SEARCH ADVERTISEMENT SELECTION BASED ON USER ACTIONS
BACKGROUND
[0001] Advertising is an established way for merchants to reach an audience of
potential customers. Ideally, merchants desire to have their advertisements directed to
individuals that are likely to purchase the merchants' products and services. Accordingly,
advertising has gone through an evolution of targeting approaches. Initially, advertising
was often generally untargeted. However, over time, merchants began to target potential
customers in a number of ways. For instance, merchants often target a particular audience
by selectively choosing the medium (e.g., newspapers, magazines, radio, television, etc.)
in which to advertise. Additionally, advertising is often targeted by selecting the particular
environment (e.g., specific shows on television, specific web pages on the Internet, etc.)
within which the advertisements will appear.
[0002] Search advertising, in which advertisements are presented in conjunction
with search results in response to user search queries, is a more recent form of advertising.
Search advertising is often considered to be a targeted form of advertising because the
advertisements are directed to the specific user search queries. In particular, when an end
user submits a search query to a search engine, keywords are identified based on the terms
of the search query and/or based on content of the search results. The keywords are used
for selecting advertisements that are presented in conjunction with the search results on a
search results page returned in response to the end user's query. Although advertisements
are selected based on relevance to the end users' search queries, end users may not be
interested in the advertisements selected and presented in conjunction with the search
results. Nonetheless, the advertisements are static on the search results page and only
reflect relevance to the search query.
SUMMARY
[0003] This summary is provided to introduce a selection of concepts in a
simplified form that are further described below in the Detailed Description. This
summary is not intended to identify key features or essential features of the claimed
subject matter, nor is it intended to be used as an aid in determining the scope of the
claimed subject matter.
[0004] Embodiments of the present invention relate to providing advertisements
based on user activity or inactivity corresponding with search results. Search results are
provided for presentation to an end user and tracked for user activity. If user activity is
detected, new advertisements may be selected based on a user context, which comprises a
context gleaned from the user activity. In some embodiments, the search results are
provided in clusters, and the user context corresponds with search results of a cluster
selected by the end user. In some embodiments, dominant concepts are identified from the
user context using a concept graph, and advertisements are selected based on the dominant
concepts. In further embodiments, new advertisements may be provided in response to the
absence of user activity. If no user activity is detected for a predetermined period of time,
new advertisements may be selected based on the absence of user activity. New
advertisements may be used to replace advertisements currently displayed on the search
results page.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] The present invention is described in detail below with reference to the
attached drawing figures, wherein:
[0006] FIG. 1 is a block diagram of an exemplary computing environment suitable
for use in implementing embodiments of the present invention;
[0007] FIG. 2 is a block diagram of an exemplary system in which embodiments
of the invention may be employed;
[0008] FIG. 3 is a flow diagram showing a method for selecting an advertisement
based on a user context in a search results page in accordance with an embodiment of the
present invention;
[0009] FIG. 4 is a flow diagram showing a method for replacing advertisements in
a search results page based on user context in accordance with an embodiment of the
present invention;
[0010] FIG. 5 is a flow diagram showing a method for providing advertisements
based on user actions corresponding with search result clusters in accordance with an
embodiment of the present invention;
[0011] FIGS. 6A and 6B are illustrative screen displays showing a search results
page with different advertisements based on user context in accordance with an
embodiment of the present invention;
[0012] FIG. 7 is a flow diagram showing a method for selecting keywords for
advertisement selection based on dominant concepts in a user context of a search results
page in accordance with an embodiment of the present invention;
[0013] FIG. 8 is a flow diagram showing a method for replacing advertisements in
a search results page based on user inactivity in accordance with an embodiment of the
present invention; and
[0014] FIG. 9 is a flow diagram showing a method for replacing advertisements in
a search results page based on user activity or inactivity in accordance with an
embodiment of the present invention.
DETAILED DESCRIPTION
[0015] The subject matter of the present invention is described with specificity
herein to meet statutory requirements. However, the description itself is not intended to
limit the scope of this patent. Rather, the inventors have contemplated that the claimed
subject matter might also be embodied in other ways, to include different steps or
combinations of steps similar to the ones described in this document, in conjunction with
other present or future technologies. Moreover, although the terms "step" and/or "block"
may be used herein to connote different elements of methods employed, the terms should
not be interpreted as implying any particular order among or between various steps herein
disclosed unless and except when the order of individual steps is explicitly described.
[0016] As previously indicated, embodiments of the present invention are
generally directed to providing, modifying, and/or updating advertisements for
presentation to end users based on the presence or absence of user activity corresponding
with search results. When a search query is received from an end user, search results
relevant to the user's search query are provided in response. In addition to search results,
advertisements may be selected based on the user's search query and provided for
presentation in conjunction with the search results. The presence or absence of user
activity is then tracked. Embodiments of the present invention track user activity that
occurs after a search query is received from the end user and includes tracking activity that
does not change the context of the original search query. For instance, in some
embodiments, a search results page may be provided that includes search results
aggregated into different clusters, and the end user may select a particular cluster to view
search results on the search results page. The selection doesn't change the context of the
original search query. As such, the end user can perform a number of actions on the search
results page while remaining within the context of the original search query.
[0017] In some instances, the search results and/or advertisements may be
provided to the end user on a search results page. In other instances, the search results
and/or advertisements may be provided in another environment. Accordingly, although
embodiments are primarily discussed herein in which search results and/or advertisements
are provided on a search results page, other embodiments include providing search results
and/or advertisements in other environments. Additionally, although embodiments are
primarily discussed herein in which a search results page is tracked for user activity, other
embodiments include tracking user activity in other environments and/or outside of a
search results page.
