Knowledge graph representation has been a long standing goal of artificial intelligence. In this paper,we consider a method for knowledge graph embedding of hyper-relational data, which are commonly found in knowledge...Knowledge graph representation has been a long standing goal of artificial intelligence. In this paper,we consider a method for knowledge graph embedding of hyper-relational data, which are commonly found in knowledge graphs. Previous models such as Trans（E, H, R） and CTrans R are either insufficient for embedding hyper-relational data or focus on projecting an entity into multiple embeddings, which might not be effective for generalization nor accurately reflect real knowledge. To overcome these issues, we propose the novel model Trans HR, which transforms the hyper-relations in a pair of entities into an individual vector, serving as a translation between them. We experimentally evaluate our model on two typical tasks—link prediction and triple classification.The results demonstrate that Trans HR significantly outperforms Trans（E, H, R） and CTrans R, especially for hyperrelational data.展开更多
Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in...Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in the training corpus. In this paper, we propose using co-reference resolution to improve the word embedding by extracting better context. We evaluate four word embeddings with considerations of co-reference resolution and compare the quality of word embedding on the task of word analogy and word similarity on multiple data sets.Experiments show that by using co-reference resolution, the word embedding performance in the word analogy task can be improved by around 1.88%. We find that the words that are names of countries are affected the most,which is as expected.展开更多
The rapid development of the internet has ushered the real world into a“media-centric”digital era where virtually everything serves as a medium.Leveraging the new attributes of interactivity,immediacy,and personaliz...The rapid development of the internet has ushered the real world into a“media-centric”digital era where virtually everything serves as a medium.Leveraging the new attributes of interactivity,immediacy,and personalization facilitated by online communication,folklore has found a broad avenue for dissemination.Among these,online social networks have become a vital channel for propagating folklore.By using social network theory,we devise a comprehensive approach known as SocialPre.Firstly,we utilize embedding techniques to capture users’low-level and high-level social relationships.Secondly,by applying an automatic weight assignment mechanism based on the embedding representations,multi-level social relationships are aggregated to assess the likelihood of a social interaction between any two users.These experiments demonstrate the ability to classify different social groups.In addition,we delve into the potential directions of folklore evolution,thus laying a theoretical foundation for future folklore communication.展开更多
Knowledge bases(KBs)are far from complete,necessitating a demand for KB completion.Among various methods,embedding has received increasing attention in recent years.PTransE,an important approach using embedding method...Knowledge bases(KBs)are far from complete,necessitating a demand for KB completion.Among various methods,embedding has received increasing attention in recent years.PTransE,an important approach using embedding method in KB completion,considers multiple-step relation paths based on TransE,but ignores the association between entity and their related entities with the same direct relationships.In this paper,we propose an approach called EP-TransE,which considers this kind of association.As a matter of fact,the dissimilarity of these related entities should be taken into consideration and it should not exceed a certain threshold.EPTransE adjusts the embedding vector of an entity by comparing it with its related entities which are connected by the same direct relationship.EPTransE further makes the euclidean distance between them less than a certain threshold.Therefore,the embedding vectors of entities are able to contain rich semantic information,which is valuable for KB completion.In experiments,we evaluated our approach on two tasks,including entity prediction and relation prediction.Experimental results show that our idea of considering the dissimilarity of related entities with the same direct relationships is effective.展开更多
This research addresses the new level-direction decomposition in the area of image watermarking as the further development of investigations. The main process of realizing a watermarking framework is to generate a wat...This research addresses the new level-direction decomposition in the area of image watermarking as the further development of investigations. The main process of realizing a watermarking framework is to generate a watermarked image with a focus on contourlet embedding representation. The approach performance is evaluated through several indices including the peak signal-to-noise ratio and structural similarity, whereby a set of attacks are carried out using a module of simulated attacks. The obtained information is analyzed through a set of images, using different color models, to enable the calculation of normal correlation. The module of the inverse of contourlet embedding representation is correspondingly employed to obtain the present watermarked image, as long as a number of original images are applied to a scrambling module, to represent the information in disorder. This allows us to evaluate the performance of the proposed approach by analyzing a complicated system, where a decision making system is designed to find the best level and the corresponding direction regarding contourlet embedding representation. The results are illustrated in appropriate level-direction decomposition. The key contribution lies in using a new integration of a set of subsystems, employed based upon the novel mechanism in contourlet embedding representation, in association with the decision making system. The presented approach is efficient compared with state-of-the-art approaches, under a number of serious attacks. A number of benchmarks are obtained and considered along with the proposed framework outcomes. The results support our ideas.展开更多
基金partially supported by the National Natural Science Foundation of China(Nos.61302077,61520106007,61421061,and 61602048)
文摘Knowledge graph representation has been a long standing goal of artificial intelligence. In this paper,we consider a method for knowledge graph embedding of hyper-relational data, which are commonly found in knowledge graphs. Previous models such as Trans（E, H, R） and CTrans R are either insufficient for embedding hyper-relational data or focus on projecting an entity into multiple embeddings, which might not be effective for generalization nor accurately reflect real knowledge. To overcome these issues, we propose the novel model Trans HR, which transforms the hyper-relations in a pair of entities into an individual vector, serving as a translation between them. We experimentally evaluate our model on two typical tasks—link prediction and triple classification.The results demonstrate that Trans HR significantly outperforms Trans（E, H, R） and CTrans R, especially for hyperrelational data.
