US20160291840A1 - Method and Apparatus for Automated Time-Adaptive Rearrangement of Visual Elements in Mobile Computing Device - Google Patents
Method and Apparatus for Automated Time-Adaptive Rearrangement of Visual Elements in Mobile Computing Device Download PDFInfo
- Publication number
- US20160291840A1 US20160291840A1 US15/180,022 US201615180022A US2016291840A1 US 20160291840 A1 US20160291840 A1 US 20160291840A1 US 201615180022 A US201615180022 A US 201615180022A US 2016291840 A1 US2016291840 A1 US 2016291840A1
- Authority
- US
- United States
- Prior art keywords
- day
- icons
- time
- user device
- computing device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
- G06F3/04817—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance using icons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2477—Temporal data queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/316—Indexing structures
- G06F16/319—Inverted lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G06F17/30551—
-
- G06F17/30598—
-
- G06F17/30622—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
- G06F3/0482—Interaction with lists of selectable items, e.g. menus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
- G06F3/04842—Selection of displayed objects or displayed text elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
- G06F3/0488—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
Abstract
A mobile computing device includes a touchscreen display, a non-transitory computer-readable memory that stores a plurality of installed native applications, and one or more processors. A method performed on the mobile device includes displaying, on the touchscreen display, a screen that includes icons allowing the installed native applications to be launched. The method includes determining that a first set of native applications are launched at the mobile computing device at or near a first time of day. The method includes determining whether a current time of day matches the first time of day. The method includes, upon determining that the current time of day matches the first time of day, rearranging the icons on the screen into a first arrangement of icons, wherein the first arrangement of icons displays the icons for the first set of native applications above a line.
Description
- The present application is a continuation of U.S. patent application Ser. No. 14/446,991 filed on Jul. 30, 2014, which application claims the benefit of U.S. Provisional Application No. 61/921,245, filed on Dec. 27, 2013. The entire disclosures of the applications referenced above are incorporated by reference.
- This disclosure relates to methods and apparatuses for determining an active persona of a user device.
- User devices, such as smartphones, tablet PCs, cellular telephones, and portable digital assistants have become widespread and essential in consumers' everyday lives. Moreover, users personalize their user devices by downloading applications to their device or accessing applications via a web-browser. Each user device can have multiple users, and each user can use the device for a number of different reasons. For example, a user can use a user device for playing games, watching movies, listening to music, checking financial information, editing documents, and messaging friends.
- One aspect of the disclosure provides a method for determining an active persona of a user device. The active persona is indicative of a predicted objective of the user of the user device. The method includes determining a first user device state of a user device. The first state defines an action currently being performed on the user device and a set of observations corresponding to the user device. The method further includes the first user device state to a model and determining the active persona of the user device based on the output of the model. The model is configured to receive a user device state and outputs probability values associated with one or more clusters, each cluster being indicative of one or more items of interest corresponding to the user device. The method further includes performing a procedure using the active persona of the user device. The method can be performed by a processing device.
- Implementations of the disclosure may include one or more of the following features. In some implementations, the method includes capturing a second user device state of the user device and updating the model based on the first user device state and the second user device state. The second user device state is subsequent to the first user device state and defines a subsequent action being performed on the user device and one or more subsequent observations. In some examples, the model is a Hidden Markov Model.
- The first user device state may be based on an active process of the user device. In some implementations, the active process is one of an active application being executed by the user device, and active media content being played by the user device, or an active search query being provided by the user device. Determining the first user device state may include using an inverted index to determine an active cluster to which the active process corresponds. The cluster is indicative of an active item of interest corresponding to the user device.
- In some examples, the method further includes determining a plurality of items of interest of the user device, providing the plurality of items of interest to a persona server, and receiving
- Determining the first user device state may include obtaining output values of one or more peripheral devices of the user device. Performing a procedure using the active persona may include providing the active persona to a search engine as part of a search query, providing the active persona to an application server in communication with the user device, or arranging a home screen of the user device based on the active persona.
- The active persona may be a vector indicating a plurality of probabilities output by the model. Each element of the vector may device will transition from a current item of interest to associated with the probability value provide a probability of transitioning from the first state to a second state. In these implementations, each element of the vector corresponds to a different state. Obtaining the active persona may include obtaining a vector indicating a plurality of probabilities from the model and mapping the vector to a category based on one or more rules.
- Another aspect of the disclosure provides a user device that determines its active persona. The user device includes a storage device including one or more computer readable mediums that store a model that receives a user device state and outputs probability values associated with one or more clusters. Each cluster is indicative of one or more items of interest corresponding to the user device. The user device further includes a processing device executing computer readable instructions. When executed by the processing device, the computer-readable instructions cause the user device to determine a first user device state of a user device, feed the first user device state to the model, obtain an active persona of the user device based on the output of the model, and perform a procedure using the active persona. The first user device state defines an action currently being performed on the user device and a set of observations corresponding to the user device. The active persona is indicative of a predicted objective of the user of the user device.
- According to some implementations of the present disclosure, the computer readable instructions further cause the processing device to capture a second user device state after determining the active persona of the user device and to update the model based on the first user device state and the second user device state. The second user device state is subsequent to the first user device state and defines a subsequent action being performed on the user device and one or more subsequent observations. In some examples, the model is a Hidden Markov Model.
- According to some implementations, the action being performed by the user device is based on an active process of the user device. The active process is one of an active application being executed by the user device, an active media content being played by the user device, or an active search query being provided by the user device.
- In some implementations, determining the first user device state includes determining an active cluster to which the active process corresponds using an inverted index. The active cluster is indicative of an active item of interest corresponding to the user device. In some of these implementations, the computer readable instructions further cause the processing device to determine a plurality of items of interest of the user device, provide the plurality of items of interest to a persona server, and to
- Determining the first user device state may include obtaining output values of one or more peripheral devices of the user device. Performing a procedure using the active persona may include providing the active persona to a search engine as part of a search query, providing the active persona to an application server in communication with the user device, or arranging a home screen of the user device based on the active persona.
- The active persona may be a vector indicating a plurality of probabilities output by the model. Each element of the vector may indicate a probability that the user device will transition from a current item of interest to a cluster associated with the probability value that contains a second item of interest. In these implementations, each element of the vector corresponds to a different state. Obtaining the active persona may include obtaining a vector indicating a plurality of probabilities from the model and mapping the vector to a category based on one or more rules.
- The details of one or more implementations of the disclosure are set forth in the accompanying drawings and the description below. Other aspects, features, and advantages will be apparent from the description and drawings, and from the claims.
