programming interface (API) usage specification, which includes the conditions, calling sequences, and semantic relationships of the API, is important for verifying its correct usage, which is in turn critical for ensuring the security and availability of the target program. However, existing techniques either mine the co-occurring relationships of multiple APIs without considering their semantic relationships, or they use data flow and control flow information to extract semantic beliefs on API pairs but difficult to incorporate when mining specifications for multiple APIs. Hence, we propose an API specification mining approach that efficiently extracts a relatively complete list of the API combinations and semantic relationships between APIs. This approach analyzes a target program in two stages. The first stage uses frequent API set mining based on frequent common API identification and filtration to extract the maximal set of frequent context-sensitive API sequences. In the second stage, the API relationship graph is constructed using three semantic relationships extracted from the symbolic path information, and the specifications containing semantic relationships for multiple APIs are mined. The experimental results on six popular open-source code bases of different scales show that the proposed two-stage approach not only yields better results than existing typical approaches, but also can effectively discover the specifications along with the semantic relationships for multiple APIs. Instance analysis shows that the analysis of security-related API call violations can assist in the cause analysis and patch of software vulnerabilities.

Multi-Organ Foundation Model for Universal Ultrasound Image Segmentation with Task Prompt and Anatomical Prior.

Semantic segmentation of ultrasound (US) images with deep learning has played a crucial role in computer-aided disease screening, diagnosis and prognosis. However, due to the scarcity of US images and small field of view, resulting segmentation models are tailored for a specific single organ and may lack robustness, overlooking correlations among anatomical structures of multiple organs. To address these challenges, we propose the Multi-Organ FOundation (MOFO) model for universal US image segmentation. The MOFO is optimized jointly from multiple organs across various anatomical regions to overcome the data scarcity and explore correlations between multiple organs. The MOFO extracts organ-invariant representations from US images. Simultaneously, the task prompt is employed to refine organ-specific representations for segmentation predictions. Moreover, the anatomical prior is incorporated to enhance the consistency of the anatomical structures. A multi-organ US database, comprising 7039 images from 10 organs across various regions of the human body, has been established to evaluate our model. Results demonstrate that the MOFO outperforms single-organ methods in terms of the Dice coefficient, 95% Hausdorff distance and average symmetric surface distance with statistically sufficient margins. Our experiments in multi-organ universal segmentation for US images serve as a pioneering exploration of improving segmentation performance by leveraging semantic and anatomical relationships within US images of multiple organs.

Etymology, History and Development of the Semantics of Food Names in Ukrainian Chronicles of the End of the 17th and the Beginning of the 18th Century

The article examines the etymology, history and semantic development of food names found in late 17th- and early 18th-century Ukrainian chronicles, including those by Samovydetsia, S. Velichko, and H. Hrabianka. Taking a synchronic-diachronic approach, the author analyzes these food nominations within the lexical-semantic domains of generic food names and dish names, establishing semantic relationships such as polysemy, synonymy and variation. This analysis traces the evolution of the Ukrainian language’s lexical composition and reveals its nationally distinct characteristics. By following these food nominations from their earliest written attestations through the period depicted in the chronicles, the study clearly illustrates the semantic development of the given food terminology. The author notes that words of Proto-Slavic origin comprise a significant portion of the food terminology used in late 17th- and early 18th-century Ukrainian chronicles. Borrowings from various languages are also used, including Polish, Old Slavic, Turkic, Russian, and German (occasionally by way of Polish), as well as Italian via German. The presence of these loanwords is explained by key historical events in Ukraine across different periods that shaped the development of the language. Taken together, the linguistic evidence across these chronicles illustrates the codification of a fledgling Ukrainian literary tradition firmly rooted in popular verbal traditions.

MeSH2Matrix: combining MeSH keywords and machine learning for biomedical relation classification based on PubMed

Biomedical relation classification has been significantly improved by the application of advanced machine learning techniques on the raw texts of scholarly publications. Despite this improvement, the reliance on large chunks of raw text makes these algorithms suffer in terms of generalization, precision, and reliability. The use of the distinctive characteristics of bibliographic metadata can prove effective in achieving better performance for this challenging task. In this research paper, we introduce an approach for biomedical relation classification using the qualifiers of co-occurring Medical Subject Headings (MeSH). First of all, we introduce MeSH2Matrix, our dataset consisting of 46,469 biomedical relations curated from PubMed publications using our approach. Our dataset includes a matrix that maps associations between the qualifiers of subject MeSH keywords and those of object MeSH keywords. It also specifies the corresponding Wikidata relation type and the superclass of semantic relations for each relation. Using MeSH2Matrix, we build and train three machine learning models (Support Vector Machine [SVM], a dense model [D-Model], and a convolutional neural network [C-Net]) to evaluate the efficiency of our approach for biomedical relation classification. Our best model achieves an accuracy of 70.78% for 195 classes and 83.09% for five superclasses. Finally, we provide confusion matrix and extensive feature analyses to better examine the relationship between the MeSH qualifiers and the biomedical relations being classified. Our results will hopefully shed light on developing better algorithms for biomedical ontology classification based on the MeSH keywords of PubMed publications. For reproducibility purposes, MeSH2Matrix, as well as all our source codes, are made publicly accessible at https://github.com/SisonkeBiotik-Africa/MeSH2Matrix.

Relation semantic fusion in subgraph for inductive link prediction in knowledge graphs

Inductive link prediction (ILP) in knowledge graphs (KGs) aims to predict missing links between entities that were not seen during the training phase. Recent some subgraph-based methods have shown some advancements, but they all overlook the relational semantics between entities during subgraph extraction. To overcome this limitation, we introduce a novel inductive link prediction model named SASILP (Structure and Semantic Inductive Link Prediction), which comprehensively incorporates relational semantics in both subgraph extraction and node initialization processes. The model employs a random walk strategy to calculate the structural scores of neighboring nodes and utilizes an enhanced graph attention network to determine their semantic scores. By integrating both structural and semantic scores, SASILP strategically selects key nodes to form a subgraph. Furthermore, the subgraph is initialized with a node initialization technique that integrates information about neighboring relations. The experiments conducted on benchmark datasets demonstrate that SASILP outperforms state-of-the-art methods on inductive link prediction tasks, and verify the effectiveness of our approach.