Semantic Method Research Articles

Application of semantic analysis in strategic marketing using the mind map tool

The article discusses the marketing use of a tool used to find solutions and systematize information as mind maps. Due to the increase in the volume and flow of information, a high degree of laboriousness of the analysis of texts in natural languages appears. To solve marketing problems and develop effective marketing strategies, the processing and analysis of the information received is of great importance. For more efficient use of tools, like mind maps, and minimization of manual data processing, there are requests such as combining maps into one from several experts or different map iterations from one expert. A technique is proposed for solving the problem of combining information from mind maps into one by semantic analysis methods, consisting of five stages. An algorithm for combining a set of mind maps into one using the methods of semantic text analysis is presented. To test the methodology, an experiment was conducted, during which more than 30 mind maps were analyzed. Based on the results of the analysis, a combined mind map based on the proposed methodology is presented and the problem of combining a number of heterogeneous maps into one is solved using semantic analysis. The article uses such methods, as literature analysis, experiment, semantic analysis of the text.

Towards machine-readable semantic-based E-business contract representations using Network of Timed Automata (NTA)

The smart contract is generally stored and executed by a computer program in the blockchain platform. However, the representation of program code is unfriendly for non-software developers since the process of a smart contract involves multiple parties such as lawyers, which is not easy to understand for most humans. To consider a broad acceptance of contracts, we design a novel contract representation to narrow the gap between smart contracts and e-business contracts. This paper proposes a novel semantic-based e-business contract model - Simple Natural Contract (SimNC) to represent a universal contract created by a Supervised Sentence Contract (SSC), which is inputted via a Semantic Input Method (SIM) with strict grammar from a human-understandable natural language contract. Then, the SSC is analyzed through Machine Natural Language (MNL) to enhance contract semantic understanding by enabling case grammar for crossing language parties. In doing so, SimNC analyses various deontic components, and combines them with the operational aspects of a legal contract, to achieve a common and better understanding between hard-code and natural language. In addition, we apply the SimNC into a Network of Timed Automata (NTA) for supporting automation, which builds a formal model including temporal constraints and then translates into an executable SimNC-NTA model. This paper aims to provide a bridge between natural language contracts and e-business contracts, making them universal and intelligible.

Historical information fusion of dense multi-source satellite image time series for flood extent mapping

Existing lightweight flood extent mapping (FEM) methods, leveraging satellite imagery, offer respite from the burden of extensive data processing. However, these methods often grapple with suboptimal accuracy of the FEM results. To tackle this challenge, this paper introduces an innovative lightweight historical information fusion framework aimed at enhancing FEM results through the utilization of dense multi-source satellite image time series (MSSITS). Firstly, we propose an ultra-lightweight convolutional neural network comprising fundamental network modules, such as convolution layers, batch normalization layers, and parametric rectified linear unit layers. This network swiftly extracts initial FEM results from dense MSSITS. Secondly, we devise a historical information fusion strategy to refine the initial FEM results of the current image by fusing historical FEM results exhibiting high temporal correlation. During the fusion stage, we harness the complementary attributes of dense MSSITS. Specifically, the fusion of diverse satellite sensor data aids in mitigating false alarms in the initial FEM results, thereby facilitating the acquisition of refined FEM results. This approach effectively diminishes the impact of noise interferences, including terrain shadows and surging flood waves in synthetic aperture radar images, as well as cloud occlusions and cloud shadows in optical images. Experimental findings, based on the dense MSSITS of four authentic flood events, demonstrate that our proposed method yields relatively superior FEM performance while maintaining a low FEM complexity in comparison with commonly used semantic segmentation methods.

Semantics of the lexemes «freedom» and «will» in the Old Russian language as a reflection of the value dominants of the historical era (based on the material of dictionaries)

The article is devoted to identifying the connection of meanings within the semantic structure of the words «freedom» and «will» in the Old Russian language, indirectly reflecting the system of values and beliefs of the corresponding historical era. The relevance of the research is due to the insufficient study of the patterns of interpreting abstract concepts in the Russian language of the older generation. Uses a complex of linguistic methods, such as analysis of dictionary definitions, semantic field method, contextual analysis, etc. The semantics of the lexemes «freedom» and «will» in the Old Russian language is characterized by syncretism, which determines the complex connection of the meanings of these units. Thus, the word «freedom» reflects a number of correlated ideas about this phenomenon: 1) the internal (psychological) state of freedom; 2) social state of freedom; 3) causation of social freedom; 4) causation of internal freedom; 5) a person is a carrier of a state. The corresponding semantic field, the core of which is the word «freedom», in the Old Russian language has an isostructural organization. The semantic structure of the word «will» is less extensive, but it also represents the connection of this concept with the internal (mental), social and religious spheres. It is shown that freedom and will are interpreted in the Old Russian language as significant axiological phenomena that determine the self-awareness of an individual and the norms of his behavior in various spheres of life. The semantics of the words «freedom» and «will» reflect ideas about the close interaction of two principles: sacred (religious) and profane (psychological and social). At the same time, these lexemes emphasize one or another specified aspect: «freedom» primarily correlates with the social and religious spheres, while the «will» turns out to be associated with the internal (psychological) sphere of desire.

