{"title":"Exploration of TransE in a Distributed Environment","authors":"Meiyan Lu, L. Liao, Feng Zhang, Dandan Song","doi":"10.1109/ICDCS47774.2020.00190","DOIUrl":null,"url":null,"abstract":"Knowledge graph is popular in knowledge mining fields. TransE uses the structure information of triples $\\left( {\\overrightarrow {{e_h}} + \\overrightarrow {{e_r}} \\approx \\overrightarrow {{e_t}} } \\right)$ to embed knowledge graphs into a continuous vector space, which is a very important component in knowledge representations. However, current TransE models are only implemented on single-node machines. With the explosive growth of data volumes, single-node TransE cannot meet the demand for data processing of large knowledge graphs, so a distributed TransE is urgently needed. In this poster, we propose a distributed TransE written in MPI, which can run on HPC clusters. In our experiments, our distributed TransE exhibits high-performance speedup and accuracy.","PeriodicalId":158630,"journal":{"name":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","volume":"228 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDCS47774.2020.00190","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Knowledge graph is popular in knowledge mining fields. TransE uses the structure information of triples $\left( {\overrightarrow {{e_h}} + \overrightarrow {{e_r}} \approx \overrightarrow {{e_t}} } \right)$ to embed knowledge graphs into a continuous vector space, which is a very important component in knowledge representations. However, current TransE models are only implemented on single-node machines. With the explosive growth of data volumes, single-node TransE cannot meet the demand for data processing of large knowledge graphs, so a distributed TransE is urgently needed. In this poster, we propose a distributed TransE written in MPI, which can run on HPC clusters. In our experiments, our distributed TransE exhibits high-performance speedup and accuracy.