A large semantic gap exists between content based index retrieval(CBIR) and high-level semantic,additional semantic information should be attached to the images,it refers in three respects including semantic represent...A large semantic gap exists between content based index retrieval(CBIR) and high-level semantic,additional semantic information should be attached to the images,it refers in three respects including semantic representation model,semantic information building and semantic retrieval techniques.In this paper,we introduce an associated semantic network and an automatic semantic annotation system.In the system,a semantic network model is employed as the semantic representation model,it uses semantic Key words,linguistic ontology and low-level features in semantic similarity calculating.Through several times of users' relevance feedback,semantic network is enriched automatically.To speed up the growth of semantic network and get a balance annotation,semantic seeds and semantic loners are employed especially.展开更多
In recent years, more and more foreigners begin to learn Chinese characters, but they often make typos when using Chinese. The fundamental reason is that they mainly learn Chinese characters from the glyph and pronunc...In recent years, more and more foreigners begin to learn Chinese characters, but they often make typos when using Chinese. The fundamental reason is that they mainly learn Chinese characters from the glyph and pronunciation, but do not master the semantics of Chinese characters. If they can understand the meaning of Chinese characters and form knowledge groups of the characters with relevant meanings, it can effectively improve learning efficiency. We achieve this goal by building a Chinese character semantic knowledge graph (CCSKG). In the process of building the knowledge graph, the semantic computing capacity of HowNet was utilized, and 104,187 associated edges were finally established for 6752 Chinese characters. Thanks to the development of deep learning, OpenHowNet releases the core data of HowNet and provides useful APIs for calculating the similarity between two words based on sememes. Therefore our method combines the advantages of data-driven and knowledge-driven. The proposed method treats Chinese sentences as subgraphs of the CCSKG and uses graph algorithms to correct Chinese typos and achieve good results. The experimental results show that compared with keras-bert and pycorrector + ernie, our method reduces the false acceptance rate by 38.28% and improves the recall rate by 40.91% in the field of learning Chinese as a foreign language. The CCSKG can help to promote Chinese overseas communication and international education.展开更多
In Chinese question answering system, because there is more semantic relation in questions than that in query words, the precision can be improved by expanding query while using natural language questions to retrieve ...In Chinese question answering system, because there is more semantic relation in questions than that in query words, the precision can be improved by expanding query while using natural language questions to retrieve documents. This paper proposes a new approach to query expansion based on semantics and statistics Firstly automatic relevance feedback method is used to generate a candidate expansion word set. Then the expanded query words are selected from the set based on the semantic similarity and seman- tic relevancy between the candidate words and the original words. Experiments show the new approach is effective for Web retrieval and out-performs the conventional expansion approaches.展开更多
This paper presents a semantic metadata description model for grid services, and furthermore, a service relevance based matching algorithm is proposed. Some rea sonable parameters interpreting services characterizatio...This paper presents a semantic metadata description model for grid services, and furthermore, a service relevance based matching algorithm is proposed. Some rea sonable parameters interpreting services characterizations are considered to advise matching process, whose weightiness and service relevance dramatically contribute to the decision making of service matching degree, And what＇s more, matching is secondly executed on the bases of experience adaptation of matching factors, which improves matching degree in gird service retrieve greatly.展开更多