[0018] When an end user performs an action on the search results page or
otherwise, a new user context is provided by the user action. For instance, in an
embodiment in which search result clusters are provided, when an end user selects a
particular cluster, the user context may include the selected cluster and/or the search
results of that cluster. However, as noted above, the query context remains the same
regardless of the user selection. The user actions and user context provide an indication of
the end user's interest and therefore provides better information for selecting
advertisements in which the end user may be interested. When a user action is detected, a
user context is identified and used to select new advertisements. In some embodiments,
dominant concepts within the user context are identified using a concept graph and
employed to select advertisements. The newly selected advertisements may replace
advertisements previously presented on the search results page. As such, as the user
performs actions on the search results page, advertisements that correspond with the user
context of the search results page are presented.
[0019] Further embodiments of the present invention select new advertisements for
presentation on a search results page based on the absence of user activity for a
predetermined period of time. This recognizes that if an end user does not select an
advertisement on a search results page after a certain period of time, the end user is not
likely interested in that advertisement. As such, if no user activity is detected within a
search results page for a predetermined period of time, new advertisements are selected
based on the inactivity and are used to replace advertisements currently displayed on the
search results page.
[0020] Accordingly, in one aspect, an embodiment of the present invention is
directed to one or more computer-readable media storing computer-useable instructions
that, when used by one or more computing devices, cause the one or more computing
devices to perform a method. The method includes receiving a search query submitted by
an end user. The method also includes providing a plurality of search results in response
to the search query submitted by the end user, wherein the plurality of search results are
grouped into a plurality of clusters, each cluster including a different subset of search
results. The method further includes identifying a user activity corresponding with one of
the plurality of clusters, the user activity being indicative of user interest in a selected
cluster. The method also includes identifying a user context based on the selected cluster.
The method further includes selecting one or more advertisements for presentation based
on the user context. The method still further includes providing the one or more
advertisements for presentation to the end user.
[0021] In another embodiment, an aspect of the invention is directed to one or
more computer-readable media storing computer-useable instructions that, when used by
one or more computing devices, cause the one or more computing devices to perform a
method. The method includes receiving a search query submitted by an end user and
providing a search results page for presentation to the end user in response to the search
query submitted by the end user, the search results page including a plurality of search
results and an initial advertisement selected based at least in part on the search query
submitted by the end user. The method also includes tracking the search results page for
user activity and detecting a user action by the end user within the search results page.
The method further includes identifying a user context based on the user action in
response to detecting a user action within the search results page. The method also
includes identifying a set of search results within the user context, identifying a plurality
of concepts within text corresponding with the set of search results, and analyzing the
plurality of concepts using a concept graph to identify one or more dominant concepts.
The method further includes selecting one or more keywords based on the one more
dominant concepts and selecting a new advertisement based on the one or more keywords.
The method still further includes replacing the initial advertisement within the search
results page with the new advertisement.
[0022] A further embodiment of the present invention is directed to one or more
computer-readable media storing computer-useable instructions that, when used by one or
more computing devices, cause the one or more computing devices to perform a method.
The method includes receiving a search query submitted by an end user. The method also
includes providing a search results page in response to the search query, the search results
page including a plurality of search results and an initial advertisement. The method
further includes determining that a predetermined time period has expired without
detecting user activity within the search results page. The method also includes selecting a
new advertisement in response to determining that the predetermined time period has
expired without detecting user activity within the search results page. The method further
includes replacing the initial advertisement within the search results page with the new
advertisement.
[0023] Having briefly described an overview of embodiments of the present
invention, an exemplary operating environment in which embodiments of the present
invention may be implemented is described below in order to provide a general context for
various aspects of the present invention. Referring initially to FIG. 1 in particular, an
exemplary operating environment for implementing embodiments of the present invention
is shown and designated generally as computing device 100. Computing device 100 is but
one example of a suitable computing environment and is not intended to suggest any
limitation as to the scope of use or functionality of the invention. Neither should the
computing device 100 be interpreted as having any dependency or requirement relating to
any one or combination of components illustrated.
[0024] The invention may be described in the general context of computer code or
machine-useable instructions, including computer-executable instructions such as program
modules, being executed by a computer or other machine, such as a personal data assistant
or other handheld device. Generally, program modules including routines, programs,
objects, components, data structures, etc., refer to code that perform particular tasks or
implement particular abstract data types. The invention may be practiced in a variety of
system configurations, including hand-held devices, consumer electronics, generalpurpose
computers, more specialty computing devices, etc. The invention may also be
practiced in distributed computing environments where tasks are performed by remoteprocessing
devices that are linked through a communications network.
[0025] With reference to FIG. 1, computing device 100 includes a bus 110 that
directly or indirectly couples the following devices: memory 112, one or more processors
114, one or more presentation components 116, input/output ports 118, input/output
components 120, and an illustrative power supply 122. Bus 110 represents what may be
one or more busses (such as an address bus, data bus, or combination thereof). Although
the various blocks of FIG. 1 are shown with lines for the sake of clarity, in reality, these
blocks represent logical, not necessarily actual, components. For example, one may
consider a presentation component such as a display device to be an I/O component. Also,
processors have memory. We recognize that such is the nature of the art, and reiterate that
the diagram of FIG. 1 is merely illustrative of an exemplary computing device that can be
used in connection with one or more embodiments of the present invention. Distinction is
not made between such categories as "workstation," "server," "laptop," "hand-held
device," etc., as all are contemplated within the scope of FIG. 1 and reference to
"computing device."
