Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval sy...Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval system mainly relies on the efficiency and accuracy of the classification models.This research addresses the challenge of enhancing the image retrieval system by developing a novel approach,EfficientNet-Convolutional Neural Network(EffNet-CNN).The key objective of this research is to evaluate the proposed EffNet-CNN model’s performance in image classification,image mining,and CBIR.The novelty of the proposed EffNet-CNN model includes the integration of different techniques and modifications.The model includes the Mahalanobis distance metric for feature matching,which enhances the similarity measurements.The model extends EfficientNet architecture by incorporating additional convolutional layers,batch normalization,dropout,and pooling layers for improved hierarchical feature extraction.A systematic hyperparameter optimization using SGD,performance evaluation with three datasets,and data normalization for improving feature representations.The EffNet-CNN is assessed utilizing precision,accuracy,F-measure,and recall metrics across MS-COCO,CIFAR-10 and 100 datasets.The model achieved accuracy values ranging from 90.60%to 95.90%for the MS-COCO dataset,96.8%to 98.3%for the CIFAR-10 dataset and 92.9%to 98.6%for the CIFAR-100 dataset.A validation of the EffNet-CNN model’s results with other models reveals the proposed model’s superior performance.The results highlight the potential of the EffNet-CNN model proposed for image classification and its usefulness in image mining and CBIR.展开更多
An association rules mining method based on semantic relativity is proposed to solve the problem that there are more candidate item sets and higher time complexity in traditional association rules mining.Semantic rela...An association rules mining method based on semantic relativity is proposed to solve the problem that there are more candidate item sets and higher time complexity in traditional association rules mining.Semantic relativity of ontology concepts is used to describe complicated relationships of domains in the method.Candidate item sets with less semantic relativity are filtered to reduce the number of candidate item sets in association rules mining.An ontology hierarchy relationship is regarded as a directed acyclic graph rather than a hierarchy tree in the semantic relativity computation.Not only direct hierarchy relationships,but also non-direct hierarchy relationships and other typical semantic relationships are taken into account.Experimental results show that the proposed method can reduce the number of candidate item sets effectively and improve the efficiency of association rules mining.展开更多
Computational techniques have been adopted in medi-cal and biological systems for a long time. There is no doubt that the development and application of computational methods will render great help in better understan...Computational techniques have been adopted in medi-cal and biological systems for a long time. There is no doubt that the development and application of computational methods will render great help in better understanding biomedical and biological functions. Large amounts of datasets have been produced by biomedical and biological experiments and simulations. In order for researchers to gain knowledge from origi- nal data, nontrivial transformation is necessary, which is regarded as a critical link in the chain of knowledge acquisition, sharing, and reuse. Challenges that have been encountered include: how to efficiently and effectively represent human knowledge in formal computing models, how to take advantage of semantic text mining techniques rather than traditional syntactic text mining, and how to handle security issues during the knowledge sharing and reuse. This paper summarizes the state-of-the-art in these research directions. We aim to provide readers with an introduction of major computing themes to be applied to the medical and biological research.展开更多
In this paper we propose a novel model ＂recursive directed graph＂ based on feature structure, and apply it to represent the semantic relations of postpositive attributive structures in biomedical texts. The usages o...In this paper we propose a novel model ＂recursive directed graph＂ based on feature structure, and apply it to represent the semantic relations of postpositive attributive structures in biomedical texts. The usages of postpositive attributive are complex and variable, especially three categories： present participle phrase, past participle phrase, and preposition phrase as postpositire attributive, which always bring the difficulties of automatic parsing. We summarize these categories and annotate the semantic information. Compared with dependency structure, feature structure, being recursive directed graph, enhances semantic information extraction in biomedical field. The annotation results show that recursive directed graph is more suitable to extract complex semantic relations for biomedical text mining.展开更多