多模态生成式摘要往往采用序列到序列(Seq2Seq)框架,目标函数在字符级别优化模型,根据局部最优解生成单词,忽略了摘要样本全局语义信息,使得摘要与多模态信息产生语义偏差,容易造成事实性错误。针对上述问题,提出一种基于语义相关性分...多模态生成式摘要往往采用序列到序列(Seq2Seq)框架,目标函数在字符级别优化模型,根据局部最优解生成单词,忽略了摘要样本全局语义信息,使得摘要与多模态信息产生语义偏差,容易造成事实性错误。针对上述问题,提出一种基于语义相关性分析的多模态摘要模型。首先,在Seq2Seq框架基础上对多模态摘要进行训练,生成语义多样性的候选摘要;其次,构建基于语义相关性分析的摘要评估器,从全局的角度学习候选摘要之间的语义差异性和真实评价指标ROUGE(Recall-Oriented Understudy for Gisting Evaluation)的排序模式,从而在摘要样本层面优化模型;最后,不依赖参考摘要,利用摘要评估器对候选摘要进行评价,使得选出的摘要与源文本在语义空间中尽可能相似。实验结果表明,在公开数据集MMSS上,相较于MPMSE(Multimodal Pointer-generator via Multimodal Selective Encoding)模型,所提模型在ROUGE-1、ROUGE-2、ROUGE-L评价指标上分别提升了3.17、1.21和2.24个百分点。展开更多
A relevance vector machine (RVM) based fault diagnosis method was presented for non-linear circuits. In order to simplify RVM classifier, parameters selection based on particle swarm optimization (PSO) and preprocessi...A relevance vector machine (RVM) based fault diagnosis method was presented for non-linear circuits. In order to simplify RVM classifier, parameters selection based on particle swarm optimization (PSO) and preprocessing technique based on the kurtosis and entropy of signals were used. Firstly, sinusoidal inputs with different frequencies were applied to the circuit under test (CUT). Then, the resulting frequency responses were sampled to generate features. The frequency response was sampled to compute its kurtosis and entropy, which can show the information capacity of signal. By analyzing the output signals, the proposed method can detect and identify faulty components in circuits. The results indicate that the fault classes can be classified correctly for at least 99% of the test data in example circuit. And the proposed method can diagnose hard and soft faults.展开更多
近年来,随着基于位置的社交网络(Location-Based Social Network, LBSN)不断发展,POI序列推荐逐渐成为近年来研究的热点问题.现有的POI序列推荐方法仅仅按照时间的先后顺序建模用户历史签到序列,默认用户POI轨迹中连续POI之间具有相等...近年来,随着基于位置的社交网络(Location-Based Social Network, LBSN)不断发展,POI序列推荐逐渐成为近年来研究的热点问题.现有的POI序列推荐方法仅仅按照时间的先后顺序建模用户历史签到序列,默认用户POI轨迹中连续POI之间具有相等的时间间隔,忽略了用户签到记录之间的时间间隔影响.另外,POI之间的地理距离以及语义信息也是影响推荐准确性的重要因素.基于此,本文提出自注意力下时空-语义相融合的POI序列推荐模型(POI sequence recommendation model based on the integration of spatiotemporal and semantics under self-attention, SA-TDS-PRec).首先,根据用户的实际签到时间建模POI轨迹.其次,融合POI绝对位置、时空间隔以及语义相关信息.最后利用自注意力机制捕捉用户动态偏好的演化,从而提高POI推荐的准确性.在公开数据集Gowalla和Yelp上进行可扩展实验.结果表明,该模型优于目前主流的基准模型,有效提升推荐结果准确性.展开更多
文摘A large semantic gap exists between content based index retrieval(CBIR) and high-level semantic,additional semantic information should be attached to the images,it refers in three respects including semantic representation model,semantic information building and semantic retrieval techniques.In this paper,we introduce an associated semantic network and an automatic semantic annotation system.In the system,a semantic network model is employed as the semantic representation model,it uses semantic Key words,linguistic ontology and low-level features in semantic similarity calculating.Through several times of users' relevance feedback,semantic network is enriched automatically.To speed up the growth of semantic network and get a balance annotation,semantic seeds and semantic loners are employed especially.
文摘In recent years, more and more foreigners begin to learn Chinese characters, but they often make typos when using Chinese. The fundamental reason is that they mainly learn Chinese characters from the glyph and pronunciation, but do not master the semantics of Chinese characters. If they can understand the meaning of Chinese characters and form knowledge groups of the characters with relevant meanings, it can effectively improve learning efficiency. We achieve this goal by building a Chinese character semantic knowledge graph (CCSKG). In the process of building the knowledge graph, the semantic computing capacity of HowNet was utilized, and 104,187 associated edges were finally established for 6752 Chinese characters. Thanks to the development of deep learning, OpenHowNet releases the core data of HowNet and provides useful APIs for calculating the similarity between two words based on sememes. Therefore our method combines the advantages of data-driven and knowledge-driven. The proposed method treats Chinese sentences as subgraphs of the CCSKG and uses graph algorithms to correct Chinese typos and achieve good results. The experimental results show that compared with keras-bert and pycorrector + ernie, our method reduces the false acceptance rate by 38.28% and improves the recall rate by 40.91% in the field of learning Chinese as a foreign language. The CCSKG can help to promote Chinese overseas communication and international education.