-
FIG. 1 is a drawing illustrating an example system of devices for determining and using an active persona of a user device. -
FIG. 2 is a schematic illustrating example components of a persona server. -
FIG. 3 is a schematic illustrating example components of a user device. -
FIG. 4 is a drawing illustrating an example of a Hidden Markov Model. -
FIG. 5 is a flow chart illustrating an example set of operations for a method for determining an active persona of a user device. - Like reference symbols in the various drawings indicate like elements.
-
FIGS. 1-3 illustrate anexample system 10 for predicting anactive persona 100 of auser device 300 and the components thereof. Thesystem 10 can include a persona server 200 (FIG. 2 ) and one or more user devices 300 (FIG. 3 ) in communication with a network 110. In some implementations, thepersona server 200 is configured to provide data that allows auser device 300 to determine itsactive persona 100. Anactive persona 100 of auser device 300 can be indicative of an objective of a user of thedevice 300 at a time when the user is using thedevice 300. Auser device 300 can have one or more users and each user can have one or more objectives when using thedevice 300. For example, a family having a mother, a father, and a child may all use the same tablet computer 300 a. The father may use the tablet computer 300 a for reading the news in the mornings (e.g., “news objective”), watching sports in the evenings (e.g., “sports fan objective”), and playing violent video games at night (e.g., “mature video game objective”). The mother may use the tablet computer 300 a for scheduling tasks in the mornings (e.g., “planning objective”) and checking stocks in the evening (e.g., “finance objective”). Finally, the child may use the tablet computer 300 a for watching child-appropriate movies in the afternoons (e.g., “child movie watcher objective”) and playing educational games in the evening (e.g., “educational game objective”). The explicit objectives provided above are provided for example only. Further, the foregoing examples are trends and not necessarily always adhered to by the users. Theuser device 300 utilizes the data and a user device state of theuser device 300 to predict theactive persona 100 of theuser device 300. The user device state can include an action currently being performed on the user device 300 (e.g., an active application currently being used on theuser device 300, anactive media content 326 being played by theuser device 300, or anactive search query 324 being provided by the user device 300) and a set of observations corresponding to the user device 300 (e.g., a location of theuser device 300, a speed of theuser device 300, a day of the week, a time of the day). Theuser device 300 can use the active persona to perform a procedure, such as communicating theactive persona 100 to a service provider or rearranging the user home screen based on theactive persona 100. - By knowing (or likely knowing) the
active persona 100 of thedevice 300, service providers, such assearch engines 120,application providers 130, and advertisers, can provide more relevant content to the user.Search engines 120 can utilize anactive persona 100 of a user to increase the relevancy of search results. For example, if theuser device 300 provides asearch query 324 of “schedules,” asearch engine 120 can provide better results to theuser device 300 if it knows theactive persona 100 of theuser device 300. In particular, thesearch engine 120 cannot decipher what the user means when entering the term “schedules.” By having a prediction as to theactive persona 100 of the user, thesearch engine 120 can better understand the user's intentions when he or she enters the term “schedules.” Drawing from the example above, thesearch engine 120 can return a list of applications that provide schedules of the major sports leagues (e.g., NFL, MLB, NHL, and NBA) when theactive persona 100 corresponds to that of a sports fan. Similarly, when theactive persona 100 of thedevice 300 corresponds to that of a “planner,” then thesearch engine 120 can return a list of applications that are directed to scheduling tasks and appointments. Theactive persona 100 of thedevice 300 can also help services determine which advertisements to transmit to thedevice 300. For example, if theactive persona 100 is “violent video game persona,” a video streaming service can embed an advertisement for a new first person shooter game. Similarly, if theactive persona 100 is “children's games persona,” anapplication server 130 can transmit an advertisement for a new blockbuster cartoon movie. The foregoing examples are not intended to limit the uses ofactive personas 100 of adevice 300. - In some implementations, the
persona server 200 and auser device 300 work in combination to determine anactive persona 100 of thedevice 300. Thepersona server 200 can identifyclusters 238 that are used to identify potentialactive personas 100 of adevice 300. Acluster 238 can refer to a group of data records that have one or more similar attributes. In some implementations, eachcluster 238 contains the same type of data records. Examples of data records can include, but are not limited to, application data records, query data records, and media data records. - In some implementations, the
persona server 200 identifiesclusters 238 ofsimilar applications 322, search queries 324, ormedia contents 326.Clusters 238 are clustered based on the attributes of the data records being clustered. For instance, thepersona server 200 can cluster the application records according to the application publisher, application type, application subtype, the ratings of the applications, and the popularity of the applications. Similarly, thepersona server 200 can cluster the query records according to the actual query, the location of the query, and the action taken in response to the query results. - When the
persona server 200 identifies acluster 238, thepersona server 200 can also assign a cluster identifier (“cluster ID”) to the identifiedcluster 238. Each cluster ID corresponds to asingle cluster 238 and eachcluster 238 has only one cluster ID. As more data records are added to acluster 238, the make-up of thecluster 238 changes, but the cluster ID of thecluster 238 should remain the same. - In some implementations, the
persona server 200 is further configured toindexes 240 corresponding to theclusters 238. In some implementations, theindexes 240 are inverted indexes that associate a data record to the cluster IDs of theclusters 238 to which the data record belongs. In some implementations, an inverted index of a particular item may be represented in the following form: <ITEM_ID->[cluster_ID1, cluster_ID2, . . . cluster_ID_n], where there are n clusters that are associated with the particular item. For instance, if a certain application (e.g., the ANGRY BIRDS® application by Rovio Games) belongs to two different clusters 238 (“Published By Rovio” and “Popular Games”), the twodifferent clusters 238 can be indexed to the data record of certain applications. In some implementations, each cluster in the inverted index may have a weight associated therewith. For example, an example inverted index of a particular item may be represented in the following form: <ITEM_ID->[(cluster_ID1, weight1), cluster_ID2, weight2), . . . cluster_ID_n, weight_n)], where there are n clusters that are associated with the item. Each weight may be a value indicating a relative importance of the cluster to a given item (e.g., relative to all other possible item-cluster tuples). - In some implementations, the
persona server 200 provides some or all of theindexes 240 to theuser devices 300. In some of these implementations, thepersona server 200 selects theindexes 240 that are provided to theuser devices 300 based on items of interest of eachuser device 300. Examples of items of interest can include applications of theuser device 300, queries/logs of theuser device 300, and histories ofmedia contents 326 played on theuser device 300. Theuser device 300 provides the list of the items of interest (or data that is indicative of the items of interest) to thepersona server 200 and thepersona server 200 transmits theindexes 240 relevant to the items of interest. - The