[0026] Computing device 100 typically includes a variety of computer-readable
media. Computer-readable media can be any available media that can be accessed by
computing device 100 and includes both volatile and nonvolatile media, removable and
non-removable media implemented in any method or technology for storage of
information such as computer-readable instructions, data structures, program modules or
other data. Computer-readable media includes, but is not limited to, RAM, ROM,
EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks
(DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk
storage or other magnetic storage devices, or any other medium which can be used to store
the desired information and which can be accessed by computing device 100.
Combinations of any of the above should also be included within the scope of computerreadable
media.
[0027] Memory 112 includes computer- storage media in the form of volatile
and/or nonvolatile memory. The memory may be removable, nonremovable, or a
combination thereof. Exemplary hardware devices include solid-state memory, hard
drives, optical-disc drives, etc. Computing device 100 includes one or more processors
that read data from various entities such as memory 112 or I/O components 120.
Presentation component(s) 116 present data indications to an end user or other device.
Exemplary presentation components include a display device, speaker, printing
component, vibrating component, etc.
[0028] I/O ports 118 allow computing device 100 to be logically coupled to other
devices including I/O components 120, some of which may be built in. Illustrative
components include a microphone, joystick, game pad, satellite dish, scanner, printer,
wireless device, etc.
[0029] Referring now to FIG. 2, a block diagram is provided illustrating an
exemplary system 200 in which embodiments of the present invention may be employed.
It should be understood that this and other arrangements described herein are set forth only
as examples. Other arrangements and elements (e.g., machines, interfaces, functions,
orders, and groupings of functions, etc.) can be used in addition to or instead of those
shown, and some elements may be omitted altogether. Further, many of the elements
described herein are functional entities that may be implemented as discrete or distributed
components or in conjunction with other components, and in any suitable combination and
location. Various functions described herein as being performed by one or more entities
may be carried out by hardware, firmware, and/or software. For instance, various
functions may be carried out by a processor executing instructions stored in memory.
[0030] Among other components not shown, the system 200 may include a user
device 202, a search engine 204, and an advertisement system 206. Each of the
components shown in FIG. 2 may be embodied on any type of computing device, such as
computing device 100 described with reference to FIG. 1, for example. The components
may communicate with each other via a network 208, which may include, without
limitation, one or more local area networks (LANs) and/or wide area networks (WANs).
Such networking environments are commonplace in offices, enterprise-wide computer
networks, intranets, and the Internet. It should be understood that any number of user
devices, search engines, and advertisement systems may be employed within the system
200 within the scope of the present invention. Each may comprise a single device or
multiple devices cooperating in a distributed environment. For instance, the system 200
may comprise multiple devices arranged in a distributed environment that collectively
provide the functionality of the search engine 204 and advertisement system 206 described
herein. Additionally, other components not shown may also be included within the system
200.
[0031] In accordance with embodiments of the present invention, a user may
employ the user device 202 to submit search queries to the search engine 204 and, in
response, receive search results and advertisements. For instance, the user may employ a
web browser on the user device 202 to access a search input web page and enter a search
query. As another example, the user may enter a search query via a search input box
provided by a search engine toolbar located, for instance, within a web browser, the
desktop of the user device 202, or other location. One skilled in the art will recognize that
a variety of other approaches may also be employed for providing a search query within
the scope of embodiments of the present invention.
[0032] When the search engine 204 receives a search query from a user device,
such as the user device 202, the search engine 204 performs a search on a search system
index 210 and/or other data storage containing searchable content. Accordingly, the search
engine 204 identifies a number of search results in response to the received search query.
In some embodiments, as will be described in further detail below, the search engine 204
may also aggregate the search results into a number of clusters and provide a search results
page that allows an end user to navigate the clusters.
[0033] Advertisements are also selected by the advertisement system 206 for
inclusion in the search results page as will be described in further detail below. A search
results page that includes search results with selected advertisements is returned to the user
device 202 in response to the search query. As will be described in further detail below,
the end user may perform actions on the search results page or otherwise that trigger new
user contexts. Each time a new user context is identified, the advertisement system 206
selects new advertisements for presentation on the search results page. Additionally, the
advertisement system 206 may identify user inactivity for a threshold period of time and
select new advertisements for presentation based on the absence of user activity.
[0034] As shown in FIG. 2, the advertisement system 206 includes a query context
identifying component 212, a user activity tracking component 214, a user context
identifying component 216, a keyword selection component 218, an advertisement
selection component 220, and an advertisement delivery engine 222. The query context
identifying component 212 operates to identify the query context for a search query
received from a user device, such as the user device 202. The query context may include
one or more search terms of the search query. In some embodiments, the query context
may include other information, such as, for instance, the time at which the search query
was submitted or the location of the end user submitting the search query. Other
information may also be identified as part of the query context. The query context may be
used for advertisement selection. For instance, when a search query is received, the query
context may be used to select initial advertisements for presentation on the search results
page before any user action is identified. In some embodiments, the query context may
also be used to select advertisements in conjunction with a user context identified after a
user action on the search results page.
