{"title":"Global and Personalized Query Probability for Obfuscation-Based Web Search","authors":"Hongya Wang, Wenyan Liu, Xiaoling Wang, Yingjie Zhang","doi":"10.1109/ICBK50248.2020.00045","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00045","url":null,"abstract":"Over the past few years, the volumes of digital information have increased dramatically. Web search engines become indispensable to our daily lives due to their capability of filtering relevant information. The web search engines collect users’ online behavior data which implies their personalized interest, even contains sensitive information. A malicious attacker may benefit from advertising or manipulating political campaigns, which poses a serious threat to public security. In this paper, we propose two new methods to generate dummy queries base on global and personalized queries’ probability. The generated anonymous queries set has greater information entropy which helps to protect user’s intent in web search. Experiment results verify the validity of our methods.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122600848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xuegang Hu, Jianxing Liao, Peipei Li, Junwei Lv, Lei Li
{"title":"Learning Wasserstein Distance-Based Gaussian Graphical Model for Multivariate Time Series Classification","authors":"Xuegang Hu, Jianxing Liao, Peipei Li, Junwei Lv, Lei Li","doi":"10.1109/ICBK50248.2020.00025","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00025","url":null,"abstract":"Multivariate time series classification occupies an important position in time series data mining tasks and has been applied in many fields. However, due to the statistical coupling between different variables of Multivariate Time Series (MTS) data, traditional classification methods cannot find complex dependencies between different variables, so most existing methods perform not well in MTS classification with many variables. Thus, in this paper, a novel model-based classification method is proposed, called Wasserstein Distance-based Gaussian Graphical Model classification (WD-GGMC), which converts the original MTS data into two important parameters of the Gaussian Graphical Model: the sparse inverse covariance matrix and the mean vector. Among them, the former is the most important parameter, which contains the information between variables and solved by Alternating Direction Method of Multipliers (ADMM). Furthermore, the Wasserstein Distance is applied as the similarity measure for different subsequences because it can measure the similarity between different distributions. Experimental results on the eight public MTS datasets demonstrate the effectiveness of the proposed method in MTS classification.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126275516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"GeLaiGeLai: A visual platform for analysis of Classical Chinese Poetry based on Knowledge Graph","authors":"Yuting Wei, Huazheng Wang, Jiaqi Zhao, Yutong Liu, Yun Zhang, Bin Wu","doi":"10.1109/ICBK50248.2020.00078","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00078","url":null,"abstract":"Classical Chinese poetry contains many precious historical and cultural information. However, the knowledge of classical Chinese poetry is highly fragmented. The statistics of imageries and allusions are often incomplete. Most of related works do not analyse the knowledge of poetry from the perspective of archaic Chinese words. It is hard to determine whether words are semantically related. Therefore, to solve these problems, “GeLaiGeLai” has been set up here, which is a system for data analysis of classical Chinese poetry based on knowledge graph. On the one hand, the platform is able to quickly and accurately find new words in ancient Chinese corpus through AP-LSTM-CRF, which is a new word detection method that first generates frequent character sequences using improved Apriori algorithm and then uses Bi-LSTM-CRF model which could generate the segmentation probability of every position of the sentence to further judge whether each frequent character sequence is a true new word. On the other hand, we visualize the knowledge graph and analyse the commonly-used word and emotions of poets. At the same time, the platform complements the characteristics of poetry, using knowledge graph to solve the problem of knowledge fragmentation and making it more systematic. With the knowledge graph, the performance of many reasoning and analysis tasks about classical Chinese poetry can be improved, such as determining the theme of poetry and analyzing the emotion of poetry, which proves the knowledge graph is helpful to understand classical Chinese poetry.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121885365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Knowledge Fragment Cleaning in a Genealogy Knowledge Graph","authors":"Guliu Liu, Lei Li","doi":"10.1109/ICBK50248.2020.00079","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00079","url":null,"abstract":"As an important topic of artificial intelligence, knowledge graphs have a wide range of applications such as semantic search, intelligent question answering, and visual decision support. Among them, the genealogy knowledge graph, as a kind of domain knowledge graph, has important application value in genetic disease analysis, population behavior analysis, etc. In the case of multiple data sources and multi-person collaboration, the construction of a genealogy knowledge graph involves the techniques of knowledge representation, knowledge acquisition, and knowledge fusion. In the knowledge fusion process, there are many situations such as the lack and chaos of a relationship, redundant entities, the isolation of some entities and knowledge fragments. How to effectively detect and process these problematic knowledge fragments is significant to the construction of a genealogy knowledge graph. In response to this problem, we propose a method for cleaning the problematic knowledge fragments in a genealogy knowledge graph. The method consists of three phases. In phase 1, we propose a method for detecting and analyzing the problematic knowledge fragments. In phase 2, we design a method for supplementing the entity-relationship of people for different error patterns and a method fusion method for the aligned entity. In phase 3, for the cleaning of isolated knowledge fragments, we propose an entity alignment method based on the father-son relationship and people’s names and a connection method of isolated knowledge fragments. Finally, we conduct experiments on a family tree dataset of the Huapu System, and the experimental results indicate the effectiveness and practicality of the method.