user device 300 can utilize theindexes 240, the user device state of theuser device 300, and one or more machine learnedmodels 400 to predict anactive persona 100 of the user. In some implementations, eachuser device 300 monitors itself and creates/updates models 400 that predict theactive persona 100 of thedevice 300 based on the actions being performed by the user device and the observations of theuser device 300 at a given time. In some implementations, the machine learnedmodel 400 is a Hidden Markov Model (“HMM”). Other types of machine learnedmodels 400 may be implemented as well. For example, theuser device 300 can create/update a Bayesian model to predict the probabilities of each state. - A machine learned
model 400 receives a user device state as input (e.g., an action being performed on theuser device 300, a current time, a day of the week, accelerometer readings, current active application, mostrecent search query 324 sent by the user device 300) and outputs one or more predictedclusters 238. In some implementations, each predictedcluster 238 includes a score (e.g., a probability) indicating a likelihood that theuser device 300 is in the predictedcluster 238. Thus, in some implementations, anactive persona 100 can be represented by a vector of probabilities. For example, an active persona may be defined by the following vector, AP(Cj)=[S0C0, S1C1, S2C2, . . . , SnCn], where AP is theactive persona 100 and Si is the probability that the user device will transition from an item of interest in the jth cluster, Cj, to an item of interest contained in the ith cluster, Cj. Additionally or alternatively, theactive persona 100 can be represented by a category to which the vector of probabilities maps or matches. - When the
user device 300 determines anactive persona 100, theuser device 300 can communicate theactive persona 100 to thepersona server 200 or another server (e.g., asearch engine 120 or an application server 130). Furthermore, theuser device 300 can take actions based on the predictedactive persona 100. For instance, if theuser device 300 predicts that theactive persona 100 is likely game related (e.g., the most likely transition or transitions is/are to a cluster corresponding to a game), theuser device 300 can reconfigure its home screen of theuser device 300 to display icons and widgets of gaming applications. Similarly, if theuser device 300 predicts that theactive persona 100 is likely business related, theuser device 300 can reconfigure its home screen to display icons and widgets of business related applications. - While the
persona server 200 may be configured to determine theactive persona 100, theuser device 300 can be used to determine theactive persona 100 so that theuser device 300 maintains themodel 400. In this way, the user's information such as locations and habits do not need to be exported from theuser device 300, thereby maintaining the user's privacy. Furthermore, when transmitting theactive persona 100 of the user to a remote device (e.g., a third party server), theuser device 300 can conceal the identity of the user and/or theuser device 300 before transmitting theactive persona 100. -
FIG. 2 illustrates an example of apersona server 200. In some implementations, thepersona server 200 is a system of one or more computing devices. Thepersona server 200 may be further integrated into a larger computing system, such as asearch engine 120, anapplication server 130, or an ad server. In an example implementation, thepersona server 200 includes aprocessing device 210, anetwork interface 220, and astorage device 230. Theprocessing device 210 can execute aclustering module 212, anindexer 214, and anindex selector 216, all of which can be implemented as computer-readable instructions. - The
processing device 210 includes one or more physical processors that execute computer-readable instructions (e.g., executable computer code) in communication with a memory device (e.g. RAM and/or ROM) that stores the computer-readable instructions. In implementations where theprocessing device 210 includes two or more processors, the two or more processors can operate in an individual or distributed manner. In some implementations, some of the executed components are implemented at different physical devices or locations. For instance, theclustering module 212 may execute at a first location/device and theindexer 214 and theindex selector 216 may execute at a second location/device. - The
network interface device 220 includes one or more devices that can perform wired or wireless (e.g., Wi-Fi or cellular) communication. Examples of thenetwork interface device 220 include, but are not limited to, a transceiver configured to perform communications using the IEEE 802.11 wireless standard, an Ethernet port, a wireless transmitter, and a universal serial bus (USB) port. - The
storage device 230 can include one or more computer readable storage mediums (e.g., hard disk drives and/or flash memory drives). The storage mediums can be located at the same physical location or device or at different physical locations or devices. Thestorage device 230 can store application records 232, query records 236,media content records 234,clusters 238, andindexes 240. - Application data records 232 correspond to known (e.g., previously discovered) applications. An application can refer to computer readable instructions that cause a computing device to perform a task. In some examples, an application may be referred to as an “app.” Example applications include, but are not limited to, word processing applications, spreadsheet applications, messaging applications, media streaming applications, social networking applications, and games.
- Each application data record 232 can include application metadata that defines the attributes of the application. Examples of attributes of an application can include, but are not limited to, a name of the application, an operating system that the application is programmed for, a number of downloads of the application, a popularity of the application, a type of the application, a subtype of the application, a rating of the application, a publisher of the application, an expected usage amount of the application, and a number of uninstalls of the application.
-
Query data records 236 correspond to previously received search queries 324. The search queries 324 can be received by asearch engine 120 associated with thepersona server 200 or from a third-party source. Eachquery data record 236 can include query metadata that defines attributes of asearch query 324. Examples of attributes of asearch query 324 can include, but are not limited to, the query itself, a geographic location of theuser device 300 providing the query, and an action taken in response to the query (e.g., what the user selected). -
Media content records 234 correspond to known media contents (e.g., movies, songs, albums, TV shows). Eachmedia content record 234 can include media metadata that defines the attributes of the media content. Examples of attributes of a media content can include, but are not limited to, a name of the media content, a type of the media content, a genre of the media content, a format of the media content, an author of the media content, a publisher of the media content, a popularity of the media content, a rating of the media content, and a length of the media content. - The
clustering module 212 clusters data records. Theclustering module 212 receives as input a set of data records to be clustered and may further receive a list of one or more attributes by which to cluster the data records. The output of theclustering module 212 is one ormore clusters 238. The outputtedclusters 238 may identify the members (i.e., data records) of thecluster 238. Theclustering module 212 can execute any clustering algorithm, such as k-nearest neighbors, k-means clustering, bi-clustering, or text-based clustering. Theclustering module 212 can cluster each set of the application records 232, the query records 236, and themedia records 234 individually. That is, theclustering module 212 clusters the application records 232 with other application records 232, query records 236 withother query records 236, andmedia records 234 withother media records 234. The attributes that theclustering module 212 uses to cluster the data records can be manually selected by a human or can be learned using a statistical process, such as principal component analysis (PCA). - The