[0035] The user activity tracking component 214 operates to track user activity on
the search results page. The identification of user activity on the search results page
triggers identification of a user context by the user context identifying component 216 and
selection of new advertisements by the advertisement selection component 220. Any user
action within a search result page or otherwise that provides further context for selecting
advertisements may be employed within embodiments of the present invention. By way
of example only and not limitation, in some embodiments, the search results page may
aggregate search results into different clusters. When an end user selects to view a
particular cluster, the selection of that cluster is viewed as a user action providing context
for selecting advertisements. In particular, the search results within the selected cluster
provide further context for advertisement selection. As another example, an end user may
perform an action that indicates interest in a particular search result, such as hovering a
cursor over a search result or selecting a control causing additional information regarding
a search result to be displayed within the search results page. The user action would
indicate end user interest in the particular search result providing further context for
advertisement selection. A variety of additional user actions within a search results page
may also be used to provide further context for advertisement selection in accordance with
various embodiments of the present invention. Additionally, in some embodiments user
actions outside a search results page may be tracked and used to determine the user
context. For instance, an end user's physical change in location (e.g., as tracked by a GPS
function of a mobile device) may be an action that defines a new user context.
[0036] In some embodiments, the user activities to be tracked by the user activity
tracking component 214 to determine whether any user action has occurred may be
configurable by the search system provider. For instance, some user activities, such as
scrolling up and down a window may not be considered as a user action within the search
results page. However, other user activities such as selecting a search result or an
advertisement would be considered a user action. In some embodiments, only actions that
change the context of the search results page are considered and tracked.
[0037] The keyword selection component 218 selects keywords based on the user
context identified by the user context identifying component 216. In some embodiments,
the keyword selection component 218 selects keywords based only on a current user
context. In other embodiments, the keyword selection component 218 selects keywords
based on previous user contexts in addition to the current user context. For instance, an
end user may perform multiple actions within a search results page, each action resulting
in a different user context. As noted above, the query context may be also be used in
conjunction with the user context to select keywords.
[0038] In some embodiments of the present invention, the keyword selection
component 218 selects keywords by identifying dominant concepts within the user context
using a concept graph stored in a knowledge base 224, as will be described in further
detail below. The concept graph comprises a collection of concepts and relationships
among the concepts. In particular, concepts are represented as nodes within the concept
graph and relationships among the concepts are represented as edges between the nodes.
Additionally, the edges are configured to represent the strength of relationships between
concepts. The concept graph may be generated by mining data from a corpus of
documents to identify concepts and the relationships of concepts and stored in the
knowledge base 224.
[0039] The advertisement selection component 220 queries an advertisement
inventory 226 using selected keywords to select advertisements that will be delivered for
presentation. The advertisement inventory 226 may store advertisements and metadata
associated with each advertisement. The metadata stored for an advertisement may include
information used in advertisement selection, such as for instance, bid values from
advertisers, click-through rates, etc. In embodiments of the present invention, the
advertisement selection component 220 selects advertisements based on relevance of the
advertisements to selected keywords and/or based on monetization (i.e., an estimate of the
extent to which the advertisements will generate advertising revenue). Auction processes
currently employed by advertising systems may be used for selecting advertisements.
Such processes are well known to those skilled in the art and therefore will not be
discussed in further detail herein.
[0040] The advertisement system 206 also includes an advertisement delivery
engine 222 that facilitates delivery of selected advertisements for presentation. In
embodiments, a search results page delivered to a user device, such as the user device 202,
is configured such that advertisements presented on the search results page may be
replaced. As such, when user activity is identified on the search results page and a new
user context is determined, new advertisements are selected and delivered by the
advertisement delivery engine 222 to replace previous advertisements presented on the
search results page.
[0041] Turning to FIG. 3, a flow diagram is provided illustrating an overall
method 300 for selecting an advertisement based on user actions within a search results
page in accordance with an embodiment of the present invention. As shown at block 302,
the method 300 includes identifying a query context. As discussed previously, the query
context comprises information regarding the search query submitted by the end user. In
some embodiments, the query context includes the search terms entered by the end user as
a search query. In addition to or in lieu of search terms entered by the end user, the query
context in some embodiments may include other information regarding a user search, such
as, for instance, the time an end user submits a search or the location of the end user
submitting the search. A variety of additional information regarding a search query
submitted by an end user may be identified as part of the query context in various
embodiments of the present invention.
[0042] In addition to identifying a query context, a user context is identified, as
shown at block 304. The user context is based on user actions within a search results page
or other environment. Any user action that provides further context for selecting
advertisements may be employed within some embodiments of the present invention. By
way of example only and not limitation, in some embodiments, the search results page
may aggregate search results into different clusters. When an end user selects to view a
particular cluster, the selection of that cluster is viewed as a user action providing context
for selecting advertisements. The search results within the selected cluster may provide
further context for advertisement selection. As another example, an end user may perform
an action that indicates interest in a particular search result, such as hovering a cursor over
a search result or selecting a control to cause additional information regarding a search
result to be displayed within the search results page. The user action would indicate end
user interest in the particular search result providing further context for advertisement
selection. A variety of additional user actions within a search results page or other
environments may also be used to provide further context for advertisement selection in
accordance with various embodiments of the present invention.
[0043] In some embodiments, only the current user context corresponding with a
current user action is identified at block 304. In other embodiments, a history of user
contexts corresponding with a history of user actions within the search results page is
identified at block 304. For instance, in an embodiment in which the search results page is
provided with a number of clusters of search results, the end user may sequentially select
clusters to view the subset of search results associated with each cluster. Each time the
end user selects a cluster, a new user context is provided. As such, in some embodiments,
the process may identify not only the current user context at block 304 based on the cluster
currently selected by the end user, but also previous user contexts corresponding with
clusters previously selected by the end user.