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127706736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Self-Attention and Dynamic Convolution Hybrid Model for Neural Machine Translation","authors":"Zhebin Zhang, Sai Wu, Gang Chen, Dawei Jiang","doi":"10.1109/ICBK50248.2020.00057","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00057","url":null,"abstract":"In sequence-to-sequence learning, models based on the self-attention mechanism dominate the network structures used for neural machine translation. Recently, convolutional networks have been demonstrated to perform excellently on various translation tasks. Despite the fact that self-attention and convolution have different strengths in modeling sequences, few efforts have been devoted to combining them. In this work, we propose a hybrid model that benefits from both mechanisms. We combine a self-attention module and a dynamic convolution module by taking a weighted sum of their outputs where the weights can be dynamically learned by the model during training. Experimental results show that our hybrid model outperforms baseline models built solely on either of these two mechanisms. And we produce new state-of-the-art results on IWSLT’15 English-German dataset.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127893078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Heterogeneous Dynamic Graph Attention Network","authors":"Qiuyan Li, Yanlei Shang, Xiuquan Qiao, Wei Dai","doi":"10.1109/ICBK50248.2020.00064","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00064","url":null,"abstract":"Network embedding (graph embedding) has become the focus of studying graph structure in recent years. In addition to the research on homogeneous networks and heterogeneous networks, there are also some methods to attempt to solve the problem of dynamic network embedding. However, in dynamic networks, there is no research method specifically for heterogeneous networks. Therefore, this paper proposes a heterogeneous dynamic graph attention network (HDGAN), which attempts to use the attention mechanism to take the heterogeneity and dynamics of the network into account at the same time, so as to better learn network embedding. Our method is based on three levels of attention, namely structural-level attention, semantic-level attention and time-level attention. Structural-level attention pays attention to the network structure itself, and obtains the representation of structural-level nodes by learning the attention coefficients of neighbor nodes. Semantic-level attention integrates semantic information into the representation of nodes by learning the optimal weighted combination of different meta-paths. Time-level attention is based on the time decay effect, and the time feature is introduced into the node representation by neighborhood formation sequence. Through the above three levels of attention mechanism, the final network embedding can be obtained.Through experiments on two real-world heterogeneous dynamic networks, our models have the best results, proving the effectiveness of the HDGAN model.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133230503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Siamese BERT Model with Adversarial Training for Relation Classification","authors":"Zhimin Lin, Dajiang Lei, Yuting Han, Guoyin Wang, Weihui Deng, Yuan Huang","doi":"10.1109/ICBK50248.2020.00049","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00049","url":null,"abstract":"Relation classification is a very important Natural Language Processing (NLP) task to classify the relations from the plain text. It is one of the basic tasks of constructing a knowledge graph. Most existing state-of-the-art methods are primarily based on Convolutional Neural Networks(CNN) or Long Short-Term Memory Networks(LSTM). Recently, many pre-trained Bidirectional Encoder Representation from Transformers (BERT) models have been successfully used in the sequence labeling and many NLP classification tasks. Relation classification is different in that it needs to pay attention to not only the sentence information but also the entity pairs. In this paper, a Siamese BERT model with Adversarial Training (SBERT-AT) is proposed for relation classification. Firstly, the features of the entities and the sentence can be extracted separately to improve the performance of relation classification. Secondly, the adversarial training is applied to the SBERT architecture to improve the robustness. Lastly, the experimental results demonstrate that we achieve significant improvement compared with the other methods on real-world datasets.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133346910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Adaptive Regularization-Incorporated Latent Factor Analysis","authors":"Xin Luo, Ye Yuan, Di Wu","doi":"10.1109/ICBK50248.2020.00074","DOIUrl":"https://doi.org/10.1109/ICBK50248.2020.00074","url":null,"abstract":"The valuable knowledge contained in High-dimensional and Sparse (HiDS) matrices can be efficiently extracted by a latent factor (LF) model. Regularization techniques are widely incorporated into an LF model to avoid overfitting. The regularization coefficient is very crucial to the prediction accuracy of models. However, its tuning process is time-consuming and boring. This study aims at making the regularization coefficient of a regularized LF model self-adaptive. To do so, an adaptive particle swarm optimization (APSO) algorithm is introduced into a regularized LF model to automatically select the optimal regularization coefficient. Then, to enhance the global search capability of particles, we further propose an APSO and particle swarm optimization (PSO)-incorporated (AP) algorithm, thereby achieving an AP-based LF (APLF) model. Experimental results on four HiDS matrices generated by real applications demonstrate that an APLF model can achieve an automatic selection of regularization coefficient, and is superior to a regularized LF model in terms of prediction accuracy and computational efficiency.","PeriodicalId":432857,"journal":{"name":"2020 IEEE International Conference on Knowledge Graph (ICKG)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133747646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}