clustering module 212 can cluster the data records in any suitable manner. In some implementations, theclustering module 212 can initially cluster data records into general “category clusters.” During this type of clustering, theclustering module 212 can produce several high-level clusters 238 based on all available information. The information by which theclustering module 212 clusters the data records can include metadata contained in thedata records 232, 236, 238 (e.g., the attributes collected from 3rd party sources or manually added to the records). In some implementations, this type of clustering can be done in a supervised manner, whereby a user manually selects the attributes, or representative data records by which the data records are clustered. Furthermore, theclustering module 212 can cluster eachcluster 238 intosub-clusters 238. When theclustering module 212 completes clustering thegeneral clusters 238, theclustering module 212 can continue to cluster eachcluster 238 to identify as many levels as desired. Furthermore, theclusters 238 do not need to make “human sense.” Put another way, the relationships between data records in a sub-cluster 238 do not need to be readily apparent to a human observer. - In some implementations, the
clustering module 216 can identifypossible clusters 238 by leveraging user activity. For example, theclustering module 212 can focus on items of interest to a given geo-region. Such clustering can be done by levels, whereby theclustering module 212 begins clustering based on country, then based on state/province, then urban area, then a particular postal code (if useful). In addition to geo-region, theclustering module 212 can identifyclusters 238 of data records that are common to a group of users based on general collaborative filtering. In these scenarios, theclusters 238 can be identified using anonymized data (i.e., data which cannot be attributed to any particular user). - In some implementations, the
clustering module 212 can identify applicationspecific clusters 238. In these implementations, theclustering module 212 can group application data records 232 with a common publisher, developer, or franchise (e.g., all of the applications in the ANGRY BIRDS® series). Further, theclustering module 212 can group application data records 232 that are related to a common franchise (e.g., include applications with cheat codes for ANGRY BIRDS®, or ANGRY BIRDS® wallpaper themes). - In some implementations, the foregoing
clusters 238 can be subdivided based on other attributes. Theclusters 238 can be divided according to, for example, current popularity, long-term popularity, or price. Theclustering module 212 can produce many sets of overlappingclusters 238. For example,clusters 238 relating to “Rovio applications” may have a large amount of overlap with acluster 238 relating to the “ANGRY BIRDS®” franchise. - The
clustering module 212 further provides a cluster ID to each newly identifiedcluster 238. Theclustering module 212 can be configured to determine whether thecluster 238 is newly formed or merely updated with new data records. If thecluster 238 is newly formed, theclustering module 212 generates a new cluster ID. The cluster ID can be any unique identifier, such as a string of alphanumeric characters. If acluster 238 is preexisting and merely being updated, theclustering module 212 does not generate a new cluster ID. The cluster IDs may or may not have properties associated therewith. For example, a “publisher=Rovio” property or a “popular games” property may be associated with the cluster ID. Furthermore, theclustering module 212 can assign a version to eachcluster 238. In this way, when acluster 238 is updated, theclustering module 212 can increase the version number of thecluster 238. The version can be used to differentiate personas generated from nowoutdated clusters 238 ifclusters 238 change over time (e.g., currently popular games). Additionally or alternatively, theclustering module 212 can mergeclusters 238 with very high overlap (e.g., more than 80% or 85% overlap) in order to reduce the index size. - The
indexer 214 creates theindex 240 based on the identifiedclusters 238. In some implementations, theindex 240 is an inverted index. For each data record, theindexer 214 identifies theclusters 238 to which the data record belongs and creates anindex entry 242 based on the cluster IDs of the identifiedclusters 240 and an identifier of the data record (e.g., a record ID). For instance, if a data record “XYZ” is a member of “cluster_A” and of “cluster_B,” theinverted index entry 242 of data record XYZ is “XYZ: {cluster_A, cluster_B}.” - In some implementations, the
index selector 216 determines whichindex entries 242 to transmit to adevice 300. Theindex selector 216 can select theindex entries 242 to transmit in any suitable manner. In some implementations, theindex selector 216 can receive indicators of items of interest of auser device 300 from theuser device 300. An item of interest can include a list ofapplications 322 executed/accessed by the device (referred to “a list of applications”), list of search queries 324 provided by the user device 300 (referred to as “list of search queries), and/or a list ofmedia contents 326 played by the user device 300 (referred to as a “list of media contents 326). Theindex selector 216 can retrieve theindex entries 242 corresponding to each of the items of interest. Further, in some implementations, theindex selector 216 can cull theindex entries 242 based on a number of factors to reduce the number ofindex entries 242 that are transmitted to theuser device 300 in theindex 240. For example, theindex selector 216 keepsindex entries 242 that containclusters 238 which match to at least a minimum number ofuser devices 300. For example, if a particular application, search query, or media content is extremely rare, then theindex selector 216 does not include theindex entry 242 corresponding to the rare application, search query, or media content in theindex 240 that theindex selector 216 transmits to theuser device 300. - In some implementations, the
index selector 216 receives values from thedevice 300, whereby the values are indicative of two ormore applications 322, two or more search queries 324, and/or two ormore media contents 326. In these implementations, the actual contents of the transmitting device 300 (i.e., the exact applications installed on thedevice 300 orrecent search queries 324 transmitted from the device 300) are obscured from thepersona server 200. The transmittinguser device 300 can determine the values using, for example, a hash table or a bit mask. In such a scenario, theserver 200 can maintain a reverse lookup table that determinespossible applications 322, search queries 324, and/ormedia contents 326 that correspond to a received value. Theserver 200 can then retrieve theindex entries 242 of thepossible applications 322, search queries 324, and/ormedia contents 326. In some implementations, theindex selector 216 can cull theindex entries 242 that are transmitted to theuser device 300. -
FIG. 3 illustrates an example of auser device 300 configured to predict anactive persona 100. Theexample user device 300 can include a processing device 310 in communication with astorage device 320, anetwork interface 330, auser interface 340, and one or moreperipheral devices 350. Auser device 300 can include other components not shown without departing from the scope of the disclosure. - The processing device 310 can include one or more processors that execute computer-executable instructions and associated memory (e.g., RAM and/or ROM) that stores the computer-executable instructions. In implementations where the processing device 310 includes more than one processor, the processors can execute in a distributed or individual manner. The processing device 310 can execute an
index requestor 312, amodel builder 314, abehavior monitor 316, and apersona predictor 318, all of which can be implemented as computer-readable instructions. - The
storage device 320 can include one or more computer-readable mediums (e.g., hard disk drive or flash memory drive). Thestorage device 320 can be in communication with the processing device 310, such that the processing device 310 can retrieve any needed data therefrom. Thestorage device 320 stores theindex 240 that theuser device 300 receives from thepersona server 200 and one ormore models 400. Thestorage device 320 can further storeapplications 322, previous search queries 324, andmedia contents 326. - The
