首页|Patent Application Titled 'Building Time-Decayed Line Graphs For Direct Embeddin g Of Continuous-Timed Interactions In Generating Time-Aware Recommendations' Pub lished Online (USPTO 20240311623)
Patent Application Titled 'Building Time-Decayed Line Graphs For Direct Embeddin g Of Continuous-Timed Interactions In Generating Time-Aware Recommendations' Pub lished Online (USPTO 20240311623)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Reporters obtained the following quote from the background information supplied by the inventors: "Recent years have seen significant improvements in hardware a nd software platforms for modeling interactions between computing devices across computer networks. For example, some conventional systems utilize temporal netw orks, which are graphs augmented with a time value for each edge, to model a var iety of phenomena involving timed interactions. Additionally, conventional syste ms utilize machine learning models with temporal networks to generate various cl assifications or predictions. Conventional machine learning models generate thes e predictions using vector representations, or embeddings, for the nodes of the temporal network. For example, most conventional systems generate node embedding s that are utilized by various machine learning models to generate predictions o r classifications. In a few cases, some conventional systems indirectly generate edge embeddings by aggregating the node embeddings (e.g., by taking the entrywi se product of the two node embeddings corresponding to an edge). In many cases, conventional systems also require discretized time representations. Notably, the conventional approaches of deriving node embeddings from temporal networks, ind irectly deriving edge embeddings by aggregating node embeddings, and/or utilizin g discretized time representations lead to deficiencies in system accuracy, effi ciency, and flexibility.
NETL
NSTL