This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are tr...This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are trained and constructed with STTS （Semantic Template Training System）, which are taken as the bridge to realize the transition from various low-level media feature to object semantic. Furthermore, we put forward a kind of double layers metadata structure to efficaciously store and manage mined low-level feature and high-level semantic. This model has broad application in lots of domains such as intelligent retrieval engine, medical diagnoses, multimedia design and so on.展开更多
The integration of the two fast-developing scientific research areas Semantic Web and Web Mining is known as Semantic Web Mining. The huge increase in the amount of Semantic Web data became a perfect target for many r...The integration of the two fast-developing scientific research areas Semantic Web and Web Mining is known as Semantic Web Mining. The huge increase in the amount of Semantic Web data became a perfect target for many researchers to apply Data Mining techniques on it. This paper gives a detailed state-of-the-art survey of on-going research in this new area. It shows the positive effects of Semantic Web Mining, the obstacles faced by researchers and propose number of approaches to deal with the very complex and heterogeneous information and knowledge which are produced by the technologies of Semantic Web.展开更多
Based on the definition of component ontology, an effective component classification mechanism and a facet named component relationship are proposed. Then an application domain oriented, hierarchical component organiz...Based on the definition of component ontology, an effective component classification mechanism and a facet named component relationship are proposed. Then an application domain oriented, hierarchical component organization model is established. At last a hierarchical component semantic network (HCSN) described by ontology interchange language(OIL) is presented and then its function is described. Using HCSN and cooperating with other components retrieving algorithms based on component description, other components information and their assembly or composite modes related to the key component can be found. Based on HCSN, component directory library is catalogued and a prototype system is constructed. The prototype system proves that component library organization based on this model gives guarantee to the reliability of component assembly during program mining.展开更多
In this paper, a finite state machine approach is followed in order to find the semantic similarity of two sentences. The approach exploits the concept of bi-directional logic along with a semantic ordering approach. ...In this paper, a finite state machine approach is followed in order to find the semantic similarity of two sentences. The approach exploits the concept of bi-directional logic along with a semantic ordering approach. The core part of this approach is bi-directional logic of artificial intelligence. The bi-directional logic is implemented using Finite State Machine algorithm with slight modification. For finding the semantic similarity, keyword has played climactic importance. With the help of the keyword approach, it can be found easily at the sentence level according to this algorithm. The algorithm is proposed especially for Nepali texts. With the polarity of the individual keywords, the finite state machine is made and its final state determines its polarity. If two sentences are negatively polarized, they are said to be coherent, otherwise not. Similarly, if two sentences are of a positive nature, they are said to be coherence. For measuring the coherence (similarity), contextual concept is taken into consideration. The semantic approach, in this research, is a totally contextual based method. Two sentences are said to be semantically similar if they bear the same context. The total accuracy obtained in this algorithm is 90.16%.展开更多
Efficient preparation and assembly guidance for complex products relies heavily on semantic information in assembly process documents.This information encompasses various levels of elements and complex semantic relati...Efficient preparation and assembly guidance for complex products relies heavily on semantic information in assembly process documents.This information encompasses various levels of elements and complex semantic relationships.However,there is currently a scarcity of effective modeling techniques to express these documents'inherent assembly process knowledge.This study introduces a method for constructing an Assembly Process Knowledge Graph of Complex Products(APKG-CP)utilizing text mining techniques to tackle the challenges of high costs,low efficiency,and difficulty reusing process knowledge.Developing the assembly process knowledge graph involves categorizing entity and relationship classes from multiple levels.The Bert-BiLSTM-CRF model integrates BERT(bidirectional encoder representations from transformers),BiLSTM(bidirectional long short-term memory),and CRF(conditional random field)to extract knowledge entities and relationships in assembly process documents automatically.Furthermore,the knowledge fusion method automatically instantiates the assembly process knowledge graph.The proposed construction method is validated by constructing and visualizing an assembly process knowledge graph using data from an aerospace enterprise as an example.Integrating the knowledge graph with the assembly process preparation system demonstrates its effectiveness for process design.展开更多