network interface 330 includes one or more devices that are configured to communicate with the network 110. Thenetwork interface 330 can include one or more transceivers for performing wired or wireless communication. Examples of thenetwork interface 330 can include, but are not limited to, a transceiver configured to perform communications using the IEEE 802.11 wireless standard, an Ethernet port, a wireless transmitter, and a universal serial bus (USB) port. - The
user interface 340 includes one or more devices that receive input from and/or provide output to a user. Theuser interface 340 can include, but is not limited to, a touchscreen, a display, a QWERTY keyboard, a numeric keypad, a touchpad, a microphone, and/or speakers. -
Peripheral devices 350 are hardware components that are included in theuser device 300 to perform specific functions. Theuser device 300 can have any number ofperipheral devices 350, including an accelerometer, a gyroscope, a global positioning system (GPS) module, a camera, a video camera, and/or a thermometer. - The index requestor 312 requests an
index 240 from thepersona server 200. In some implementations, the index requestor 312 scans theuser device 300 for items of interest. Items of interest can includeapplications 322 installed (native apps) or accessed (web apps) on theuser device 300, search queries 324 sent by thedevice 300 to asearch engine 120, andmedia contents 326 stored on or immediately accessible to theuser device 300. In some implementations, the index requestor 312 partially obscures each item of interest to protect the privacy of the user or users of theuser device 300. For instance, the index requestor 312 can apply a lossy hash function to an indicator of an item of interest. Additionally or alternatively, the index requestor 312 can apply a bit mask to one or more bits of the indicator. The index requestor 312 transmits the indicators of the items of interest to thepersona server 200. In some implementations, the index requestor 312 can include one or more timestamps with each indicator. For example, for an item of interest, a timestamp can indicate when the item of interest was loaded onto the device 300 (e.g., when anapplication 322 was installed or amedia content 326 was downloaded), when the item of interest was originated from the device 300 (e.g., when asearch query 324 was transmitted from the device 300), and/or when the item of interest was last used (e.g., the last time theapplication 322 was launched, themedia content 326 was played, or thesearch query 324 was sent to a search engine 120). In response, thepersona server 200 returns anindex 240 corresponding to the items of interest. Put another way, thepersona server 200 can return anindex 240 containingindex entries 242 that are keyed by the items of interest of theuser device 300. The index requestor 312 can receive theindex 240 and store the receivedindex 240 on thestorage device 320. - The
model builder 314 generates and updates one or more machine learnedmodels 400 used to predict theactive persona 100 of thedevice 300. In some implementations, the machine learnedmodel 400 is a Hidden Markov Model (HMM) or a Bayesian model. Themodel builder 314 initially creates themodel 400 based on the possible user device states of theuser device 300 and updates themodel 400 as theuser device 300 is used by one or more users. The user device states can correspond to thevarious clusters 238 that the user's actions may implicate. For example, theapplications 322 that theuser device 300 launches, themedia contents 326 that theuser device 300 plays, and the search queries 324 that theuser device 300 outputs can implicate various cluster IDs ofclusters 238 identified in theindex 240. The implicatedclusters 238 may be indicative of theactive persona 100 of theuser device 300. - The
model 400 predicts a transition from the current cluster implicated by the user device state of theuser device 300 based on the current user device state of thedevice 300. Themodel 400 can take into account current observations as well as how theuser device 300 has been used historically (i.e., the learned element of the model 400). For example, if the user of theuser device 300 typically opens an email application on weekday mornings and then opens a newspaper application immediately afterward, themodel 400 learns this behavior. In this way, when themodel 400 receives a set of observations indicating that it is a weekday morning and the user has opened up the email application, themodel 400 can predict that theuser device 300 is likely to launch anewspaper application 322 shortly thereafter (or another application in thesame cluster 238 as the newspaper application), or at least that it is more probable that the user will launch anewspaper application 322 as opposed to agaming application 322. Themodel 400 can output one or more probability values corresponding to different clusters that theuser device 300 may transition to (including the current cluster). In some implementations, the output of themodel 400 is structured in a vector, whereby each element of the vector represents a different cluster, and the value at each element is a probability that theuser device 300 will transition to that cluster. -
FIG. 4 illustrates an example of a machine learnedmodel 400. In the illustrated example, the machine learnedmodel 400 is an HMM. Themodel 400 can include n states 410, S={1, . . . , n} (e.g., clusters) and m observation sets 420, O=[o1, . . . , ot _ max] (e.g., user device states). In the illustrated example, there are threeclusters arrows 440 betweendifferent clusters 410 represent state transition probabilities, andlines 450 betweenclusters 410 and the observations 420 represent output probabilities. The state transition probability indicates a likelihood that theuser devices 300 changes from thecurrent cluster 410 to anext cluster 410. The output probabilities represent a likelihood that a given user device state 420 corresponds to a givencluster 410. Acurrent cluster 410 can refer to the current cluster of the user device 300 (e.g., a cluster of an application or media content currently being executed/played by the user device 300). A user device state 420 can include any number of features, including the last application launched by theuser device 300, the state of the accelerometer, the location of theuser device 300, the current weather conditions, the power state of thedevice 300, the conditions of the user device an hour ago, the conditions of the user device 30 minutes ago, or other suitable observations. Themodel 400 predicts the likelynext cluster 410 orclusters 410 of theuser device 300 given the current user device state 420 at a given time. For example, at t=0, the phone's screen is off, the previously predictedcluster 410 was “Violent Video Games,” and it is 9 AM on Saturday. Themodel 400 has learned that on Saturday mornings, thepersona 100 is likely to correlate to “watch children movies.” As soon as the phone screen is turned on, based on theprevious cluster 410, and the time of day and thepossible personas 100, themodel 400 predicts thenext cluster 410 likely correlates to “watch children movies.” This time, however, it is not the child, but rather the father who plays violent games, the first thing he does is launch a specific game in the “violent video game cluster.” In such a scenario, themodel 400 updates thecurrent cluster 410 and then predicts thenext cluster 410 corresponds to “violent video games.” - The
model builder 314 generates the machine learned model 400 (e.g., HMM or Bayesian model) using the cluster IDs identified in theindex entries 242 as thestates 410 of themodel 400. In particular, themodel builder 314 creates astate 410 for eachcluster 238 that is to be included in themodel 400. In some implementations, themodel builder 314 can analyze the receivedindex 240 to identify the set of allpossible clusters 238 for theuser device 300 and creates astate 410 for each identifiedcluster 238. In other implementations, themodel builder 314 identifies the set 420 of allpossible clusters 238 and selects a subset of theclusters 238, such that the selectedclusters 238 collectively include all or substantially all of the items of interest of theuser device 300. In these implementations, themodel builder 314 may be configured to minimize the amount of intersection of items of interest between the selectedclusters 238. Initially, themodel builder 314 has not trained themodel 400 with any data. Accordingly, at the inception of themodel 400, themodel builder 314 associates equal probabilities with each of the transitions from aparticular cluster 410 to each of the otherpossible clusters 410. Furthermore, because thedevice 300 can reside inmultiple clusters 410 at the same time (e.g., acluster 238 for popular applications and acluster 238 for first person shooter games), the sum of the probabilities of all of the possible transitions from aparticular cluster 410 can be greater than one. In some implementations, themodel builder 314 aggregates user behavior or other activity data. In these implementations, themodel builder 314 utilizes the aggregated user behavior to seed the initial probabilities of themodel 400, as opposed to allclusters 238 being equally probable. - The