LogGT: Cross-system log anomaly detection via heterogeneous graph feature and transfer learning

Automated system log anomaly detection plays a crucial role in ensuring service reliability. Existing methods incompletely utilize structured log entries, resulting in the loss of key information such as components and time. Besides, due to the limitations of labeled data, models trained by a single system are difficult to apply to other systems. Therefore, we propose a cross-system log anomaly detection method named LogGT, which simultaneously models log events, components and time, leveraging labeled system knowledge to achieve anomaly detection in unlabeled systems. Specifically, we firstly design a heterogeneous graph to accurately represent the interactions between different events and components in the log sequence. Then, in order to avoid noise interference and conduct cross-system semantic analysis, we employ BERT to extract log sentence vectors, and Normalizing Flow is used to optimize them for smoother node embedding. A Graph Transformer Architecture with Time Intervals (GTAT) is proposed to model heterogeneous graphs by integrating time feature, allowing for a comprehensive analyze of execution order and time anomalies. Additionally, we design a semantic weighting method and utilize a novel domain-adapted transfer learning technology to effectively transfer the heterogeneous graph features of the source system to the target system. Experimental results demonstrate that LogGT outperforms five log anomaly detection methods, achieving an average anomaly detection F1-score higher than 0.95. Moreover, the AUC value of GTAT exceeds the sequence model by more than 2.3%.

An unsupervised semantic segmentation method that combines the ImSE-Net model with SLICm superpixel optimization

ABSTRACT In the field of remote sensing, using a large amount of labeled image data to supervise the training of fully convolutional networks for the semantic segmentation of images is expensive. However, using a small amount of labeled data can lead to reduced network performance. This paper proposes an unsupervised semantic segmentation method that combines the ImSE-Net model with SLICm superpixel optimization. First, the ImSE-Net model is used to extract semantic features from the image to obtain rough semantic segmentation results. Then, the SLICm superpixel segmentation algorithm is used to segment the input image into superpixel images. Finally, an unsupervised semantic segmentation model (UGLS) is used to combine high-level abstract semantic features with detailed information on superpixels to obtain edge-optimized semantic segmentation results. Experimental results show that compared with other semantic segmentation algorithms, our method more effectively handles unbalanced areas, such as object boundaries, and achieves better segmentation results, with higher semantic consistency.

SSGAN: A Semantic Similarity-Based GAN for Small-Sample Image Augmentation

Image sample augmentation refers to strategies for increasing sample size by modifying current data or synthesizing new data based on existing data. This technique is of vital significance in enhancing the performance of downstream learning tasks in widespread small-sample scenarios. In recent years, GAN-based image augmentation methods have gained significant attention and research focus. They have achieved remarkable generation results on large-scale datasets. However, their performance tends to be unsatisfactory when applied to datasets with limited samples. Therefore, this paper proposes a semantic similarity-based small-sample image augmentation method named SSGAN. Firstly, a relatively shallow pyramid-structured GAN-based backbone network was designed, aiming to enhance the model’s feature extraction capabilities to adapt to small sample sizes. Secondly, a feature selection module based on high-dimensional semantics was designed to optimize the loss function, thereby improving the model’s learning capacity. Lastly, extensive comparative experiments and comprehensive ablation experiments were carried out on the “Flower” and “Animal” datasets. The results indicate that the proposed method outperforms other classical GANs methods in well-established evaluation metrics such as FID and IS, with improvements of 18.6 and 1.4, respectively. The dataset augmented by SSGAN significantly enhances the performance of the classifier, achieving a 2.2% accuracy improvement compared to the best-known method. Furthermore, SSGAN demonstrates excellent generalization and robustness.