[0044] As shown at block 306, an advertisement is selected based on the identified
query context and user context. Advertisements may be selected based on the query
context and user context in a variety of different manners. In accordance with some
embodiments of the present invention, one or more keywords are selected based on the
query context and user context. For instance, in embodiments in which an end user selects
a cluster of search results, keywords that are relevant to the original search query and the
search results within the selected cluster are identified. As another example, in
embodiments in which a user action indicates end user interest in a particular search result,
keywords relevant to the original search query and identified search result are identified.
[0045] Keywords may be selected based on the current user context or based on a
history of user contexts within the search result page. As an example of the former,
keywords may be selected based only on a subset of search results corresponding with a
currently selected cluster. As an example of the latter, keywords may be selected based on
the subset of search results corresponding with each cluster the end user has selected
within the search results page. In some embodiments, different weightings may be applied
to each user context in the history of user contexts. For instance, a weighting may be
applied based on the order of the user contexts in the history. Earlier user contexts may be
given less weight than more recent user contexts, and the current user context may be
given the greatest weighting. As another example, a weighting may be applied based on a
length of time the end user views each user context. The longer the end user views a user
context within the search results page, the greater the weighting for that user context. A
variety of other weightings may be applied in various embodiments of the present
invention.
[0046] In some embodiments, keyword selection may also take into account
monetization considerations. In particular, the process may select keywords that will
result in the selection of advertisements that are determined to be more likely to generate
greater advertising revenue.
[0047] One or more keywords identified from the query context and user context
are used to select an advertisement for presentation within the search results page. Any
number of advertisements may be selected in accordance with various embodiments of the
present invention. In some embodiments, the top N search results are selected and
presented within the search results page. In other embodiments, search results satisfying a
particular threshold of relevance to the keywords are selected. In further embodiments,
search results having a relevance that is significantly greater than the relevance of other
search results are selected. Any and all such variations are contemplated to be within the
scope of embodiments of the present invention. The selected advertisements are provided
for presentation on the search results page, as shown at block 308. In other embodiments,
the selected advertisements may be presented in another environment.
[0048] In embodiments, advertisements selected based on user actions can be used
to replace advertisements currently displayed on the search results page or other
environment. For instance, as an end user performs various actions within a search results
page, the advertisements on the search results page may be replaced with each user action.
Accordingly, referring next to FIG. 4, a flow diagram is provided that illustrates a method
400 for replacing advertisements displayed on a search results page in accordance with an
embodiment of the present invention. As shown at block 402, a search query is received
that may include one or more search terms. A query context is identified for the search
query, as shown at block 404. The query context may include the one or more search
terms of the search query. In some embodiments, the query context may include
additional information regarding the search, such as, for instance, a time the search query
was submitted or a location of the end user submitting the search query.
[0049] An initial set of one or more advertisements is selected based on the query
context, as shown at block 406. A search results page that includes search results and the
initial set of advertisements is generated and provided to the end user who submitted the
search query, as shown at block 408.
[0050] After the search results page is provided to the end user, the search results
page is tracked for user actions within the search results page, as shown at block 410. In
some embodiments, some user activities, such as scrolling up and down a window may not
be considered as a user action within the search results page. However, other user
activities such as selecting a search result or an advertisement would be considered a user
action. The user activities to be tracked to determine whether any user action has occurred
for the purposes of the method 400 may be configurable by the search system provider.
[0051] A determination is made at block 412 regarding whether a user action has
been made in the search results page. If a user action is not detected, the process continues
to track the search results page for a user action. Alternatively, if a user action is detected,
a user context is identified based on the user action, as shown at block 414.
[0052] A new set of one or more advertisements is selected based on the user
context at block 416. At least one advertisement currently displayed within the search
results page is replaced with a new advertisement, as shown at block 418. In some
embodiments, only a single advertisement is displayed on the search results page, and the
single advertisement is replaced with a new advertisement. In other embodiments, multiple
advertisements are displayed on the search results page. In some instances, all
advertisements are replaced with new advertisements. In other instances, only a portion of
the advertisements are replaced with new advertisements.
[0053] As previously discussed, some embodiments are directed to providing
advertisements for a search results page that aggregates search results into clusters. As
used herein, the term "cluster" is used to refer to any grouping of search results, which
may include, for instance, categories of search results, search result types (e.g., news,
image, etc.), as well as any other search result aggregation. The clusters allow users to
view different search results while remaining within the context of the original search.
[0054] In embodiments, advertisements may be selected based on a user context
resulting from a user action associated with a cluster of search results. FIG. 5 provides a
flow diagram illustrating a method 500 for providing advertisements based on user actions
corresponding with search result clusters in accordance with an embodiment of the present
invention.
[0055] As shown at block 502, a search query is received from an end user.
Search results relevant to the search query are determined, and the search results are
aggregated into a number of clusters, as shown at block 504. A number of techniques may
be employed for aggregating search results into clusters. In some embodiments, semantic
query expansion is performed based on the search query submitted by the end user. This
may include generating a number of expanded queries based on the original search query
and performing a search using each expanded query. Each search would result in a set of
search results, and each set of search results may be considered a cluster.
[0056] A search results page is generated that comprises a user interface that
allows the end user to navigate the different clusters and view search results within each
cluster. The search results page is provided to the end user, as shown at block 506.
