User interface for visualizing search data

Patent metadata

Field	Value
Publication number	US-2019324981-A1
Application number	US-201916378198-A
Country	US
Kind code	A1
Filing date	Apr 8, 2019
Priority date	Apr 20, 2018
Publication date	Oct 24, 2019
Grant date	—

Abstract

A user interface (UI) for visualizing search data provides techniques for grouping and organizing aggregate data that shows the categories of topics included in search queries from a large number of individual users. Raw search queries are categorized into one of a number of topical categories. The search queries are assigned to a geographic location based on geolocations of computing devices generating the search queries. The UI presents a map that shows the number of search queries per topical category for each geographic location displayed in the current UI view. As a result of this UI design, a user can easily understand the interaction between geographic location and frequency of search query topics. Trends in the geographic distribution of searches and in the categories of topics searched are also easily understood from this UI design by changing the time range of the search queries displayed.

First claim

1 . A system for classifying and visualizing search queries based on geolocation, the system comprising:
one or more processing units; and memory storing computer-executable instructions that, when executed by the one or more processing units, cause the system to perform acts comprising:
receiving a plurality of raw search queries including query contents and geolocations;
generating word embeddings for the plurality of raw search queries;
classifying the plurality of raw search queries into one of a plurality of categories based on the query contents and the word embeddings;
determining a number of raw search queries for a plurality of geographic regions for respective ones of the plurality of categories based on results of the classifying and the geolocations;
generating a user interface (UI) indicating a metric derived at least in part from the number of the raw search queries for the plurality of categories for ones of the geographic regions represented in the UI; and
modifying the UI by applying a filter which removes data values for the geographic regions represented in the UI based on a criterion of the filter.

2 . The system of claim 1 , wherein the raw search queries comprise job search queries and the plurality of categories comprise job categories.

3 . The system of claim 1 , wherein the criterion of the filter comprises a socio-economic dimension of the geographic regions represented in the UI.

4 . The system of claim 1 , wherein the raw search queries comprise Internet searches and the geolocations are determined by reverse Internet protocol (IP) lookup.

5 . The system of claim 1 , wherein the raw search queries further include timestamps and the UI displays a change over time of the frequency of raw search queries per respective ones of the plurality of categories.

6 . The system of claim 1 , wherein the UI comprises a map including at least a subset of the plurality of geographic regions and the metric of the raw search queries is represented in the UI by visual characteristics of individual ones of the subset of the plurality of geographic regions.

7 . The system of claim 1 , wherein receiving the plurality of raw search queries occurs in substantially real-time as the raw search queries are generated and the UI indicating the metric of the raw search queries is updated substantially in real-time.

8 . A method for classifying a raw search query:
identifying the raw search query as belonging to a general class of queries based on a keyword present in query contents of the raw search query; representing the raw search query as a multidimensional feature vector; and classifying the multidimensional feature vector into one of a plurality of categories using a machine learning classifier.

9 . The method of claim 8 , further comprising removing stop words and the keyword from the query contents of the raw search query wherein the identifying word embeddings is performed on the raw search query after removal of the stop words and the keyword.

10 . The method of claim 8 , wherein the raw search query is a query directed to an Internet search engine and the word embeddings are generated by a neural network trained on other Internet search queries.

11 . The method of claim 8 , wherein the machine learning classifier is a support vector machine trained on labeled data.

12 . The method of claim 8 , wherein the plurality of categories includes architecture/engineering, art, business, construction, education, finance, food, healthcare, leisure/hospitality, manufacturing, retail, science, technology, and transportation.

13 . The method of claim 8 , wherein the raw search query comprises a geolocation and further comprising:
assigning the raw search query to a geographic region based on the geolocation; and counting a total number of search queries, including the raw search query, in the geographic region that are classified in a same one of the plurality of categories.

14 . A computer-readable storage medium containing computer-readable instructions that, when executed by one or more processing units, cause the one or more processing units to perform acts comprising:
receiving categorized search data comprising categories and geolocations for a plurality of searches; grouping the categorized search data into a plurality of geographic regions based on the geolocations; receiving a selection of a single category of the categories; assigning visual characteristics to the geographic regions, the visual characteristics representing a frequency of searches in the plurality of searches categorized in the single category relative to the frequency of searches in the single category in other geographic regions; and generating a user interface (UI) including representations of the geographic regions with the visual characteristics.

15 . The computer-readable storage medium of claim 14 , wherein the UI includes a map portion and the geographic regions shown in the map portion comprise one of a country, a state, a province, a county, a metropolitan region, or a city.

16 . The computer-readable storage medium of claim 14 , wherein the UI includes a map portion and a unit of the geographic regions changes with a zoom level applied to the map portion.

17 . The computer-readable storage medium of claim 14 , wherein the categorized search data is categorized using a machine learning classifier that analyzes word embeddings generated from query contents of the plurality of searches.

18 . The computer-readable storage medium of claim 14 , wherein the categorized search data further comprises timestamps and the computer-readable instructions cause the one or more processing units to perform further acts comprising:
receiving an indication of a time period; and changing the visual characteristics based on the time period.

19 . The computer-readable storage medium of claim 14 , wherein the computer-readable instructions cause the one or more processing units to perform further acts comprising:
receiving a selection of a criterion of a filter; and modifying the UI by applying a first modified visual characteristic to a first subset of the geographic regions associated with an upper range of the criterion of the filter and applying a second modified visual characteristic to a second subset of the geographic regions associated with a lower range of the criterion of the filter.

20 . The computer-readable storage medium of claim 14 , wherein the UI includes a map portion and a text portion and the computer-readable instructions cause the one or more processing units to perform further acts comprising:
receiving an indication of one or more words in the text portion; and changing the map portion of the UI by changing the selection of the single category, a criterion of a filter applied to the map portion of the UI, or the geographic regions included within the map portion of the UI.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Counts Scott
• Moeur Roberta R
• Von Veh Curtis N
• Cranshaw Justin Brooks
• Chancellor Stevie Nicole
• Carbary Anthony

Classifications

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N3/08

  Learning methods · CPC title

• G06F18/2431

  Multiple classes · CPC title

• G06F16/358 Primary

  Browsing; Visualisation therefor · CPC title

• G06F40/30

  Semantic analysis · CPC title