model builder 314 trains themodels 400 with observations corresponding to thedevice 300 and the actions of the user commanding theuser device 300. Examples of observations can include, but are not limited to, a current time, a day of the week, an active application (i.e., the application currently being used by the user), active media content, an active search query, the output of the accelerometer, a video signal from the camera, the type of network connection that theuser device 300 is currently using, and the current location of thedevice 300. The types of observations used to train themodel 400 can vary based on the type of theuser device 300. In addition to the time of day, the day of the week, and the active application, the designer of the model 400 (e.g., a developer) can select which types of observations to include in the training data. For example, if theuser device 300 is a smartphone, theperipheral devices 350 of theuser device 300 may include a camera, a gyroscope, an accelerometer, and a global positioning system (GPS) receiver. In such a scenario, themodel builder 314 can be configured to record the output of theseperipheral devices 350. The output of the gyroscope, accelerometer, and/or GPS devices can be used to infer whether theuser device 300 is in transit, and if so, what type of transit (e.g., running, walking, driving). Similarly, the output of the camera can be used to infer the identity of the user of the device 300 (e.g., mother, father, or child). - Examples of actions of a user commanding the
user device 300 can include a command to turn thedevice 300 on, a command to launch anapplication 322, a command to play amedia content 326, and/or a command to provide asearch query 324 to a remote server. Further, the lack of a command from the user can be interpreted as an instruction to continue performing the same function or activity. Themodel builder 314 treats each of these commands as transitions between clusters 410 (e.g., from the home screen to an application, from an application to another application, and playing a media file from a media player application). Each time thecurrent cluster 410 of theuser device 300 transitions (e.g., theuser device 300 launches a new application or sends a search query 324), themodel builder 314 updates themodel 400 with the observation, thecluster 410 to which theuser device 300 transitioned, and thecluster 410 from which theuser device 300 transitioned. When the model builder adds new data (e.g., observed conditions and state transitions) to the HMM 400, themodel builder 314 can calculate new probabilities for all or some of the potential state transitions from theprevious cluster 410 of theuser device 300. - The behavior monitor 316 monitors the
user device 300 and the commands issued by the user. The behavior monitor 316 can execute as a background process. The behavior monitor 316 may be configured to monitor thenetwork interface 330, theuser interface 340, theperipheral devices 350, and/or any applications being executed by the processing device 310. The behavior monitor 316 can maintain a set of current observations based on the monitoring. For example, the behavior monitor 316 can monitor whether theuser device 300 is communicating over Wi-Fi or over a cellular network, the output of the accelerometer of theuser device 300 over a recent period of time, the most recent GPS coordinates of theuser device 300, and/or an image captured by the camera of theuser device 300. - The behavior monitor 316 further monitors the
current cluster 410 of thedevice 300. Thecurrent state 410 of theuser device 300 can define an action currently being performed on theuser device 300. Thecurrent cluster 410 of thedevice 300 can be represented using a cluster ID corresponding to thecluster 238 that is currently implicated by theuser device 300. In some scenarios, the current action may implicate more than one cluster (e.g., an application belongs to a “violent games” cluster and a “popular games” cluster). The behavior monitor 316 can identify thecurrent cluster 410 by determining theactive application 322,active media content 326, oractive search query 324 and referencing the correspondingindex entry 242 of theinverted index 240. Theinverted index 240 outputs the cluster ID or IDs of thecluster 238 orclusters 238 to which theactive application 322,active media content 326, oractive search query 324 belongs. The behavior monitor 316 can provide the set of observations to themodel builder 314 and thepersona predictor 318. - The
persona predictor 318 predicts theactive persona 100 of theuser device 300 based on the current user device state of theuser device 300. Thepersona predictor 318 can input the current user device state of theuser device 300 to the machine learnedmodel 400, and themodel 400 outputs a set of probabilities. Each probability is associated with adifferent cluster 238. In an example implementation, themodel 400 can output a vector, AP, in the form of <P(CJ->C0), P1(CJ->C1), P1(CJ->C2), . . . PN(CJ->CN-1)> where there are N possible clusters that theuser device 300 can transition to from the current cluster, CJ, of thedevice 300, including remaining in the current cluster (i.e., CJ->CJ). In some implementations, theactive persona 100 is set equal to the output vector. Put another way, theactive persona 100 can be represented by the combination of probabilities of transitioning to each possible cluster from the current cluster. In some situations, the sum of the probabilities may exceed one because theuser device 300 can exist in multiple clusters at the same time. For example, if the user is watching a popular new horror movie, the popular horror movie may implicate acluster 238 corresponding to newly released movies and acluster 238 corresponding to horror movies. - Additionally or alternatively, the
persona predictor 318 can match the vectors output by themodel 400 to persona categories. Persona categories can be predetermined categories. Examples of predetermined categories can include: “productivity applications,” “electronic reader applications,” “zombie shooter games,” “popular games,” “media player applications,” “shopping,” “organization applications,” “play kids music,” “play popular TV shows,” and “searching for movies.” The foregoing list of categories is provided for example only and any category can be used. Thepersona predictor 318 can match a vector to a category in any suitable manner. In some implementations, thepersona predictor 318 can reference a set of rules for matching the vector to a category. For example, a rule can state that if the probability that the user device will transition to an item of interest in afirst cluster 238 is greater than 0.8, then the output vector matches to category A. Furthermore, in some implementations, the categories do not need to be “human understandable.” The rules for matching a vector to a category can be learned over time by a machine learner or hard-coded. - Table 1 illustrates an example use of a model 400 (e.g., an HMM). The
model 400 receives a current user device state (i.e., the current cluster of the user and features corresponding to user device 300) and determines theactive persona 100 based thereon. -
TABLE 1 INPUT CURRENT CLUSTER (last output) Time, t Cat1 Cat2 Cat3 Cat4 0 0 0 0 0 1 0 0 0 0 2 0.95 0 0 0 3 0 1 0 0 . . . N 0 0 0 0.95 Features Cluster 5 minutes ago Time, t Cat1 Cat2 Cat3 Cat4 0 0 0 0 0 1 0 0 0 0 2 0 1 0 0 3 0 0 0 0 . . . N 0 0 0 0 Features Cluster 30 minutes ago Time/step Cat1 Cat2 Cat3 Cat4 0 0 0 0 0 1 0 0 0 0 2 0 1 0 0 3 0 1 0 0 . . . N 0 0 0 1 Features Device features Time, t Idle Screen Accelerometer 0 1 0 0 1 1 0 0 2 0 1 0 3 1 0 1 . . . N 0 1 1 User-location in the past 30 minutes Time, t Moving fast? Still? At Loc 1At Loc2 At Loc3 0 0 1 1 0 0 1 0 1 1 0 0 2 0 1 1 0 0 3 0 0 1 0 0 . . . N 1 0 0 1 0 Time Away from Time, t home? 0-6 6-9 9-12 12-15 15-18 18-21 21-24 DOW 0 0 1 0 0 0 0 0 0 M 1 0 0 1 0 0 0 0 0 M 2 0 0 1 0 0 0 0 0 M 3 0 0 1 0 0 0 0 0 M N 1 0 1 0 0 1 0 0 F Output NEXT STATE Time, t Cat1 Cat2 Cat3 Cat4 0 0.1 0.1 0.1 0.1 1 0.8 0.15 0 0.05 2 0 0.95 0 0 3 0 0.95 0 0 N 0 0.1 0.8 0.1 - In the example of Table 1, the user device can operate in four or more clusters 410 (e.g., Cat1=“News and Weather,” Cat2=“Game Mode,” Cat3=“Entertainment,” Cat4=“Productivity Apps” . . . ). Each of the