[0057] A user selection of a cluster within the search results page is detected, as
shown at block 508. In response to detecting selection of the cluster, a user context is
identified based on the cluster at block 510. In some embodiments, the user context
corresponds with the subset of search results within the selected cluster. In other
embodiments, each cluster corresponds to a concept to which each search result within the
cluster pertains. In such embodiments, the user context may correspond with the concept
of the selected cluster. In further embodiments, the selected cluster may have been
generated using an expanded query. In such embodiments, the user context may
correspond with the expanded query. Any and all such variations are contemplated to be
within the scope of embodiments of the present invention.
[0058] One or more advertisements are selected based on the current user context,
as shown at block 512. In some embodiments, the advertisements are selected based only
on the current user context. In other embodiments, the advertisement selection may also
take into account previous user contexts within the search results page. For instance, the
end user may have previously selected to view other search result clusters within the
search results page. Each time the end user selected a cluster, a user context was generated
corresponding with the search results of the selected cluster. In embodiments, weighting
may be applied to each user context. The weighting may be based on an order of the user
contexts, user view time of the user contexts, and/or a variety of other factors. The one or
more advertisements are provided for presentation on the search results page at block 514.
[0059] By way of illustration, FIGS. 6A and 6B include exemplary screen displays
showing presentation of advertisements selected based on a current search result cluster in
accordance with embodiments of the present invention. It will be understood and
appreciated by those of ordinary skill in the art that the screen displays of FIGS. 6A and
6B are provided by way of example only and are not intended to limit the scope of the
present invention in any way.
[0060] Referring initially to FIG. 6A, an exemplary screen display is provided that
shows a search results page 600. The search results page 600 includes search results
returned in response to the search query "taylor swift" 602. As shown in FIG. 6A, the
search results page 600 includes a number of tabs 604, 606, 608, 610. Each tab
corresponds with a cluster of search results and includes a title identifying the cluster.
[0061] The end user may view search results corresponding with a cluster by
selecting one of the tabs 604, 606, 608, 610. In FIG. 6A, the end user has selected the
"Country Music" tab 604 such that search results within that cluster are presented on the
search results page 600 (search results have been omitted from FIG. 6A for clarity
purposes). As such, the user context corresponding with the search results page 600
includes the concept "Country Music" and/or the search results within the "Country
Music" cluster. An advertisement 612 has been selected based on the concept "Country
Music" and/or the search results within the "Country Music" cluster and is presented on
the search results page 600.
[0062] FIG. 6B illustrates the search results page 600 after the user has selected
the "Songs and Lyrics" tab 606. Search results corresponding with the "Country Music"
cluster are no long displayed. Instead, search results corresponding with the "Songs and
Lyrics" cluster are now presented within the search results page (search results have been
omitted from FIG. 6B for clarity purposes). As such, a new user context is presented
corresponding with the "Songs and Lyrics" concept and/or the search results within the
"Songs and Lyrics" cluster. A new advertisement 614 has been selected based on the
"Songs and Lyrics" cluster and replaces the previous advertisement 612 in the search
results page.
[0063] Some embodiments of the present invention select advertisements based on
user context by identifying dominant concepts within the user context of the search results
page using a concept graph. Referring to FIG. 7, a flow diagram is provided that illustrates
a method 700 for identifying dominant concepts in the user context of a search results
page and selecting advertisements based on the identified dominant concepts in
accordance with an embodiment of the present invention. As shown at block 702, search
results within a current user context (and/or previous user contexts) are identified.
Concepts within each search result (i.e., the user context) are identified at block 704. Each
concept is identified by analyzing text corresponding with the search results to identify the
concepts. In various embodiments of the invention, the concepts may be identified from
text of search result snippets, documents corresponding with the search results, or other
locations.
[0064] The concepts are analyzed in conjunction with a concept graph at block 706
to determine dominant concepts. As discussed previously, a concept graph comprises a
collection of concepts and relationships among the concepts. Concepts are represented as
nodes within the concept graph and relationships among the concepts are represented as
edges between the nodes. The edges also represent the strength of relationships between
the concepts.
[0065] In various embodiments, dominant concepts may be determined based on
the frequency with which each concept appears in the text of the search results of the user
context and the strength of relationship of concepts within the concept graph. In some
embodiments, the strength of relationships of identified concepts to one another is
analyzed. In other embodiments, a main concept or topic is identified for the user context
(e.g., based on the query context and/or a concept corresponding with a selected cluster)
and each concept's strength to the main concept or topic is analyzed.
[0066] Any number of dominant concepts may be identified at block 706. In some
embodiments, the top N concepts are determined to be dominant concepts. In other
embodiments, any concept satisfying a predetermined threshold is determined to be a
dominant concept. Any and all such variations are contemplated to be within the scope of
embodiments of the present invention.
[0067] After identifying one or more dominant concepts, an expanded graph of
concepts is generated from the concept graph for each dominant concept, as shown at
block 708. In particular, the location of each dominant concept is identified in the concept
graph and related concepts that have a strong relationship to the dominant concepts are
identified from the concept graph. Based on the expanded graph(s), keywords mapped to
the expanded graph(s) are identified and ranked, as shown at block 710. One or more of
the ranked keywords are selected for use in advertisement selection at block 712.