长文本关系识别在科技情报与数字人文领域中具有重要作用,是实现知识重组向知识发现转变的关键。然而,由于长文本存在上下文跨度大、语义线索分散、实体指代复杂等特征,传统大语言模型(large language model,LLM)在处理该类文本时,易出...长文本关系识别在科技情报与数字人文领域中具有重要作用,是实现知识重组向知识发现转变的关键。然而,由于长文本存在上下文跨度大、语义线索分散、实体指代复杂等特征,传统大语言模型(large language model,LLM)在处理该类文本时,易出现上下文理解不足、语义偏移以及幻觉等问题,使得长文本在科技情报与人文计算等领域的实际应用中尚未更好地实现内容增值。为了解决上述问题,首先,本文依据关系触发词的聚类结果构建实体关系体系;其次,针对长文本特征,设计基于LLM-BERT(large language model-bidirectional encoder representations from transformers)协同框架的长文本关系识别算法,提升语义关联性;再其次,融合预训练模型、深度学习网络、注意力机制处理文本特征的优势,构建BERT-CNN-BiLSTM-MHA(BCBM)模型,深层次挖掘文本语义;最后,结合模型置信度和文本相似度,设计一种摘要质量评估机制,以缓解LLM幻觉。研究结果表明,该算法实测效果优于传统模型,能在一定程度上缓解LLM在处理长文本时易产生的上下文理解不足、语义偏移和幻觉等问题。展开更多
In the era of information technology,recommendation systems play a crucial role in information filtering and user preference identification.Notably,the auxiliary information provided by online social platforms offers ...In the era of information technology,recommendation systems play a crucial role in information filtering and user preference identification.Notably,the auxiliary information provided by online social platforms offers significant support for enhancing the performance of recommendation systems.Based on the hypothesis that socially connected users share similar preferences,inte-grating social relationships as supplementary information into recommendation algorithms can significantly enhance recommendation accuracy while effectively mitigating the cold-start prob-lem.However,existing social recommendation systems primarily rely on explicit social relation-ships as auxiliary information,often overlooking the value of potential social connections.Research indicates that users with potential social links may also possess valuable preference information.We believe that mining potential social relationships can provide valuable auxiliary information,thereby enhancing the performance of recommendation systems.To address this issue,we propose a social recommendation model based on social semantic mining and denoising(SSMD).Specifically,we propose an encoder-decoder architecture to learn explicit social user representations and mine potential social relationships.Considering the potential noise in these implicit connections,we design a denoising module that utilizes user preference information to filter unreliable social links.Furthermore,we implement cross-view information alignment be-tween the potential social graph and interaction graph through an auxiliary loss function.Extensive experiments conducted on multiple public datasets demonstrate that our SSMD method outperforms various baseline approaches with significant improvements.展开更多
基金The authors extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid University,Kingdom of Saudi Arabia,for funding this work through the Small Research Group Project under Grant Number RGP.1/316/45.
文摘Content-Based Image Retrieval(CBIR)and image mining are becoming more important study fields in computer vision due to their wide range of applications in healthcare,security,and various domains.The image retrieval system mainly relies on the efficiency and accuracy of the classification models.This research addresses the challenge of enhancing the image retrieval system by developing a novel approach,EfficientNet-Convolutional Neural Network(EffNet-CNN).The key objective of this research is to evaluate the proposed EffNet-CNN model’s performance in image classification,image mining,and CBIR.The novelty of the proposed EffNet-CNN model includes the integration of different techniques and modifications.The model includes the Mahalanobis distance metric for feature matching,which enhances the similarity measurements.The model extends EfficientNet architecture by incorporating additional convolutional layers,batch normalization,dropout,and pooling layers for improved hierarchical feature extraction.A systematic hyperparameter optimization using SGD,performance evaluation with three datasets,and data normalization for improving feature representations.The EffNet-CNN is assessed utilizing precision,accuracy,F-measure,and recall metrics across MS-COCO,CIFAR-10 and 100 datasets.The model achieved accuracy values ranging from 90.60%to 95.90%for the MS-COCO dataset,96.8%to 98.3%for the CIFAR-10 dataset and 92.9%to 98.6%for the CIFAR-100 dataset.A validation of the EffNet-CNN model’s results with other models reveals the proposed model’s superior performance.The results highlight the potential of the EffNet-CNN model proposed for image classification and its usefulness in image mining and CBIR.