clusters 410 can correspond to different sets of items of interest (e.g., applications, media contents, and/or queries). Furthermore, Table 1 provides information regarding N or more different times (t=0, t=1, t=2, t=3 . . . t=N). Themodel 400 receives, as input, the current cluster of the user device 300 (e.g., the current category being used by the user). Themodel 400 further receives an observation set 420 (i.e., features corresponding to the user and/or the user device 300). In the example of Table 1, themodel 400 receives previous clusters of the user device 300 (e.g., the cluster of the user device five minutes ago and thirty minutes ago), a present location of the device 300 (e.g., the location in the previous thirty minutes), whether thedevice 300 is moving quickly at the present location (e.g., is the accelerometer signal indicative of a speed greater than a threshold speed), the status of the screen, the status of the accelerometer, and the status of the device 300 (e.g., idle/in use). - In the illustrated example, at t=0 (i.e., Monday morning between midnight and 6:00 AM), the
device 300 is idle and is not accelerating. The user is “quiet” and there is no strong prediction as to the category of thedevice 300. Thus, theactive persona 100 may be represented as <0.1, 0.1, 0.1, 0.1 . . . >. At t=1 (i.e., Monday morning at 6 AM), the user is idle but themodel 400 is most likely in Cat1 (e.g., “News and Weather”), but also may be in Cat2 (e.g., “Games”). Thus, theactive persona 100 may be represented by <0.8, 0.15, 0, 0.05 . . . >. The foregoingactive persona 100 may be learned by themodel 400 if the user typically checks the news and/or the weather upon waking up, but sometimes plays a game. At t=0 (i.e., Monday morning at 7:00 AM), the user is active and currently playing a game. The output of themodel 400 changes to predict that the user is in Cat2 (e.g., “Games”). In this instance, the prediction at t=1 was incorrect. Themodel builder 314 can update themodel 400 accordingly, such that the next time theuser device 300 encounters the same or similar observation set 420, the probability that the user is operating in a game mode is increased. At t=3 (i.e., Monday morning at 8:00 AM), theuser device 300 is at Loc1 (e.g., the home of the user) and moving, but not moving fast (i.e., the accelerometer signal is on, but does not exceeds a threshold). At t=3, themodel 400 outputs a prediction that theuser device 300 is likely in Cat2 (e.g., gaming). Theactive persona 100 may be <0, 0.95, 0, 0, . . . >. At t=N, (i.e., Friday at 5:00 PM) theuser device 300 is idle, but the user is moving fast. In this scenario, the user may be driving home from work (Loc2). At t=5, the user has used the productivity applications (e.g., email application) in the previous three hours. Themodel 400 predicts that the user is likely to switch to Cat3 (e.g., “Entertainment), and is less likely to play a game or use a productivity application (each with a 10% probability). At t=5, theactive persona 100 may be <0, 0.1, 0.8, 0.1 . . . >. The example of Table 1 is not intended to limit the scope of the disclosure. The category lists and the features are provided for example and additional alternative categories and features may be implemented in themodel 400. - The
active persona 100 of theuser device 300 can be used for many different functions. In some implementations, a search application can request theactive persona 100 from thepersona predictor 318 when the search application is launched or otherwise made active. Thepersona predictor 318 provides anactive persona 100 of theuser device 300 to the search application, which in turn includes theactive persona 100 with the metadata of thesearch query 324. In this way, when thesearch engine 120 executes the search using thesearch query 324, theactive persona 100 can provide additional context to thesearch query 324. For example, if thesearch query 324 is (“watch lions”) and theactive persona 100 indicates that the user is more likely (e.g., 0.8 probability) searching for sports related content than searching for children's educational content (0.05), thesearch engine 120 can return search results that indicate applications that allow users to watch live sports (e.g., watch the Detroit Lions play football) over applications that have animal related videos (e.g., watch videos of lions). In this situation, thepersona predictor 318 can continue to update theactive persona 100 of theuser device 300 as theactive persona 100 changes. - In another example, the user may be using a free application (e.g., music streaming application) that derives revenue from displaying ads to users. In this example, an
application server 130 of the free application can request theactive persona 100 of the user from thepersona predictor 318. Thepersona predictor 318 can provide theactive persona 100 to theapplication server 130. In response to theactive persona 100, theserver 130 can select an advertisement to provide to the user based on theactive persona 100. For instance, if theactive persona 100 indicates that the user is likely to watch a popular TV show after listening to music, the free application can display or stream an ad for a new popular TV show. Alternatively, if theactive persona 100 indicates that the user is likely to play a violent video game after listening to music, then theapplication server 130 may provide an ad for a new first person shooter game. - In yet another example, the operating system of the
user device 300 may request theactive persona 100 of theuser device 300. Thepersona predictor 318 can provide theactive persona 100 to the operating system, which can use theactive persona 100 to reconfigure the home screen displayed by theuser interface 340. For instance, if theactive persona 100 indicates that the user is likely to watch a movie, the operating system can reconfigure the home screen to display the media player application icons prominently. Alternatively, if theactive persona 100 indicates that the user is likely to open email or use a document viewer, the operating system can reconfigure the home screen to display business-related application icons prominently. - The
active persona 100 can be used for any other number of functions. In this way, the use ofactive personas 100 can enhance the user experience of theuser device 300. Furthermore, in some implementations the user can manually select his or heractive persona 100. In these implementations, the user can input his or heractive persona 100 via a graphical user interface of theuser device 300. Theuser device 300 can then use the manually inputtedactive persona 100 in any of the manners described above. - Referring now to
FIG. 5 , anexample method 500 for predicting anactive persona 100 of auser device 300 is illustrated. Themethod 500 is explained with reference to the components of theuser device 300. It is understood that any other suitable components may execute themethod 500. - At
operation 510, the behavior monitor 316 captures a current user device state of auser device 300. In order to determine the current user device state of theuser device 300, the behavior monitor 316 determines what the active process of theuser device 300 is to capture thecurrent state 410. That is, the behavior monitor 316 can identify theactive application 322,active media content 326, oractive search query 324. The behavior monitor 316 then references theinverted index entry 242 of the active process to identify the cluster ID or IDs of theclusters 238 implicated by the active process. The cluster ID or IDs represent the currentuser device state 410 or states 410 of theuser device 300. As discussed, thepersona server 200 may select theindex entries 242 so as to minimize collisions, but there may be overlap between one ormore clusters 238. - In addition to the active process, the behavior monitor 316 captures the current observations of the
