Automatic database analysis

Patent metadata

Field	Value
Publication number	US-2019108230-A1
Application number	US-201816156414-A
Country	US
Kind code	A1
Filing date	Oct 10, 2018
Priority date	Oct 10, 2017
Publication date	Apr 11, 2019
Grant date	—

Abstract

A method and system may be implemented for automatically analyzing data in a database. The method and system may receive a current context of the database. The method and system may identify one or more columns of utility based on the current context and generate a current context based on the one or more columns of utility. The method and system may generate one or more exploration queries. The method and system may explore the one or more exploration queries to generate an exploration result set. The method and system may generate one or more insights. The one or more insights may be based on the current context, the exploration result set, or both. The method and system may rank the insights. The method and system may display, transmit, or store the one or more insights based on the rank.

First claim

What is claimed is:

1 . A method for analyzing data in a database, the method comprising:
identifying a current context for accessing data from a low-latency database, wherein the current context includes a requested result set satisfying a requested search criterion; generating an exploration query based on the current context, wherein generating the exploration query includes identifying a column from the low-latency database as a column of utility in response to a determination that a probabilistic utility for the column satisfies a defined utility criterion, wherein the probabilistic utility for the column is based on the current context; generating an exploration result set based on the exploration query; generating a plurality of insights based on the exploration result set; ranking the plurality of insights; and outputting at least one insight from the plurality of insights based on the ranking.

2 . The method of claim 1 , wherein identifying a column of utility further comprises:
incorporating a user preference to identify the column of utility.

3 . The method of claim 2 , wherein the user preference is associated with a user, a group of users, or all users.

4 . The method of claim 1 , wherein identifying a column of utility further comprises:
incorporating system usage data to identify the column of utility.

5 . The method of claim 4 , wherein the system usage data includes a count of a search term has been used.

6 . The method of claim 1 , wherein identifying a column of utility further comprises:
incorporating user feedback to identify the column of utility.

7 . The method of claim 1 , wherein the plurality of insights are ranked by statistical significance.

8 . The method of claim 1 , wherein the plurality of insights are personalized for each user based on a search history of the user, a user profile, a group profile, or a data characteristic.

9 . The method of claim 1 , further comprising:
identifying an algorithm of utility based on the current context; and applying the algorithm of utility to generate the exploration query.

10 . The method of claim 9 , wherein the algorithm of utility is an outlier detection algorithm, a cross correlation algorithm, a trend analysis algorithm, or a comparative analysis algorithm.

11 . The method of claim 9 , further comprising:
updating the algorithm of utility based on the generated insight.

12 . The method of claim 9 , further comprising:
updating the algorithm of utility based on user feedback.

13 . The method of claim 1 , wherein the insight includes a visualization.

14 . The method of claim 1 , wherein the insight includes a natural language narrative that explains what is meaningful in the data.

15 . A system for generating an insight, the system comprising:
an insight unit configured to:
identify a current context for accessing data from a low-latency database, wherein the current context includes a requested result set satisfying a requested search criterion;
generate an exploration query based on the current context, wherein generating the exploration query includes identifying a column from the low-latency database as a column of utility in response to a determination that a probabilistic utility for the column satisfies a defined utility criterion, wherein the probabilistic utility for the column is based on the current context;
generate an exploration result set based on the exploration query;
generate a plurality of insights based on the exploration result set; and
rank the plurality of insights; and
output at least one insight from the plurality of insights based on the ranking.

16 . The system of claim 15 , wherein the insight unit is further configured to incorporate a user preference to identify the column of utility.

17 . The system of claim 15 , wherein the insight unit is further configured to incorporate system usage data to identify the column of utility.

18 . The system of claim 15 , wherein the insight unit is further configured to:
identify an algorithm of utility based on the current context; and apply the algorithm of utility to generate the exploration query.

19 . The system of claim 15 , wherein the display is configured to display the insight, wherein the insight includes a visualization.

20 . The of claim 15 , wherein the display is configured to display the insight, wherein the insight includes a natural language narrative that explains what is meaningful in the data.

Assignees

• Thoughtspot Inc

Inventors

• Prakash Amit
• Chen Antony Chuxiao
• Jha Gunjan
• Jaggi Jasmeet Singh
• Ghosh Manoj Krishna
• Piratla Pavan Ram
• Dorairaj Pradeep
• Agrawal Sanjay

Classifications

• G06F17/30554

  Physics · mapped topic

• G06F17/3053 Primary

  Physics · mapped topic

• G06F17/30315

  Physics · mapped topic

• G06F16/2462

  Approximate or statistical queries · CPC title

• G06F16/248

  Presentation of query results · CPC title