基金The National Natural Science Foundation of China(No.50674086)Specialized Research Fund for the Doctoral Program of Higher Education(No.20060290508)the Science and Technology Fund of China University of Mining and Technology(No.2007B016)
文摘An association rules mining method based on semantic relativity is proposed to solve the problem that there are more candidate item sets and higher time complexity in traditional association rules mining.Semantic relativity of ontology concepts is used to describe complicated relationships of domains in the method.Candidate item sets with less semantic relativity are filtered to reduce the number of candidate item sets in association rules mining.An ontology hierarchy relationship is regarded as a directed acyclic graph rather than a hierarchy tree in the semantic relativity computation.Not only direct hierarchy relationships,but also non-direct hierarchy relationships and other typical semantic relationships are taken into account.Experimental results show that the proposed method can reduce the number of candidate item sets effectively and improve the efficiency of association rules mining.
文摘Computational techniques have been adopted in medi-cal and biological systems for a long time. There is no doubt that the development and application of computational methods will render great help in better understanding biomedical and biological functions. Large amounts of datasets have been produced by biomedical and biological experiments and simulations. In order for researchers to gain knowledge from origi- nal data, nontrivial transformation is necessary, which is regarded as a critical link in the chain of knowledge acquisition, sharing, and reuse. Challenges that have been encountered include: how to efficiently and effectively represent human knowledge in formal computing models, how to take advantage of semantic text mining techniques rather than traditional syntactic text mining, and how to handle security issues during the knowledge sharing and reuse. This paper summarizes the state-of-the-art in these research directions. We aim to provide readers with an introduction of major computing themes to be applied to the medical and biological research.
基金Supported by the National Natural Science Foundation of China(61202193,61202304)the Major Projects of Chinese National Social Science Foundation(11&ZD189)the Chinese Postdoctoral Science Foundation(2013M540593,2014T70722)
文摘In this paper we propose a novel model ＂recursive directed graph＂ based on feature structure, and apply it to represent the semantic relations of postpositive attributive structures in biomedical texts. The usages of postpositive attributive are complex and variable, especially three categories： present participle phrase, past participle phrase, and preposition phrase as postpositire attributive, which always bring the difficulties of automatic parsing. We summarize these categories and annotate the semantic information. Compared with dependency structure, feature structure, being recursive directed graph, enhances semantic information extraction in biomedical field. The annotation results show that recursive directed graph is more suitable to extract complex semantic relations for biomedical text mining.
基金Supported by the National Basic Research Program of China 973 Program (2007CB310801)the Specialized Research Fund for the Doctoral Program of Higer Education of China (20070486064)+1 种基金the Natural Science Foundation of Hubei Province (2007ABA038)the Programme of Introducing Talents of Discipline to Universities (B07037)
文摘This paper presents a cross-media semantic mining model （CSMM） based on object semantic. This model obtains object-level semantic information in terms of maximum probability principle. Then semantic templates are trained and constructed with STTS （Semantic Template Training System）, which are taken as the bridge to realize the transition from various low-level media feature to object semantic. Furthermore, we put forward a kind of double layers metadata structure to efficaciously store and manage mined low-level feature and high-level semantic. This model has broad application in lots of domains such as intelligent retrieval engine, medical diagnoses, multimedia design and so on.
文摘The integration of the two fast-developing scientific research areas Semantic Web and Web Mining is known as Semantic Web Mining. The huge increase in the amount of Semantic Web data became a perfect target for many researchers to apply Data Mining techniques on it. This paper gives a detailed state-of-the-art survey of on-going research in this new area. It shows the positive effects of Semantic Web Mining, the obstacles faced by researchers and propose number of approaches to deal with the very complex and heterogeneous information and knowledge which are produced by the technologies of Semantic Web.
文摘Based on the definition of component ontology, an effective component classification mechanism and a facet named component relationship are proposed. Then an application domain oriented, hierarchical component organization model is established. At last a hierarchical component semantic network (HCSN) described by ontology interchange language(OIL) is presented and then its function is described. Using HCSN and cooperating with other components retrieving algorithms based on component description, other components information and their assembly or composite modes related to the key component can be found. Based on HCSN, component directory library is catalogued and a prototype system is constructed. The prototype system proves that component library organization based on this model gives guarantee to the reliability of component assembly during program mining.