Building facade structure extraction method based on three-dimensional laser point cloud by considering semantic information

ABSTRACT Building facade structures serve as important data support for three-dimensional (3D) building modelling. A building facade structure extraction method based on 3D Laser Point Cloud Data (LPCD) by considering semantic information is proposed. The proposed method mainly involves three steps. First, an improved 3D LPCD semantic segmentation method is introduced to extract and label point clouds as the building structure and walls. Second, structure line segments are recognized from the initial building facade structure based on the Regulated Block RANSAC (RB-RANSAC) algorithm. Third, geometric information such as distance, vector and positional relationship between structure line segments is considered to optimize the initial building facade structure. The proposed method can effectively extract and fit segment structures by considering the semantic and geometric information, obtaining significantly accurate and concise building facade structures. Case studies on 3D LPCD of local buildings and open-source datasets (Semantic3D) prove the extraction accuracy and efficiency of the proposed method.

Distraction-aware hierarchical learning for vascular structure segmentation in intravascular ultrasound images

Vascular structure segmentation in intravascular ultrasound (IVUS) images plays an important role in pre-procedural evaluation of percutaneous coronary intervention (PCI). However, vascular structure segmentation in IVUS images has the challenge of structure-dependent distractions. Structure-dependent distractions are categorized into two cases, structural intrinsic distractions and inter-structural distractions. Traditional machine learning methods often rely solely on low-level features, overlooking high-level features. This way limits the generalization of these methods. The existing semantic segmentation methods integrate low-level and high-level features to enhance generalization performance. But these methods also introduce additional interference, which is harmful to solving structural intrinsic distractions. Distraction cue methods attempt to address structural intrinsic distractions by removing interference from the features through a unique decoder. However, they tend to overlook the problem of inter-structural distractions. In this paper, we propose distraction-aware hierarchical learning (DHL) for vascular structure segmentation in IVUS images. Inspired by distraction cue methods for removing interference in a decoder, the DHL is designed as a hierarchical decoder that gradually removes structure-dependent distractions. The DHL includes global perception process, distraction perception process and structural perception process. The global perception process and distraction perception process remove structural intrinsic distractions then the structural perception process removes inter-structural distractions. In the global perception process, the DHL searches for the coarse structural region of the vascular structures on the slice of IVUS sequence. In the distraction perception process, the DHL progressively refines the coarse structural region of the vascular structures to remove structural distractions. In the structural perception process, the DHL detects regions of inter-structural distractions in fused structure features then separates them. Extensive experiments on 361 subjects show that the DHL is effective (e.g., the average Dice is greater than 0.95), and superior to ten state-of-the-art IVUS vascular structure segmentation methods.

Forest Fire Smoke Detection Based on Multiple Color Spaces Deep Feature Fusion

The drastic increase of forest fire occurrence, which in recent years has posed severe threat and damage worldwide to the natural environment and human society, necessitates smoke detection of the early forest fire. First, a semantic segmentation method based on multiple color spaces feature fusion is put forward for forest fire smoke detection. Considering that smoke images in different color spaces may contain varied and distinctive smoke features which are beneficial for improving the detection ability of a model, the proposed model integrates the function of multi-scale and multi-type self-adaptive weighted feature fusion with attention augmentation to extract the enriched and complementary fused features of smoke, utilizing smoke images from multi-color spaces as inputs. Second, the model is trained and evaluated on part of the FIgLib dataset containing high-quality smoke images from watchtowers in the forests, incorporating various smoke types and complex background conditions, with a satisfactory smoke segmentation result for forest fire detection. Finally, the optimal color space combination and the fusion strategy for the model is determined through elaborate and extensive experiments with a superior segmentation result of 86.14 IoU of smoke obtained.

PSTNet: Transformer for aggregating neighborhood features in 3D point cloud semantic segmentation of eggplant plants

Improving the quality of plant point cloud data and achieving precise segmentation is essential for effective plant phenotyping analysis and plant breeding. To ensure data quality and accuracy of segmentation, we employed a self-developed three-dimensional scanning device based on binocular vision to acquire eggplant plant point clouds with high spatial resolution. And we improved down-sampling algorithm Octree Farthest Point Sampling (OFPS) to process the original point cloud data. Furthermore, we proposed a neural network called Plant Segmentation Transformer Network (PSTNet) to achieve semantic segmentation of high spatial resolution eggplant plant datasets. PSTNet comprises the following components: (i) a Neighborhood Feature Aggregator (NPA) for storing and aggregating local neighborhood features of input points, and (ii) a cascaded Point Self-Attention module (PSA) for capturing contextual information. Experimental results demonstrate the remarkable performance of PSTNet in semantic segmentation tasks, with IoU, Precision, Recall, F1-score, and Accuracy values of 92.20 %, 95.30 %, 95.57 %, 95.43 %, and 95.15 %, respectively. Additionally, compared to the Point Cloud Transformer (PCT), which exhibited the second-best performance, PSTNet achieved improvements of 4.37, 1.47, 2.87, 2.17, and 4.22 percentage points in the abovementioned metrics. This method achieves high-precision segmentation of plant point clouds, expands the semantic segmentation method of plants, and lays a solid foundation for plant phenotypic analysis.

