Generating object morphisms during object search

Patent metadata

Field	Value
Publication number	US-12292878-B2
Application number	US-202318501588-A
Country	US
Kind code	B2
Filing date	Nov 3, 2023
Priority date	Jun 3, 2022
Publication date	May 6, 2025
Grant date	May 6, 2025

Abstract

Generating object morphisms during object search includes obtaining object-search request data, wherein the object-search request data includes object-search terms, obtaining resolved-request data representing the object-search terms, determining that a first analytical object partially consistent with the resolved-request data is available, wherein the first analytical object is consistent with a first portion of the resolved-request data, generating candidate object-morphism data with respect to the first analytical object in accordance with a second portion of the resolved-request data, outputting object-search response data including the candidate object-morphism data for presentation to a user, obtaining data indicating a selected object morphism from the candidate object-morphism data, generating a second analytical object in accordance with the first analytical object and the selected object morphism, wherein the second analytical object differs from the first analytical object, and outputting response data including the second analytical object for presentation to the user.

First claim

What is claimed is:

1. A non-transitory computer-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
obtaining first data expressing usage intent with respect to a low-latency data access and analysis system;
determining that the first data expressing usage intent includes object-search request data, wherein the object-search request data includes object-search terms; and
in response to determining that the first data expressing usage intent includes the object-search request data:
obtaining resolved-request data representing the object-search terms as a sequence of tokens;
determining that a first analytical object consistent with the resolved-request data is unavailable in the low-latency data access and analysis system;
determining that a second analytical object partially consistent with the resolved-request data is available in the low-latency data access and analysis system, wherein the second analytical object is consistent with a first portion of the resolved-request data;
obtaining first object-visualization data representing the second analytical object;
generating candidate object-morphism data with respect to the second analytical object in accordance with a second portion of the resolved-request data;
outputting object-search response data including the first object-visualization data and the candidate object-morphism data for presentation to a user;
obtaining second data expressing usage intent with respect to the low-latency data access and analysis system, the second data expressing usage intent indicating a selected object morphism from the candidate object-morphism data;
generating a third analytical object in accordance with the second analytical object and the selected object morphism; and
outputting response data, responsive to the second data expressing usage intent, including second object-visualization data representing the third analytical object for presentation to the user.

2. The non-transitory computer-readable storage medium of claim 1 , wherein:
determining that the second analytical object partially consistent with the resolved-request data is available in the low-latency data access and analysis system includes:
determining that a plurality of analytical objects respectively partially consistent with the resolved-request data is available in the low-latency data access and analysis system, wherein the plurality of analytical objects includes the second analytical object; and

generating the candidate object-morphism data includes generating respective candidate object-morphism data for respective analytical objects from the plurality of analytical objects.

3. The non-transitory computer-readable storage medium of claim 1 , wherein generating the candidate object-morphism data includes:
determining that descriptive data for the second analytical object is consistent with the first portion of the resolved-request data.

4. The non-transitory computer-readable storage medium of claim 1 , wherein generating the candidate object-morphism data includes:
obtaining predicate resolved-request data for the second analytical object, wherein the second analytical object was generated in accordance with the predicate resolved-request data.

5. The non-transitory computer-readable storage medium of claim 4 , wherein generating the candidate object-morphism data includes:
obtaining a sequence of one or more phrases from the predicate resolved-request data, wherein a respective phrase from the sequence of one or more phrases includes a sequence of one or more tokens; and
obtaining one or more candidate object morphisms by object-morphism phrasing using the sequence of one or more phrases as current phrases and using the resolved-request data as a current sequence of tokens, wherein object-morphism phrasing includes:
identifying a current phrase from the current phrases in sequential order, wherein the current phrase includes a current cardinality of tokens;
identifying one or more current token sequences from the current sequence of tokens, wherein a respective current token sequence from the current token sequences includes the current cardinality of tokens;
identifying semantic-similarity scores for the current token sequences, wherein a respective semantic-similarity score for a respective current token sequence from the current token sequences indicates a determined semantic similarity between the respective current token sequence and the current phrase;
identifying a current token sequence from the current token sequences corresponding to a maximal semantic-similarity score from the semantic-similarity scores;
in response to a determination that the maximal semantic-similarity score is less than a defined semantic similarity threshold:
identifying a data source indicated by the predicate resolved-request data with respect to the current phrase; and
determining that the data source includes a value matching the current token sequence and in response to a determination that the data source includes the value matching the current token sequence, including, in the candidate object morphisms, a candidate object morphism that includes the current token sequence;

obtaining, as updated current phrases, a difference between the current phrases and the current phrase;
obtaining, as an updated current sequence of tokens, a difference between the current sequence of tokens and the current token sequence; and
on a condition that the updated current phrases include at least one phrase:
object-morphism phrasing using the updated current phrases as the current phrases and using the updated current sequence of tokens as the current sequence of tokens.

6. The non-transitory computer-readable storage medium of claim 5 , wherein identifying the semantic-similarity scores includes:
obtaining a first dense vector of sentence embeddings representing for the current phrase; and
for a respective current token sequence from the current token sequences, obtaining a corresponding semantic-similarity score by:
obtaining a second dense vector representing the respective current token sequence; and
determining, as the corresponding semantic-similarity score, a cosine similarity between the first dense vector and the second dense vector.

7. The non-transitory computer-readable storage medium of claim 5 , wherein including the candidate object morphism in the candidate object morphisms data includes:
including, in the candidate object morphism, the predicate resolved-request data and the current token sequence;
including, in the candidate object morphism, a first portion of the predicate resolved- request data and the current token sequence, such that the second portion of the predicate resolved-request data is omitted from the candidate object morphism; or
including, in the candidate object morphism, the predicate resolved-request data and second object-visualization data, such that the first object-visualization data is omitted from the candidate object morphism.

8. A method comprising:
generating object morphisms during object search in a data access and analysis system, wherein generating the object morphisms includes:
in response to obtaining, by the data access and analysis system, resolved request data expressing, in accordance with a defined data-analytics grammar implemented by the data access and analysis system, a request to search the data access and analysis system for an analytical object, generating candidate object morphism data for a first analytical object available in the data access and analysis system;
generating a second analytical object in accordance with the first analytical object and a selected ob

Assignees

• Thoughtspot Inc

Inventors

• Kothari Rakesh
• Yadav Nikhil
• Balakavi Manikanta Venkata Rahul
• Kao Hungpin

Classifications

• G06F16/248

  Presentation of query results · CPC title

• G06F16/254

  Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses · CPC title

• G06F16/34

  Browsing; Visualisation therefor (browsing or visualisation for clustering or classification G06F16/358) · CPC title

• G06F16/2365 Primary

  Ensuring data consistency and integrity · CPC title

• G06F16/316 Primary

  Indexing structures · CPC title