IDENTIFYING ON A GRAPHICAL DEPICTION CANDIDATE POINTS AND TOP-MOVING QUERIES

• US 20090303238A1
• Filed: 06/06/2008
• Published: 12/10/2009
• Est. Priority Date: 06/06/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: LAM, ANDY, CRASWELL, NICHOLAS ERIC, WILLIAMS, HUGH EVAN, BUCKLEY, JAMIE P., Govani, Tabreez
• Granted Patent: US 8,803,882 B2
• Time in Patent Office: Days
• US Class Current: 345/440
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Computer-readable media and computerized methods for identifying candidate points on a graphical depiction of relative popularity of an entity (e.g., entertainer, sports team, and the like) are provided. Points on the graphical depiction are ranked based on a number of user-submitted web queries that reference the entity that are received during a particular time frame. Peak points and slope values (i.e., derived from an angle of inclination of inclines on the graphical depiction) may be captured by analyzing movements in the rank of an entity over time. An algorithmic process may then be applied to the peak points and slope values to determine points of interest of the entity'"'"'s popularity, such as the highest-ranked periods and/or dramatic positive movements in rank of the entity. These points of interest are selected as candidate points and are surfaced as icons on a visual representation of the graphical depiction.

Claims

1. One or more computer-readable media having computer-executable instructions embodied thereon that, when executed, perform a method for automatically selecting one or more candidate points on a graphical depiction of user-submitted queries referencing an entity, the method comprising:

receiving a user-initiated selection of the entity and a time frame in which the user-submitted queries referencing the entity are to be retrieved, wherein the time frame is divided into temporal increments;

providing the graphical depiction of a rank of the entity over the time frame, wherein the rank represents a relative popularity of the entity, in relation to a plurality of entities of a predefined category, as determined from the user-submitted queries collected at a search engine;

automatically initiating a selection procedure for identifying the one or more candidate points on the graphical depiction that are each associated with a positive movement of the rank, wherein the selection procedure comprises;

(i) identifying locations on the graphical depiction between adjacent inclines and declines as peak points and valley points;

(ii) ascertaining the rank at the temporal increments corresponding with each of the peak points and the valley points;

(iii) deriving slope values based on an angle of inclination of the inclines; and

(iv) based on the rank of the peak points and the valley points, and the derived slope values, selecting the one or more candidate points; and

surfacing the one or more selected candidate points by presenting, at a user-interface (UI) display, an indicator of each of the one or more selected candidate points on the graphical depiction.

2. The one or more computer-readable media of claim 1, wherein the plurality of entities of the predefined category comprise substantially comparable subjects of queries grouped according to at least one of people, sports teams, products, geographic locations, music, companies, tools, or types of entertainers.

3. The one or more computer-readable media of claim 1, wherein the temporal increments of the graphical depiction represent at least one of months, weeks, days, or units of time that uniformly divide the time frame.

4. The one or more computer-readable media of claim 1, wherein selecting the one or more candidate points comprises:

comparing the rank associated with each of the peak points against each other to determine a highest ranked peak point; and

identifying the highest ranked peak point with the time frame as one of the one or more candidate points.

5. The one or more computer-readable media of claim 4, wherein the method further comprises:

determining that a plurality of the peak points share the highest ranking;

tracing backward along the graphical depiction to ascertain a valley point immediately preceding each of the plurality of peak points;

calculating a rank differential between the valley point associated with each of the plurality of peak points; and

identifying a peak point of the plurality of peak points associated with a largest rank differential as one of the one or more candidate points, thereby capturing the peak point on the graphical depiction having a greatest positive movement in the rank with respect to a precedent valley point.

6. The one or more computer-readable media of claim 4, wherein comparing the rank associated with each of the identified temporal increments comprises:

determining that a plurality of the peak points share the highest ranking;

selecting a peak point, of the plurality of the highest ranked peak points, that is latest in time within a scope of the time frame; and

identifying the latest peak point as one of the one or more candidate points.

7. The one or more computer-readable media of claim 1, wherein selecting the one or more candidate points comprises:

identifying a peak point that is latest in time within a scope of the time frame;

ascertaining that the slope value of an incline preceding the latest peak point is greater than a predefined threshold slope to ensure sufficient positive movement of the rank is associated with the latest peak point, and

identifying the latest peak point as one of the one or more candidate points.

8. The one or more computer-readable media of claim 1, wherein selecting the one or more candidate points to feature comprises:

identifying a remainder of the peak points that are not identified as one of the one or more candidate points;

comparing the slope values of the inclines adjacent to the remaining peak points, and

identifying a peak point of the remaining peak points adjacent to an incline having a greatest slope value as one of the one or more candidate points.

9. The one or more computer-readable media of claim 8, wherein the method further comprising recursively identifying a peak point of the remaining peak points as one of the one or more candidate points until a predefined number of candidate points are accumulated.

10. The one or more computer-readable media of claim 8, wherein the method further comprising:

batching the user-submitted queries received during a scope of a temporal increment associated with a peak point identified as one of the one or more candidate points into query groupings based on commonality of the search terms therein;

enumerating a number of the user-submitted queries encompassed by each of the query groupings;

determining that the number of the user-submitted queries associated with each of the query groupings fails to meet a predefined threshold; and

withdrawing the identified peak point from the one or more candidate points.

