公开(公告)号：US20190065539A1

公开(公告)日：2019-02-28

申请号：US15691610

申请日：2017-08-30

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Seema Jethani ,  Mohnish Kodnani ,  Vishnusaran Ramaswamy ,  Jonathan Baggott ,  Harish Kumar Vittal Murthy

IPC: G06F17/30

Abstract: A data management system accesses a set of vectors containing binary values generates a corresponding set of sequentially ordered vector blocks. Each vector contains a set of two or more binary values and a numerical vector identifier. The data management system generates a block index based on each corresponding set of sequentially ordered vector blocks. The block index includes a set of vector block arrays, each corresponding to a respective sequential position and including one vector block from each of the sets of sequentially ordered vector blocks that are in the respective sequential position. The vector blocks in each vector block array being are ordered sequentially based on two or more sequential binary values in each respective vector block. For each vector block array, the data management system combines pairs of sequentially ordered vector blocks containing matching sets of two or more binary values into combined vector blocks.

公开(公告)号：US20190392058A1

公开(公告)日：2019-12-26

申请号：US16016788

申请日：2018-06-25

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Mohnish Kodnani ,  Alberto Ordonez Pereira ,  Ravindra Surya Lanka

IPC: G06F17/30

Abstract: Systems and methods are provided for creating and utilizing search indexes with different properties to provide efficient and flexible means for retrieving and ranking high-dimensional content. A first index and a second index are generated with a different sets of properties, which may include a permutation prefix length and a vector quantization scheme, and are each associated with a computational workload. Further, at least one index structure may be created with a posting list associated with a permutation prefix and a set of references, rather than only a permutation prefix, to balance out posting lists of variable lengths within the index. Another embodiment of the disclosure provides for expansion of search results at search time, rather than during indexing, by retrieving posting lists associated with shorter permutation prefix after an initial list is determined to have insufficient recall.

公开(公告)号：US20200012630A1

公开(公告)日：2020-01-09

申请号：US16574794

申请日：2019-09-18

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Seema Jethani ,  Mohnish Kodnani ,  Vishnusaran Ramaswamy ,  Jonathan Baggott ,  Harish Kumar Vittal Murthy

IPC: G06F16/22 ,  G06F16/25

Abstract: A data management system accesses a set of vectors containing binary values and generates vector blocks comprising binary values from each vector. Each of at least a portion of the vector blocks for each vector contain a set of two or more binary values from the vector. The data management system generates a block index based on the vector blocks. The block index includes a set of vector block arrays, each vector block array corresponding to a position in the vectors and including the binary values of a vector block from each vector. The data management system can identify relevant vectors for a target vector by generating vector blocks from the target vector and querying the block index to identify candidate vectors.

公开(公告)号：US12038896B2

公开(公告)日：2024-07-16

申请号：US17899944

申请日：2022-08-31

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Mohnish Kodnani ,  Alberto Ordonez Pereira ,  Ravindra Surya Lanka

IPC: G06F16/22 ,  G06F16/2457 ,  G06F16/9535 ,  G06F16/95

CPC classification number: G06F16/2237 ,  G06F16/24578 ,  G06F16/9535

Abstract: Systems and methods are provided for creating and utilizing search indexes with different properties to provide efficient and flexible means for retrieving and ranking high-dimensional content. A first index and a second index are generated with a different sets of properties, which may include a permutation prefix length and a vector quantization scheme, and are each associated with a computational workload. Further, at least one index structure may be created with a posting list associated with a permutation prefix and a set of references, rather than only a permutation prefix, to balance out posting lists of variable lengths within the index. Another embodiment of the disclosure provides for expansion of search results at search time, rather than during indexing, by retrieving posting lists associated with shorter permutation prefix after an initial list is determined to have insufficient recall.

公开(公告)号：US20220414079A1

公开(公告)日：2022-12-29

申请号：US17899944

申请日：2022-08-31

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Mohnish Kodnani ,  Alberto Ordonez Pereira ,  Ravindra Surya Lanka

IPC: G06F16/22 ,  G06F16/9535 ,  G06F16/2457

Abstract: Systems and methods are provided for creating and utilizing search indexes with different properties to provide efficient and flexible means for retrieving and ranking high-dimensional content. A first index and a second index are generated with a different sets of properties, which may include a permutation prefix length and a vector quantization scheme, and are each associated with a computational workload. Further, at least one index structure may be created with a posting list associated with a permutation prefix and a set of references, rather than only a permutation prefix, to balance out posting lists of variable lengths within the index. Another embodiment of the disclosure provides for expansion of search results at search time, rather than during indexing, by retrieving posting lists associated with shorter permutation prefix after an initial list is determined to have insufficient recall.

公开(公告)号：US11449484B2

公开(公告)日：2022-09-20

申请号：US16016788

申请日：2018-06-25

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Mohnish Kodnani ,  Alberto Ordonez Pereira ,  Ravindra Surya Lanka

IPC: G06F16/22 ,  G06F16/9535 ,  G06F16/2457 ,  G06F16/95

Abstract: Systems and methods are provided for creating and utilizing search indexes with different properties to provide efficient and flexible means for retrieving and ranking high-dimensional content. A first index and a second index are generated with a different sets of properties, which may include a permutation prefix length and a vector quantization scheme, and are each associated with a computational workload. Further, at least one index structure may be created with a posting list associated with a permutation prefix and a set of references, rather than only a permutation prefix, to balance out posting lists of variable lengths within the index. Another embodiment of the disclosure provides for expansion of search results at search time, rather than during indexing, by retrieving posting lists associated with shorter permutation prefix after an initial list is determined to have insufficient recall.

公开(公告)号：US10649968B2

公开(公告)日：2020-05-12

申请号：US15691610

申请日：2017-08-30

Applicant: eBay Inc.

Inventor： Roberto Daniel Konow Krause ,  Seema Jethani ,  Mohnish Kodnani ,  Vishnusaran Ramaswamy ,  Jonathan Baggott ,  Harish Kumar Vittal Murthy

IPC: G06F17/30 ,  G06F16/22 ,  G06F16/25 ,  G06F16/21

Abstract: A data management system accesses a set of vectors containing binary values generates a corresponding set of sequentially ordered vector blocks. Each vector contains a set of two or more binary values and a numerical vector identifier. The data management system generates a block index based on each corresponding set of sequentially ordered vector blocks. The block index includes a set of vector block arrays, each corresponding to a respective sequential position and including one vector block from each of the sets of sequentially ordered vector blocks that are in the respective sequential position. The vector blocks in each vector block array being are ordered sequentially based on two or more sequential binary values in each respective vector block. For each vector block array, the data management system combines pairs of sequentially ordered vector blocks containing matching sets of two or more binary values into combined vector blocks.