[0068] As discussed previously, in some embodiments of the present invention,
advertisements may be selected based on the absence of user activity as opposed to being
based on the identification of user activity within a search results page. Turning now to
FIG. 8, a flow diagram is provided that illustrates a method 800 for selecting an
advertisement based on user inactivity within a search results page in accordance with an
embodiment of the present invention. Initially, as shown at block 802, a search query is
received from an end user. A search results page is provided with one or more
advertisements in response to the search query at block 804. It is determined at block 806
that a predetermined period of time has expired in which no user action within the search
results page has been detected. In some embodiments, some user activities, such as
scrolling up and down a window may not be considered as a user action within the search
results page. However, other user activities such as selecting a search result or an
advertisement would be considered a user action. The user activities to be tracked to
determine whether any user action has occurred for the purposes of the method 800 may
be configurable by the search system provider.
[0069] The predetermined period of time is also configurable by the search engine
provider. In embodiments, the search engine provider may arbitrarily set the
predetermined period of time or may select a period of time that satisfies the search engine
provider's interests. For instance, the search engine provider may determine a period of
time that the search engine provider determines is sufficient for the end user to view the
currently displayed advertisements and exhibit no interest in the advertisements based on
the absence of user action. In some cases, the search engine provider may collect
historical information regarding end user interactions with search results pages, including
the time end users spend viewing search results and advertisements, and may use such
historical information to select the predetermined time period.
[0070] A new set of one or more advertisements is selected at block 808 based on
the determination that no user activity has been detected within the predetermined period
of time. New advertisements may be selected based on user inactivity using a number of
different approaches in various embodiments of the present invention. By way of example
only and not limitation, in some embodiments, the new advertisements may be selected
from a ranked list of advertisements from which the currently displayed set of
advertisements was selected. For instance, only the highest N advertisements may have
been previously selected for presentation. Accordingly, new advertisements may be
identified by going down the list of ranked advertisements to select the next highest
ranked advertisements that were not previously selected.
[0071] In another embodiment, new advertisements may be selected by using
different keywords for advertisement selection than were used to select the currently
presented advertisements. For instance, the process of selecting the currently presented
advertisements may have included determining a list of ranked keywords, from which only
the top N keywords were used for selecting the currently presented advertisements. In
some embodiments, different keywords from the original list of ranked keywords may be
employed for selecting new advertisements. In other embodiments, instead of using the
original list of keywords, a new set of keywords may be identified. For instance, different
algorithms for keyword selection or a different context may be used for the keyword
selection. As an example of the latter, the currently displayed advertisements may have
been selected using the current user context within the search results page (i.e., what the
end user is currently viewing within the search results page). New advertisements may be
selected by also taking into account historical user actions within the search results page.
[0072] After selecting a set of one or more new advertisements, at least one
advertisement currently displayed on the search results page is replaced with a new
advertisement, as shown at block 810. In some embodiments, only a single advertisement
is displayed on the search results page and that advertisement is replaced with a new
advertisement. In other embodiments, the search results page includes multiple
advertisements, and the process may replace all or only a portion of the currently
displayed advertisements with new advertisements.
[0073] Turning to FIG. 9, a flow diagram is provided that illustrates a method 900
for replacing advertisements on a search results page based on user activity and/or user
inactivity in accordance with an embodiment of the present invention. As shown at block
902, a search query is received. A query context is determined for the search query at
block 904. The query context may include the search terms included in the search query
and/or other information, such as, for instance, a time the search query is submitted or the
location of the end user submitting the search query.
[0074] An initial set of one or more advertisements is selected based on the query
context, as shown at block 906. Additionally, a search results page is generated that
includes the initial set of advertisements, and the search results page is provided to the end
user, as shown at block 908.
[0075] The process continues by tracking the search results page for user action, as
shown at block 910. It is determined at block 912 whether any user action has been
detected. If no user action is detected, it is determined whether a predetermined time
period has expired, as shown at block 914. If the predetermined time period has not
expired, the process continues to track the search results page for user action. If the
predetermined time period has expired, a new set of one or more advertisements is
selected at block 916 based on the absence of user activity within the predetermined
period. The new set of advertisements may be selected using any of a number of different
approaches, such as those described hereinabove. At least one currently displayed
advertisement is replaced with a new advertisement, as shown at block 918.
[0076] Alternatively, if user activity is detected at block 912, the process identifies
a user context based on the user action, as shown at block 920. A new set of one or more
advertisements is selected based on the user context at block 922. The new set of
advertisements may be selected using any of a number of different approaches, such as
those described hereinabove. At least one currently displayed advertisement is replaced
with a new advertisement, as shown at block 924.
[0077] As can be understood, embodiments of the present invention are directed to
selecting advertisements for presentation within a search results page based on user
activity or user inactivity within a search results page. The present invention has been
described in relation to particular embodiments, which are intended in all respects to be
illustrative rather than restrictive. Alternative embodiments will become apparent to those
of ordinary skill in the art to which the present invention pertains without departing from
its scope.
[0078] From the foregoing, it will be seen that this invention is one well adapted to
attain all the ends and objects set forth above, together with other advantages which are
obvious and inherent to the system and method. It will be understood that certain features
and subcombinations are of utility and may be employed without reference to other
features and subcombinations. This is contemplated by and is within the scope of the
claims.