Comparing semantic representation methods for keyword analysis in bibliometric research

Semantic representation methods play a crucial role in text mining tasks. Although numerous approaches have been proposed and compared in text mining research, the comparison of semantic representation methods specifically for publication keywords in bibliometric studies has received limited attention. This lack of practical evidence makes it challenging for researchers to select suitable methods to obtain keyword vectors for downstream bibliometric tasks, potentially hindering the achievement of optimal results. To address this gap, this study conducts an experimental comparison of various typical semantic representation methods for keywords, aiming to provide quantitative evidence for bibliometric studies. The experiment focuses on keyword clustering as the fundamental task and evaluates 22 variations of five typical methods across four scientific domains. The methods compared are co-word matrix, co-word network, word embedding, network embedding, and “semantic + structure” integration. The comparison is based on fitting the clustering results of these methods with the “evaluation standard” specific to each domain. The empirical findings demonstrate that the co-word matrix exhibits subpar performance, whereas the co-word network and word embedding techniques display satisfactory performance. Among the five network embedding algorithms, LINE and Node2Vec outperform DeepWalk, Struc2Vec, and SDNE. Remarkably, both the “pre-training and fine-tuning” model and the “semantic + structure” model yield unsatisfactory results in terms of performance. Nevertheless, even with variations in the performance of these methods, no singular approach stands out as universally superior. When selecting methods in practical applications, comprehensive consideration of factors such as corpus size and semantic cohesion of domain keywords is crucial. This study advances our understanding of semantic representation methods for keyword analysis and contributes to the advancement of bibliometric analysis by providing valuable recommendations for researchers in selecting appropriate methods.

Browsing target extraction and spatiotemporal preference mining from the complex virtual trajectories

Public Map Service Platforms (PMSPs) aggregate and disseminate the earth observation data. Leveraging spatiotemporal preference patterns derived from browsing targets within complex virtual trajectories on PMSPs aids in constructing user-profiles and comprehending their intentions. However, complex virtual trajectories, characterized by numerous trajectory points and overlapping pyramidal spatial structures, introduce inefficiencies and inaccuracies during browsing target extraction. To mitigate this, we propose an Optimized Spatial Structure Segmentation (OSSS) method that divides complex virtual trajectories into sub-trajectories with simplified spatial structures, enhancing the efficiency of browsing target extraction. Spatiotemporal reconstruction of these sub-trajectories establishes sequences of browsing targets, revealing patterns of interest transitions. Moreover, recognizing the spatial uncertainty inherent in complex virtual trajectories, which results in imprecise matching between browsing targets and spatial features, we introduce a spatial co-occurrence semantic modeling approach. This involves constructing a POI semantic space and introducing a semantic similarity matching method to reduce spatial uncertainty and refine the accuracy of mining spatiotemporal preference patterns. We evaluate these methods using real-world data from Tianditu. Results demonstrate that the OSSS improves extraction efficiency by 3.45 times and accuracy by 18.84%. Additionally, the semantic similarity approach combined with spatial co-occurrence effectively mitigates spatial uncertainty. This research contributes to advancing the intelligence of PMSPs.

EEG analysis in patients with schizophrenia based on microstate semantic modeling method.

