公开(公告)号：US20210272559A1

公开(公告)日：2021-09-02

申请号：US16805660

申请日：2020-02-28

Applicant: Intuit Inc.

Inventor： Shlomi Medalion ,  Alexander Zhicharevich ,  Yair Horesh ,  Oren Sar Shalom ,  Elik Sror ,  Adi Shalev

IPC: G10L15/18 ,  G06N3/08 ,  G06N3/04 ,  G10L15/22 ,  G10L15/16 ,  G10L15/06 ,  G10L15/197

Abstract: A method of training machine learning models (MLMs). An issue vector is generated using an issue MLM to generate a first output including first embedded natural language issue statements. An action vector is generated using an action MLM to generate a second output including related embedded natural language action statements. The issue and action MLMs are of a same type. An inner product of the first and second output is calculated, forming a third output. The third output is processed according to a sigmoid gate process to predict whether a given issue statement and corresponding action statement relate to a same call, resulting in a fourth output. A loss function is calculated from the fourth output by comparing the fourth output to a known result. The issue MLM and the action MLM are modified using the loss function to obtain a trained issue MLM and a trained action MLM.

公开(公告)号：US20210233520A1

公开(公告)日：2021-07-29

申请号：US16751867

申请日：2020-01-24

Applicant: Intuit Inc.

Inventor： Oren Sar Shalom ,  Yair Horesh ,  Alexander Zhicharevich ,  Elik Sror ,  Adi Shalev ,  Yehezkel Shraga Resheff

IPC: G10L15/187 ,  G10L15/06 ,  G10L15/22 ,  G10L15/02 ,  H04M3/50 ,  G06N20/00

Abstract: A method for improving a transcription may include identifying, in the transcription, reliable channel tokens of an utterance of a reliable channel and an unreliable channel token of an utterance of an unreliable channel, and generating, using a machine learning model, a vector embedding for the unreliable channel token and vector embeddings for the reliable channel tokens. The method may further include calculating vector distances between the vector embedding and the vector embeddings, and generating, for the unreliable channel token and using the vector distances, a score corresponding to a reliable channel token. The method may further include determining that the score is within a threshold score, and in response to determining that the score is within the threshold score, replacing, in the transcription, the unreliable channel token with the reliable channel token.

公开(公告)号：US20210149671A1

公开(公告)日：2021-05-20

申请号：US16688697

申请日：2019-11-19

Applicant: Intuit Inc.

Inventor： Oren Sar Shalom ,  Meng Chen ,  Linxia Liao ,  Yehezkel Shraga Resheff

IPC: G06F9/30 ,  G06N5/02 ,  G06N3/08

Abstract: A machine learning method. A source domain data structure and a target domain data structure are combined into a unified data structure. First data in the source domain data structure are latent with respect to second data in the target domain data structure. The unified data structure includes user vectors that combine the first data and the second data. The user vectors are transformed into a transformed data structure by applying a mapping function to the user vectors. The mapping function relates, using at least one parameter, first relationships in the source domain data structure to second relationships in the target domain data structure. The at least one parameter is based on a combination of affinity scores relating items with which the user interacted and did not interact. The transformed data structure is input into a machine learning model, from which is obtained a recommendation relating to the target domain.

公开(公告)号：US20210081454A1

公开(公告)日：2021-03-18

申请号：US16573619

申请日：2019-09-17

Applicant: Intuit Inc.

Inventor： Oren Sar Shalom ,  Alexander Zhicharevich ,  Rami Cohen ,  Yonatan Ben-Simhon

IPC: G06F16/901 ,  G06F16/953

Abstract: A method involves receiving search queries, having search terms, submitted to at least one computerized search engine. For each query, a corresponding pairwise relation in the search queries is calculated. The corresponding pairwise relation is a corresponding probability of a potential edge relationship between at least two terms. Thus, potential edges are formed. A general graph of the terms is constructed by selecting edges from the potential edges. The general graph is nodes representing the terms used in the search queries. The general graph also is edges representing semantic relationships among the nodes. A hierarchical graph is constructed from the general graph by altering at least one of the edges among the nodes in the general graph to form the hierarchical graph.