基金the Specialized Research Program Fundthe Doctoral Program of Higher Education of China (20050007023)the Natural Science Foundation of Shandong Province(Y2004G04)
文摘In Chinese question answering system, because there is more semantic relation in questions than that in query words, the precision can be improved by expanding query while using natural language questions to retrieve documents. This paper proposes a new approach to query expansion based on semantics and statistics Firstly automatic relevance feedback method is used to generate a candidate expansion word set. Then the expanded query words are selected from the set based on the semantic similarity and seman- tic relevancy between the candidate words and the original words. Experiments show the new approach is effective for Web retrieval and out-performs the conventional expansion approaches.
基金Supported by Foundation of High Technology Pro-ject of JiangSu (BG2004034)
文摘This paper presents a semantic metadata description model for grid services, and furthermore, a service relevance based matching algorithm is proposed. Some rea sonable parameters interpreting services characterizations are considered to advise matching process, whose weightiness and service relevance dramatically contribute to the decision making of service matching degree, And what＇s more, matching is secondly executed on the bases of experience adaptation of matching factors, which improves matching degree in gird service retrieve greatly.
文摘多模态生成式摘要往往采用序列到序列(Seq2Seq)框架,目标函数在字符级别优化模型,根据局部最优解生成单词,忽略了摘要样本全局语义信息,使得摘要与多模态信息产生语义偏差,容易造成事实性错误。针对上述问题,提出一种基于语义相关性分析的多模态摘要模型。首先,在Seq2Seq框架基础上对多模态摘要进行训练,生成语义多样性的候选摘要;其次,构建基于语义相关性分析的摘要评估器,从全局的角度学习候选摘要之间的语义差异性和真实评价指标ROUGE(Recall-Oriented Understudy for Gisting Evaluation)的排序模式,从而在摘要样本层面优化模型;最后,不依赖参考摘要,利用摘要评估器对候选摘要进行评价,使得选出的摘要与源文本在语义空间中尽可能相似。实验结果表明,在公开数据集MMSS上,相较于MPMSE(Multimodal Pointer-generator via Multimodal Selective Encoding)模型,所提模型在ROUGE-1、ROUGE-2、ROUGE-L评价指标上分别提升了3.17、1.21和2.24个百分点。
基金Project(Z132012)supported by the Second Five Technology-based in Science and Industry Bureau of ChinaProject(YWF1103Q062)supported by the Fundemental Research Funds for the Central Universities in China
文摘A relevance vector machine (RVM) based fault diagnosis method was presented for non-linear circuits. In order to simplify RVM classifier, parameters selection based on particle swarm optimization (PSO) and preprocessing technique based on the kurtosis and entropy of signals were used. Firstly, sinusoidal inputs with different frequencies were applied to the circuit under test (CUT). Then, the resulting frequency responses were sampled to generate features. The frequency response was sampled to compute its kurtosis and entropy, which can show the information capacity of signal. By analyzing the output signals, the proposed method can detect and identify faulty components in circuits. The results indicate that the fault classes can be classified correctly for at least 99% of the test data in example circuit. And the proposed method can diagnose hard and soft faults.
文摘近年来,随着基于位置的社交网络(Location-Based Social Network, LBSN)不断发展,POI序列推荐逐渐成为近年来研究的热点问题.现有的POI序列推荐方法仅仅按照时间的先后顺序建模用户历史签到序列,默认用户POI轨迹中连续POI之间具有相等的时间间隔,忽略了用户签到记录之间的时间间隔影响.另外,POI之间的地理距离以及语义信息也是影响推荐准确性的重要因素.基于此,本文提出自注意力下时空-语义相融合的POI序列推荐模型(POI sequence recommendation model based on the integration of spatiotemporal and semantics under self-attention, SA-TDS-PRec).首先,根据用户的实际签到时间建模POI轨迹.其次,融合POI绝对位置、时空间隔以及语义相关信息.最后利用自注意力机制捕捉用户动态偏好的演化,从而提高POI推荐的准确性.在公开数据集Gowalla和Yelp上进行可扩展实验.结果表明,该模型优于目前主流的基准模型,有效提升推荐结果准确性.