Microstate analysis enables the characterization of quasi-stable scalp potential fields on a sub-second timescale, preserving the temporal dynamics of EEG and spatial information of scalp potential distributions. Owing to its capacity to provide comprehensive pathological insights, it has been widely applied in the investigation of schizophrenia (SCZ). Nevertheless, previous research has primarily concentrated on differences in individual microstate temporal characteristics, neglecting potential distinctions in microstate semantic sequences and not fully considering the issue of the universality of microstate templates between SCZ patients and healthy individuals. This study introduced a microstate semantic modeling analysis method aimed at schizophrenia recognition. Firstly, microstate templates corresponding to both SCZ patients and healthy individuals were extracted from resting-state EEG data. The introduction of a dual-template strategy makes a difference in the quality of microstate sequences. Quality features of microstate sequences were then extracted from four dimensions: Correlation, Explanation, Residual, and Dispersion. Subsequently, the concept of microstate semantic features was proposed, decomposing the microstate sequence into continuous sub-sequences. Specific semantic sub-sequences were identified by comparing the time parameters of sub-sequences. The SCZ recognition test was performed on the public dataset for both the quality features and semantic features of microstate sequences, yielding an impressive accuracy of 97.2%. Furthermore, cross-subject experimental validation was conducted, demonstrating that the method proposed in this paper achieves a recognition rate of 96.4% between different subjects. This research offers valuable insights for the clinical diagnosis of schizophrenia. In the future, further studies will seek to augment the sample size to enhance the effectiveness and reliability of this method.

A Fibrational Tale of Operational Logical Relations: Pure, Effectful and Differential

Logical relations built on top of an operational semantics are one of the most successful proof methods in programming language semantics. In recent years, more and more expressive notions of operationally-based logical relations have been designed and applied to specific families of languages. However, a unifying abstract framework for operationally-based logical relations is still missing. We show how fibrations can provide a uniform treatment of operational logical relations, using as reference example a lambda-calculus with generic effects endowed with a novel, abstract operational semantics defined on a large class of categories. Moreover, this abstract perspective allows us to give a solid mathematical ground also to differential logical relations -- a recently introduced notion of higher-order distance between programs -- both pure and effectful, bringing them back to a common picture with traditional ones.

Learning topic emotion and logical semantic for video paragraph captioning

Video paragraph captioning aims to generate multiple descriptive sentences for videos, which strive to replicate human writing in accuracy, logicality, and richness. However, current research focuses on the accuracy and temporal order of events, ignoring emotion and other critical logical relations embedded in human language, such as causal and adversative relations. The ignorance impairs the reasonable transition across generated event descriptions and restricts the vividness of expression, resulting in a gap from the standard of human language. To resolve the problem, a framework that integrates logic and emotion representation learning is proposed to narrow the gap. Concretely, a large-scale inter-event relation corpus is constructed based on the EMVPC dataset. This corpus is named EMVPC-EvtRel (standing for “EMVPC-Event Relations”) and contains six widely-used logical relations in human writing, 127 explicit inter-sentence connectives, and over 20,000 pairs of event segments with newly annotated logical relations. A logical semantic representation learning method is developed for recognizing the dependencies between visual events, thereby enhancing the characteristics of video contents and boosting the logicality of generated paragraphs. Moreover, a fine-grained emotion recognition module is designed to uncover emotion features embedded in videos. Finally, experimental results on the EMVPC dataset demonstrate the superiority of the proposed method compared to existing state-of-the-art approaches.

Efficient multi-branch segmentation network for situation awareness in autonomous navigation

Real-time and high-precision situational awareness technology is critical for autonomous navigation of unmanned surface vehicles (USVs). In particular, robust and fast obstacle semantic segmentation methods are essential. However, distinguishing between the sea and the sky is challenging due to the differences between port and maritime environments. In this study, we built a dataset that captured perspectives from USVs and unmanned aerial vehicles in a maritime port environment and analysed the data features. Statistical analysis revealed a high correlation between the distribution of the sea and sky and row positional information. Based on this finding, a three-branch semantic segmentation network with a row position encoding module (RPEM) was proposed to improve the prediction accuracy between the sea and the sky. The proposed RPEM highlights the effect of row coordinates on feature extraction. Compared to the baseline, the three-branch network with RPEM significantly improved the ability to distinguish between the sea and the sky without significantly reducing the computational speed.

Ice-Core Micro-CT Image Segmentation with Dual Stream Spectrum Deconvolution Neural Network and Gaussian Mixture Model