基金supported by the National HighTech Research and Development(863)Program(No.2015AA015401)the National Natural Science Foundation of China(Nos.61533018 and 61402220)+2 种基金the State Scholarship Fund of CSC(No.201608430240)the Philosophy and Social Science Foundation of Hunan Province(No.16YBA323)the Scientific Research Fund of Hunan Provincial Education Department(Nos.16C1378 and 14B153)
文摘Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in the training corpus. In this paper, we propose using co-reference resolution to improve the word embedding by extracting better context. We evaluate four word embeddings with considerations of co-reference resolution and compare the quality of word embedding on the task of word analogy and word similarity on multiple data sets.Experiments show that by using co-reference resolution, the word embedding performance in the word analogy task can be improved by around 1.88%. We find that the words that are names of countries are affected the most,which is as expected.
基金supported by the National Social Science Foundation of China(No.21BTY115).
文摘The rapid development of the internet has ushered the real world into a“media-centric”digital era where virtually everything serves as a medium.Leveraging the new attributes of interactivity,immediacy,and personalization facilitated by online communication,folklore has found a broad avenue for dissemination.Among these,online social networks have become a vital channel for propagating folklore.By using social network theory,we devise a comprehensive approach known as SocialPre.Firstly,we utilize embedding techniques to capture users’low-level and high-level social relationships.Secondly,by applying an automatic weight assignment mechanism based on the embedding representations,multi-level social relationships are aggregated to assess the likelihood of a social interaction between any two users.These experiments demonstrate the ability to classify different social groups.In addition,we delve into the potential directions of folklore evolution,thus laying a theoretical foundation for future folklore communication.
基金This work was supported by the National Key Research and Development Plan of China(2017YFD0400101)the National Natural Science Foundation of China(Grant No.61502294)the Natural Science Foundation of Shanghai,Project Number(16ZR1411200).
文摘Knowledge bases(KBs)are far from complete,necessitating a demand for KB completion.Among various methods,embedding has received increasing attention in recent years.PTransE,an important approach using embedding method in KB completion,considers multiple-step relation paths based on TransE,but ignores the association between entity and their related entities with the same direct relationships.In this paper,we propose an approach called EP-TransE,which considers this kind of association.As a matter of fact,the dissimilarity of these related entities should be taken into consideration and it should not exceed a certain threshold.EPTransE adjusts the embedding vector of an entity by comparing it with its related entities which are connected by the same direct relationship.EPTransE further makes the euclidean distance between them less than a certain threshold.Therefore,the embedding vectors of entities are able to contain rich semantic information,which is valuable for KB completion.In experiments,we evaluated our approach on two tasks,including entity prediction and relation prediction.Experimental results show that our idea of considering the dissimilarity of related entities with the same direct relationships is effective.
文摘This research addresses the new level-direction decomposition in the area of image watermarking as the further development of investigations. The main process of realizing a watermarking framework is to generate a watermarked image with a focus on contourlet embedding representation. The approach performance is evaluated through several indices including the peak signal-to-noise ratio and structural similarity, whereby a set of attacks are carried out using a module of simulated attacks. The obtained information is analyzed through a set of images, using different color models, to enable the calculation of normal correlation. The module of the inverse of contourlet embedding representation is correspondingly employed to obtain the present watermarked image, as long as a number of original images are applied to a scrambling module, to represent the information in disorder. This allows us to evaluate the performance of the proposed approach by analyzing a complicated system, where a decision making system is designed to find the best level and the corresponding direction regarding contourlet embedding representation. The results are illustrated in appropriate level-direction decomposition. The key contribution lies in using a new integration of a set of subsystems, employed based upon the novel mechanism in contourlet embedding representation, in association with the decision making system. The presented approach is efficient compared with state-of-the-art approaches, under a number of serious attacks. A number of benchmarks are obtained and considered along with the proposed framework outcomes. The results support our ideas.