CLAIMS
What is claimed is:
1. One or more computer-readable media storing computer-useable
instructions that, when used by one or more computing devices, cause the one or more
computing devices to perform a method comprising:
receiving a search query submitted by an end user;
providing a plurality of search results in response to the search
query submitted by the end user, wherein the plurality of search results are grouped
into a plurality of clusters, each cluster including a different subset of search
results;
identifying a user activity corresponding with one of the plurality of
clusters, the user activity being indicative of user interest in a selected cluster;
identifying a user context based on the selected cluster; and
selecting one or more advertisements for presentation based on the
user context; and
providing the one or more advertisements for presentation to the
end user.
2. The one or more computer-readable media of claim 1, wherein the
method further includes selecting one or more initial advertisements based on the search
query, wherein the one or more initial advertisements are provided in conjunction with the
plurality of search results, and wherein the one or more advertisements are provided by
replacing the one or more initial advertisements with the one or more advertisements.
3. The one or more computer-readable media of claim 1, wherein each
subset of search results corresponding with each cluster is provided within a separate tab
selectable by the end user, and wherein identifying a user activity corresponding with one
of the plurality of clusters comprises identifying a user selection of a tab corresponding
with the selected cluster.
4. The one or more computer-readable media of claim 1, wherein
selecting the one or more advertisements comprises selecting one or more keywords based
on the user context and selecting the one or more advertisements based on the one or more
keywords, and wherein selecting the one or more keywords comprises:
identifying a plurality of search results within the user context;
identifying a plurality of concepts within text corresponding with
the plurality of search results;
analyzing the plurality of concepts using a concept graph to identify
one or more dominant concepts; and
selecting the one or more keywords based on the one or more
dominant concepts.
5. The one or more computer-readable media of claim 1, wherein
providing the plurality of search results grouped into the plurality of clusters comprises
performing a semantic query expansion of the search query submitted by the end user to
generate a plurality of expanded queries and performing a search using each expanded
query, and wherein the user context is identified based on an expanded query
corresponding with the selected cluster.
6. The one or more computer-readable media of claim 1, wherein the
one or more advertisements are selected based on at least one previous user context in
addition to the user context, and wherein selecting the one or more advertisements
comprises weighting each of the user context and the at least one previous user context
based on at least one selected from the following: an order in which each user context
occurred, and a length of time the end user viewed each user context.
7. One or more computer-readable media storing computer-useable
instructions that, when used by one or more computing devices, cause the one or more
computing devices to perform a method comprising:
receiving a search query submitted by an end user;
providing a search results page for presentation to the end user in
response to the search query submitted by the end user, the search results page
including a plurality of search results and an initial advertisement selected based at
least in part on the search query submitted by the end user;
tracking the search results page for user activity;
detecting a user action by the end user within the search results
page;
responsive to detecting a user action within the search results page,
identifying a user context based on the user action;
identifying a set of search results within the user context;
identifying a plurality of concepts within text corresponding with
the set of search results within the user context;
analyzing the plurality of concepts using a concept graph to identify
one or more dominant concepts;
selecting one or more keywords based on the one more dominant
concepts;
selecting a new advertisement based on the one or more keywords;
and
replacing the initial advertisement within the search results page
with the new advertisement.
8. The one or more computer-readable media of claim 7, wherein the
text corresponding with the set of search results comprises at least one selected from the
following: a search result snippet for a search result; and a document to which a search
result references.
9. The one or more computer-readable media of claim 7, wherein
analyzing the plurality of concepts using a concept graph to identify one or more dominant
concepts comprises identifying a main topic of the user context and identifying the one or
more dominant concepts based on strength of relationship in the concept graph between
each of the one or more dominant concepts and the main topic of the user context, and
wherein the method further comprises generating an expanded graph for at least one
dominant concept and ranking a plurality of keywords based on the expanded graph,
wherein the one or more keywords are selected based on the rankings for the plurality of
keywords.
10. The one or more computer-readable media of claim 7, wherein the
method further comprises:
continuing to track the search results page for user activity after
replacing the initial advertisement within the search results page with the new
advertisement;
detecting a new user action by the end user within the search results
page;
responsive to detecting a new user action within the search results
page, identifying a new user context based on the new user action;
selecting a second new advertisement based on the new user
context; and
replacing the new advertisement within the search results page with
the second new advertisement.
11. One or more computer-readable media storing computer-useable
instructions that, when used by one or more computing devices, cause the one or more
computing devices to perform a method comprising:
receiving a search query submitted by an end user;
providing a search results page in response to the search query, the
search results page including a plurality of search results and an initial
advertisement;
determining that a predetermined time period has expired without
detecting user activity within the search results page;
responsive to determining that the predetermined time period has
expired without detecting user activity within the search results page, selecting a
new advertisement; and
replacing the initial advertisement within the search results page
with the new advertisement.
12. The one or more computer-readable media of claim 11, wherein the
initial advertisement was selected from a list of ranked advertisements, and wherein the
new advertisement is selected from the list of ranked advertisements.
13. The one or more computer-readable media of claim 11, wherein the
initial advertisement is selected based on a current user context within the search results
page, wherein the new advertisement is selected based on a history of user contexts within
the search results page, and wherein each user context in the history of user contexts is
weighted based on at least one selected from the following: an order in which each user
selection was selected, and a length of time the end user viewed each subset of search
results corresponding with each user selection.
14. The one or more computer-readable media of claim 11, wherein the
new advertisement is selected using a first algorithm that differs from a second algorithm
used to select the initial advertisement.
15. The one or more computer-readable media of claim 11, wherein the
initial advertisement is selected using a first subset of one or more keywords selected for a
list of ranked keywords and the new advertisement is selected using a second subset of one
or more keywords selected from the list of ranked keywords.