Polar ice sheets, or ice cores, are among the most well-known natural archives that might provide crucial historical details about our planet's previous environment. An important factor in establishing the fundamental characteristics of ice, likes pore close-off, albedo, melt events, is the ice-core microstructure. To engulf these complications Ice-Core Micro-CT Image Segmentation with Dual Stream Spectrum Deconvolution Neural Network and Gaussian Mixture Model (ICMCTS-WSOA-DSSDNN-GMM)is proposed. Initially, micro scale CT images are collect from Alfred Wegener Institute ice-core storage as input. Then,data’s are given to pre-processing. In pre-processing, it enhance image brightness, remove salt pepper noise, crop outer ring (carbon fiber casing) to have only ice particles in image utilizing Federated Neural Collaborative Filtering (FNCF).The pre-processing output is given to segmentation. Here, micro scale CT image is utilized for segmenting high-resolution scans using Gaussian Mixture Model (GMM). After that, the segmented images are given to Dual Stream Spectrum Deconvolution Neural Network optimized with Water Strider Optimization Algorithm for classifying micro scale CT images as sintered snow, compacted firn and bubbly ice. The proposed ICMCTS-WSOA-DSSDNN-GMM method is executed in python. The performance of ICMCTS-WSOA-DSSDNN-GMM approach attains 16.24%, 17.90% and 27.7% high accuracy, 14.04%, 25.51% and 19.31% higher precision and 14.36%, 12.65%, 14.51% higher recall analyzed with existing techniques likes Ice-Core Micro-CT Image Segmentation With Deep Learning and Gaussian Mixture Method (ICMCTS-U-net-GMM), Computer Aided Detection of COVID 19 from CT Images Depend on Gaussian Mixture Method with Kernel Support Vector Machines Classifer (ICMCTS-KNN-GMM) and the Gmmseg: Gaussian mixture based generative semantic segmentation methods(ICMCTS-FCN-GMM) respectively.

Fused Attentive Generative Adversarial Network based Personalized English Learning Material Recommendation System utilizing Knowledge Graph

The information era has arrived swiftly in the world. Because it is the universal language, English influences how information is transmitted and exchanged throughout the world. Learning English is like having a valuable instrument for learning priceless knowledge. For this reason, it is imperative that English instruction be improved. The Fused Attentive Generative Adversarial Network based Personalized English Learning Material Recommendation System utilizing Knowledge Graph. Initially, input data are collected, given to preprocessing method. In preprocessing remove data redundancy using Generalized Correntropy Sparse Gauss–Hermite Quadrature Filter (GCSG-HQF). In pre-processed output fed to Recommendation System. Here, FAGAN is used to recommend English Learning Material. In general, FAGAN classifier does not express any optimization adaption approaches for determining optimum parameters to assure the Knowledge Graph. Here Clouded Leopard Optimization Algorithm utilized to optimize FAGAN classifier, for Personalized English Learning Material Recommendation System. In Clouded Leopard Optimization Algorithm utilized to optimise weight parameter of the FAGAN. With the use of performance metrics likes accuracy, precision, sensitivity, specificity, F1-score, ROC, computational time are analyzed. The proposed FAGAN-ELMRS-CLOA method attains 30.58%, 28.73% and 35.62%, higher accuracy, 20.48%, 24.73%, 29.32% higher computational time and 30.98%, 26.66% and 21.32% higher ROC analysed with existing models such as personalized English learning material recommendation system depend on knowledge graph (CNN-ELMRS-KG), semantic method for document classification utilizing deep neural networks with multimedia knowledge graph (DNN- ELMRS-MKG), and personalized recommendation of English learning depend on knowledge graph with graph convolutional network (PR- ELMRS-GCN) respectively.

A hybrid deep semantic mining method considering fuzzy expressions for the automatic recognition of construction safety hazard information

Safety hazards are a key consideration in construction management. The efficient recognition of safety hazard information can help managers formulate safety hazard management measures and improve the efficiency of construction safety management. However, construction site safety hazard data are stored in semistructured and unstructured text formats, which cannot be directly converted into understandable and usable information. Moreover, safety hazard text contains many fuzzy expressions, thereby increasing the difficulty of text semantic analysis; thus, how to accurately mine safety hazard information from complex and diverse text data is an urgent problem that must be solved. In consideration of this problem, we propose a bidirectional long short-term memory (BiLSTM) method with a fuzzy word vector and self-attention mechanism (FSABiLSTM) to automatically recognize safety hazard information. This method adopts TextRank and Word2vec to calculate the fuzzy word vector and process fuzzy expressions in safety hazard text. The safety hazard text semantic features are deeply extracted based on BiLSTM and a fuzzy word vector, and the extracted semantic features are analyzed via a self-attention mechanism. Actual construction safety hazard text is used to verify the reliability and applicability of the method, and the results indicate that the accuracy of this method, which outperforms existing machine learning methods, is 91.70%. In addition, the FSABiLSTM method can be used to automatically evaluate the risk degree of safety hazards; this use is beneficial to managing and controlling safety hazards. Concerning safety hazard text data, this study provides a new deep mining approach that can enhance safety management efficiency.