文摘In this paper, a finite state machine approach is followed in order to find the semantic similarity of two sentences. The approach exploits the concept of bi-directional logic along with a semantic ordering approach. The core part of this approach is bi-directional logic of artificial intelligence. The bi-directional logic is implemented using Finite State Machine algorithm with slight modification. For finding the semantic similarity, keyword has played climactic importance. With the help of the keyword approach, it can be found easily at the sentence level according to this algorithm. The algorithm is proposed especially for Nepali texts. With the polarity of the individual keywords, the finite state machine is made and its final state determines its polarity. If two sentences are negatively polarized, they are said to be coherent, otherwise not. Similarly, if two sentences are of a positive nature, they are said to be coherence. For measuring the coherence (similarity), contextual concept is taken into consideration. The semantic approach, in this research, is a totally contextual based method. Two sentences are said to be semantically similar if they bear the same context. The total accuracy obtained in this algorithm is 90.16%.
基金Supported by National Natural Science Foundation of China(Grant No.52375479)。
文摘Efficient preparation and assembly guidance for complex products relies heavily on semantic information in assembly process documents.This information encompasses various levels of elements and complex semantic relationships.However,there is currently a scarcity of effective modeling techniques to express these documents'inherent assembly process knowledge.This study introduces a method for constructing an Assembly Process Knowledge Graph of Complex Products(APKG-CP)utilizing text mining techniques to tackle the challenges of high costs,low efficiency,and difficulty reusing process knowledge.Developing the assembly process knowledge graph involves categorizing entity and relationship classes from multiple levels.The Bert-BiLSTM-CRF model integrates BERT(bidirectional encoder representations from transformers),BiLSTM(bidirectional long short-term memory),and CRF(conditional random field)to extract knowledge entities and relationships in assembly process documents automatically.Furthermore,the knowledge fusion method automatically instantiates the assembly process knowledge graph.The proposed construction method is validated by constructing and visualizing an assembly process knowledge graph using data from an aerospace enterprise as an example.Integrating the knowledge graph with the assembly process preparation system demonstrates its effectiveness for process design.
文摘长文本关系识别在科技情报与数字人文领域中具有重要作用,是实现知识重组向知识发现转变的关键。然而,由于长文本存在上下文跨度大、语义线索分散、实体指代复杂等特征,传统大语言模型(large language model,LLM)在处理该类文本时,易出现上下文理解不足、语义偏移以及幻觉等问题,使得长文本在科技情报与人文计算等领域的实际应用中尚未更好地实现内容增值。为了解决上述问题,首先,本文依据关系触发词的聚类结果构建实体关系体系;其次,针对长文本特征,设计基于LLM-BERT(large language model-bidirectional encoder representations from transformers)协同框架的长文本关系识别算法,提升语义关联性;再其次,融合预训练模型、深度学习网络、注意力机制处理文本特征的优势,构建BERT-CNN-BiLSTM-MHA(BCBM)模型,深层次挖掘文本语义;最后,结合模型置信度和文本相似度,设计一种摘要质量评估机制,以缓解LLM幻觉。研究结果表明,该算法实测效果优于传统模型,能在一定程度上缓解LLM在处理长文本时易产生的上下文理解不足、语义偏移和幻觉等问题。
基金supported by the National Natural Science Foundation of China(62077038,61672405,62176196 and 62271374).
文摘In the era of information technology,recommendation systems play a crucial role in information filtering and user preference identification.Notably,the auxiliary information provided by online social platforms offers significant support for enhancing the performance of recommendation systems.Based on the hypothesis that socially connected users share similar preferences,inte-grating social relationships as supplementary information into recommendation algorithms can significantly enhance recommendation accuracy while effectively mitigating the cold-start prob-lem.However,existing social recommendation systems primarily rely on explicit social relation-ships as auxiliary information,often overlooking the value of potential social connections.Research indicates that users with potential social links may also possess valuable preference information.We believe that mining potential social relationships can provide valuable auxiliary information,thereby enhancing the performance of recommendation systems.To address this issue,we propose a social recommendation model based on social semantic mining and denoising(SSMD).Specifically,we propose an encoder-decoder architecture to learn explicit social user representations and mine potential social relationships.Considering the potential noise in these implicit connections,we design a denoising module that utilizes user preference information to filter unreliable social links.Furthermore,we implement cross-view information alignment be-tween the potential social graph and interaction graph through an auxiliary loss function.Extensive experiments conducted on multiple public datasets demonstrate that our SSMD method outperforms various baseline approaches with significant improvements.
