公开(公告)号：WO2021262333A1

公开(公告)日：2021-12-30

申请号：PCT/US2021/032123

申请日：2021-05-13

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： LI, Mingqin ,  CHEN, Qi ,  WANG, Jingdong ,  LI, Zengzhong ,  ZHU, Jeffrey, Song ,  ZHANG, Shi ,  YADAV, Nilesh, N. ,  ZHANG, Han

IPC: G06N20/00 ,  G06F16/2237 ,  G06F16/2379 ,  G06F16/27 ,  G06F16/951 ,  G06F16/953 ,  G06N5/02

Abstract: System and methods are directed to operations associated with an approximate nearest neighbor search engine. More specifically, a vector semantically representing content to be added to a search index may be received. The search index may include a neighborhood graph having a plurality of nodes, where each node of the plurality of nodes is associated with content in a content repository. A plurality of nodes within the search index determined to be most semantically similar to the received vector semantically representing content to be added to the search index may be identified. The node corresponding to the received vector semantically representing content to be added to the search index to the search index may be added to the search index and a listing of nearest neighbors associated with each of the plurality of nodes may be updated to include an identifier associated with the added node.

公开(公告)号：WO2020060605A1

公开(公告)日：2020-03-26

申请号：PCT/US2019/038078

申请日：2019-06-20

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： LI, Wei ,  LI, Mingqin ,  CHEN, Qi ,  LUO, Wei ,  REN, Gang ,  WANG, Jingdong ,  ZHANG, Lintao

IPC: G06N3/08 ,  G06F16/33 ,  G06N3/00 ,  G06N5/00 ,  G06N5/02

Abstract: The present disclosure discloses a technique for generating nearest neighbor searching strategy for vectors based on reinforcement learning. A searching technology using vectors approximate matching may be applied in a searching engine. More particularly, the searching engine may be subjected to a training by suing a reinforcement learning so that a mapping relationship between a calculation state and a behavior action may be obtained. The searching engine may automatically generate a searching strategy for an enquiry content by using the mapping relationship.

公开(公告)号：WO2019118253A1

公开(公告)日：2019-06-20

申请号：PCT/US2018/064146

申请日：2018-12-06

Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

Inventor： HAN, Dianfei ,  HUA, Jiefeng ,  ZHANG, Dongqing ,  ZHU, Suyan ,  ZHANG, Shi ,  REN, Gang ,  TAN, Feng ,  WANG, Jingdong ,  SHEN, Hui ,  LUO, Wei ,  LI, Zengzhong ,  ZHANG, Lintao ,  CHEN, Qi ,  LI, Mingqin

IPC: G06F16/903

CPC classification number: G06F16/90335

Abstract: The present disclosure provides technical solutions related to a document recalling based the vector nearest neighbor search. The technique of vector approximate matching is applied to the searching engine. The content for searching and webpage documents may be turned into semantic vectors, respectively, and the webpage documents related to the content for searching may be obtained in a way of searching by vector approximate matching, so that a searching service, which could understand the user's intention better, may be provided without being limited by the searching method of symbol matching.