公开(公告)号：US11580179B2

公开(公告)日：2023-02-14

申请号：US16139386

申请日：2018-09-24

Applicant: salesforce.com, inc.

Inventor： Pingping Xiu ,  Sitaram Asur ,  Anjan Goswami ,  Ziwei Chen ,  Na Cheng ,  Suhas Satish ,  Jacob Nathaniel Huffman ,  Peter Francis White ,  WeiPing Peng ,  Aditya Sakhuja ,  Jayesh Govindarajan ,  Edgar Gerardo Velasco

IPC: G06N5/00 ,  G06F16/9535 ,  G06F16/35 ,  G06F16/338 ,  H04L67/63

Abstract: A method and system for recommending articles including: receiving a customer request from the customer during the session; generating case data for a case, by an article recommender app; configuring a training set based on the subject and description data of the customer request; identifying, by an artificial intelligence (AI) app, a first pool of articles from a knowledge database; identifying by at least one query, a second pool of articles from a case article database to into a merged pool of articles; assigning, by the AI app, an implicit label to one of the first pool and the second pool of the articles; applying a model derived by the AI app based on customer behavior and a set of features related to the case to classify each article of the merged pool of articles based at least in part on the predicted relevance of the article.

公开(公告)号：US20180101617A1

公开(公告)日：2018-04-12

申请号：US15730591

申请日：2017-10-11

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Francisco Borges ,  Ammar Haris

IPC: G06F17/30 ,  G06N99/00 ,  H04L29/08

CPC classification number: G06F16/9535 ,  G06N20/00 ,  H04L67/02 ,  H04L67/306

Abstract: An online system identifies and ranks records using multiple machine learning models in response to a search query. Therefore, the online system can provide selected records that are of the most relevance to a user of a client device that provided the search query. More specifically, the online system applies a first machine learning model that is of low complexity, such as a regression model. Therefore, the first machine learning model can quickly narrow down the large number of records of the online system to a first set of candidate records. The online system analyzes candidate records in the first set by applying a more complex, second machine learning model that more accurately determines records of interest for the user. In various embodiments, the online system can apply subsequent machine learning models of higher complexity for selecting and ranking records for provision to the client device.

公开(公告)号：US20180101536A1

公开(公告)日：2018-04-12

申请号：US15728938

申请日：2017-10-10

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris

IPC: G06F17/30

CPC classification number: G06F16/24578 ,  G06F16/2228 ,  G06F16/2457 ,  G06F16/248 ,  G06F16/9535

Abstract: An online system receives a search query from a user. In response to the request, the online system obtains search results matching the search query and identifies a set of attributes describing a context of the search query. The online system generates a data structure that includes a plurality of search coefficients. The search coefficients are selected based on the identified set of attributes. Some of the search coefficients have conflicting values. The online system traverses the data structure to identify the search coefficients having conflicting values. For each search coefficient having conflicting values, the online system resolves conflicts and determines a value of the search coefficient. The online system ranks search results based on the resolved values of the search coefficients.

公开(公告)号：US11327979B2

公开(公告)日：2022-05-10

申请号：US16708925

申请日：2019-12-10

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris ,  Zachary Alexander ,  Scott Thurston Rickard, Jr. ,  Clifford Z. Huang

IPC: G06F16/00 ,  G06F16/2457 ,  G06F16/9032 ,  G06F16/903 ,  G06N20/00 ,  G06N20/20

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

公开(公告)号：US20210319037A1

公开(公告)日：2021-10-14

申请号：US17359388

申请日：2021-06-25

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris

IPC: G06F16/2457 ,  G06F16/248 ,  G06F16/22 ,  G06F16/9535

Abstract: An online system receives a search query from a user. In response to the request, the online system obtains search results matching the search query and identifies a set of attributes describing a context of the search query. The online system generates a data structure that includes a plurality of search coefficients. The search coefficients are selected based on the identified set of attributes. Some of the search coefficients have conflicting values. The online system traverses the data structure to identify the search coefficients having conflicting values. For each search coefficient having conflicting values, the online system resolves conflicts and determines a value of the search coefficient. The online system ranks search results based on the resolved values of the search coefficients.

公开(公告)号：US11093511B2

公开(公告)日：2021-08-17

申请号：US15728938

申请日：2017-10-10

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris

IPC: G06F16/2457 ,  G06F16/248 ,  G06F16/22 ,  G06F16/9535

Abstract: An online system receives a search query from a user. In response to the request, the online system obtains search results matching the search query and identifies a set of attributes describing a context of the search query. The online system generates a data structure that includes a plurality of search coefficients. The search coefficients are selected based on the identified set of attributes. Some of the search coefficients have conflicting values. The online system traverses the data structure to identify the search coefficients having conflicting values. For each search coefficient having conflicting values, the online system resolves conflicts and determines a value of the search coefficient. The online system ranks search results based on the resolved values of the search coefficients.

公开(公告)号：US10552432B2

公开(公告)日：2020-02-04

申请号：US15730660

申请日：2017-10-11

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris ,  Zachary Alexander ,  Scott Thurston Rickard, Jr. ,  Clifford Z. Huang

IPC: G06F17/30 ,  G06F16/2457 ,  G06F16/9032 ,  G06F16/903 ,  G06N20/00

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

公开(公告)号：US20180101537A1

公开(公告)日：2018-04-12

申请号：US15730660

申请日：2017-10-11

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Nicholas Beng Tek Geh ,  Ammar Haris ,  Zachary Alexander ,  Scott Thurston Rickard, JR. ,  Clifford Z. Huang

IPC: G06F17/30

CPC classification number: G06F16/24578 ,  G06F16/2457 ,  G06F16/90324 ,  G06F16/90348 ,  G06N20/00

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

公开(公告)号：US20180101527A1

公开(公告)日：2018-04-12

申请号：US15730574

申请日：2017-10-11

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Ammar Haris ,  Nicholas Beng Tek Geh ,  Francisco Borges

IPC: G06F17/30

CPC classification number: G06F16/93 ,  G06F16/248 ,  G06F16/3323 ,  G06F16/986

Abstract: An online system stores documents for access by users. The online system also stores query independent information about the documents. Query independent features include data that can be used to score or rank a document independent of any terms entered as a search query. The online system periodically determines whether the values of query independent features have changed, such as by checking activity logs. The online system updates records of query independent features accordingly, and sends information about the updated records to an enterprise search platform for re-indexing. When a user sends a search query to the online system, the enterprise search platform determines whether documents are relevant to the query based on the document contents and the query independent features associated with the documents.

公开(公告)号：US20170286978A1

公开(公告)日：2017-10-05

申请号：US15085808

申请日：2016-03-30

Applicant: salesforce.com, inc.

Inventor： Jayesh Govindarajan ,  Naren Chittar

IPC: G06Q30/02 ,  G06Q50/00 ,  G06F17/30

CPC classification number: G06Q30/0202 ,  G06F16/24578 ,  G06F16/951 ,  G06Q30/0276 ,  G06Q50/01

Abstract: A trend system generates external trend information for different topics that may be related to merchant items offered by a merchant. External trend information is generated based on content items obtained from sources that are separate from the merchant. Periodically the trend system determines an interest growth rate for a topic based on the external trend information. If the interest growth rate determined for the topic is above an interest threshold, the trend system recommends to the merchant starting a campaign related to the topic.