11. The one or more computer-readable media of claim 1, wherein presenting an indicator for each of the one or more candidate points at a user-interface (UI) display of the graphical depiction comprises surfacing an icon substantially overlaying the peak points associated with each of the one or more candidate points, wherein the graphical depiction is a line graph indicating a movement in rank of the entity over time.

12. The one or more computer-readable media of claim 11, wherein the icon exposes information related to its respective peak point upon detecting a user-initiated indication thereof, wherein exposing information comprises causing an explanation object to be displayed in association with the icon, and wherein the explanation object communicates at least one representation of a user-submitted query to rationalize the indicated peak point.

13. A computerized method, implemented at a computing device, for selecting one or more candidate points based on a ranking of an entity, where the ranking is rendered at temporal increments that uniformly divide a predetermined time frame;

the method comprising;

ascertaining a rank associated with each of the temporal increments within the predetermined time frame, wherein the rank represents a relative popularity of the entity, as determined from a number of queries referencing the entity received at a search engine, in relation to a plurality of entities within a predefined category;

identifying a temporal increment having a rank associated therewith that is higher than the rank associated with a remainder of the temporal increments and selecting the identified temporal increment for inclusion into the one or more candidate peak points;

determining that a rank of a temporal increment immediately preceding a latest temporal increment, within a scope of the predetermined time frame, is lower than a rank of the latest temporal increment and selecting the latest temporal increment for inclusion into the one or more candidate points;

identifying positive movements of the rank of the entity that occur between the temporal increments based on a comparison of the rank associated with each of the temporal increments and a rank associated with a chronologically-preceding temporal increment;

comparing the identified positive movements against each other to distinguish a largest positive movement;

selecting a temporal increment at a culmination of the largest positive movement for inclusion into the one or more candidate points; and

at least temporarily, storing the one or more candidate points, in association with the corresponding temporal increments, at a data store.

14. One or more computer-readable media having computer-executable instructions embodied thereon that, when executed, perform a method for automatically selecting one or more top-moving queries associated with one or more candidate points on a graphical depiction of user-submitted queries referencing an entity, the method comprising:

retrieving the user-submitted queries containing search terms referencing the entity within a time frame of the graphical depiction;

identifying the one or more candidate points associated with a positive movement of relative popularity of the entity;

invoking a selection procedure to determine which of the retrieved user-submitted queries is associated with the one or more candidate points to surface at a graphical user interface (GUI), the selection procedure comprising;

(i) assigning the retrieved user-submitted queries to query groups according to the search terms that are common to the retrieved user-submitted queries, and enumerating the user-submitted queries assigned to each of the query groups;

(ii) removing from consideration the query groups associated with the common search terms that occur less than a predefined threshold;

(iii) ranking the query groups based on the enumeration such that the query groups associated with the common search terms that occur frequently are ranked highly; and

(iv) applying one or more filters to the ranked query groups to identify the one or more top-moving queries therefrom; and

at least temporarily, storing the one or more top-moving queries in association with the one or more candidate points.

15. The one or more computer-readable media of claim 14, wherein ranking the query groups based on the enumeration comprises:

assigning previously received user-submitted queries to prior query groups according to the search terms that are common to the previously received user-submitted queries, and enumerating the previously received user-submitted queries assigned to each of the prior query groups;

mapping the query groups to the prior query groups based on a determination that the common search terms of a query group correspond to the common search terms of a prior query group;

calculating a ratio of a number of the user-submitted queries in the query group over a number of the previously-received user-submitted queries in the prior query group that is mapped to the query group;

calculating a query score by multiplying the number of the user-submitted queries in the query group by the ratio; and

ranking the query groups based on a comparison of the calculated query score against each other.

16. The one or more computer-readable media of claim 14, wherein the method further comprises:

selecting a predefined number of query groups that are high in the ranking;

identifying the common search terms associated with the selected query groups as the one or more top-moving queries; and

surfacing the top-moving queries at the GUI in a location proximate to the graphical depiction.

17. The one or more computer-readable media of claim 14, wherein applying one or more filters to the ranked query groups to identify the one or more top-moving queries therefrom comprises:

determining that the common search terms associated with a query group are substantially comparable to a title of the entity; and

excluding the comparable common search terms from consideration for identification as the one or more top-moving queries to be surfaced.

18. The one or more computer-readable media of claim 14, wherein applying one or more filters to the ranked query groups to identify the one or more top-moving queries therefrom comprises:

accessing a manifest of phrases and terms designated as objectionable;

comparing the common search terms associated with the ranked query groups against the objectionable phrases and terms to detect a match;

excluding the matching common search terms from consideration for identification as the one or more top-moving queries to be surfaced.

19. The one or more computer-readable media of claim 14, wherein applying one or more filters to the ranked query groups to identify the one or more top-moving queries therefrom comprises:

accessing a manifest of commonly used keywords;

comparing the common search terms associated with the ranked query groups against the commonly used keywords to detect a match; and

excluding the matching common search terms from consideration for identification as the one or more top-moving queries to be surfaced.

20. The one or more computer-readable media of claim 14, wherein applying one or more filters to the ranked query groups to identify the one or more top-moving queries therefrom comprises:

comparing the common search terms identified as the one or more top-moving queries against an unselected common search term;

based on the comparison, determining that an edit distance between the common search terms identified as the one or more top-moving queries and the unselected common search term is below a predefined threshold; and

removing the unselected common search term from consideration for identification as one of the one or more top-moving queries.