user device 300. The behavior monitor 316 can capture the states of thenetwork interface device 330, theuser interface 340, and/or theperipheral devices 350. For example, the behavior monitor 316 can determine whether theuser device 300 is operating on a Wi-Fi network or on a cellular network, the output of the accelerometer over a previous amount of time (e.g., five minutes), the output of the GPS system, the time of the day, and the day of the week. Depending on the capabilities of thedevice 300 and other considerations, the behavior monitor 316 can capture other current observations as well, including a current temperature, the output of the video camera, and the amount of ambient light. The behavior monitor 316 can maintain a current observation data structure that defines the set of observations that the behavior monitor 316 captures. Each time the behavior monitor 316 captures a new observation, the behavior monitor 316 can update the data structure with the new observation. - At
operation 512, thepersona predictor 318 inputs the current user device state into amodel 400. Atoperation 514, thepersona predictor 318 obtains theactive persona 100 from themodel 400. In some implementations, themodel 400 outputs one or more vectors of probabilities. As previously indicated, each probability is associated with a different cluster and indicates a likelihood that theuser device 300 will transition to an item of interest corresponding to the cluster. In some implementations, theactive persona 100 orpersonas 100 are set equal to the output vector or vectors. In other implementations, thepersona predictor 318 matches the vector to an active persona category to obtain theactive persona 100 orpersonas 100 of theuser device 300. - At
operation 516, thepersona predictor 318 performs a procedure using theactive persona 100. In some implementations, thepersona predictor 318 can provide theactive persona 100 to a process executing on the user device 300 (e.g., the operating system of the user device 300) or to a remote device (e.g.,application server 130 or an advertisement server). In some implementations, thepersona predictor 318 can provide theactive persona 100 to a search application, which in turn includes theactive persona 100 in asearch query 324 that it provides to asearch engine 120. In some implementations, the operating system of theuser device 300 can rearrange the home screen of theuser device 300 in accordance with theactive persona 100. For instance, if theactive persona 100 indicates that the user is likely to play a violent video game, the operating system can rearrange the home screen to show icons of the violent video game on the home screen. In this way, theactive persona 100 can be used to enhance the user's overall experience of theuser device 300. - At
operation 518, the behavior monitor 316 captures a new user device state of theuser device 300. The behavior monitor 316 can monitor the action of the user of theuser device 300 after thepersona predictor 318 predicts theactive persona 100. In some implementations, the behavior monitor 316 waits until theuser device 300 changes the current active process (e.g., the user transitions to a different application or media content 326) to capture the new user device state. If the user does not change the active process for a timeout period (e.g., five minutes), the behavior monitor 316 determines that theuser device 300 remained in the same user device state. In other implementations, the behavior monitor 316 captures a new user device state of theuser device 300 in intervals of predetermined length (e.g., every two minutes). In these implementations, the new user device state may be the same as the current user device state if theuser device 410 does not change the active process during the predetermined interval. The behavior monitor 316 can capture other observations in addition to capturing the new active process. - At
operation 520, themodel builder 314 updates themodel 400. In some implementations, themodel builder 314 updates the training data of themodel 400 with the current user device state (determined at operation 510) and the new user device state (determined at operation 520) as captured by the behavior monitor 316. In this way, themodel builder 314 defines a known transition given a current user state and a subsequent new user device state. - The
method 500 can execute in this manner upon receiving a request for anactive persona 100 or at predetermined intervals (e.g., every five minutes when theuser device 300 is in use). Themethod 500 is provided for example only and variations of themethod 500 are contemplated. Further, some of the operations may be performed out of order or concurrently. - Various implementations of the systems and techniques described here can be realized in digital electronic and/or optical circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
- These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor, and can be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. As used herein, the terms “machine-readable medium” and “computer-readable medium” refer to any computer program product, non-transitory computer readable medium, apparatus and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal used to provide machine instructions and/or data to a programmable processor.
- Implementations of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Moreover, subject matter described in this specification can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a computer readable medium for execution by, or to control the operation of, data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter affecting a machine-readable propagated signal, or a combination of one or more of them. The terms “data processing apparatus,” “computing device,” and “computing processor” encompass all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them. A propagated signal is an artificially generated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal that is generated to encode information for transmission to suitable receiver apparatus.
- A computer program (also known as an application, program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.
- The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).
- Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read only memory or a random access memory or both. The essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio player, a Global Positioning System (GPS) receiver, to name just a few. Computer readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto optical disks; and CD ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.
- To provide for interaction with a user, one or more aspects of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen for displaying information to the user and optionally a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input. In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's client device in response to requests received from the web browser.
- One or more aspects of the disclosure can be implemented in a computing system that includes a backend component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a frontend component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the subject matter described in this specification, or any combination of one or more such backend, middleware, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (“LAN”) and a wide area network (“WAN”), an inter-network (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks). Communication networks can include cellular networks and/or telephone networks.
- The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. In some implementations, a server transmits data (e.g., an HTML page) to a client device (e.g., for purposes of displaying data to and receiving user input from a user interacting with the client device). Data generated at the client device (e.g., a result of the user interaction) can be received from the client device at the server.
- While this specification contains many specifics, these should not be construed as limitations on the scope of the disclosure or of what may be claimed, but rather as descriptions of features specific to particular implementations of the disclosure. Certain features that are described in this specification in the context of separate implementations can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable sub-combination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a sub-combination or variation of a sub-combination.
- Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multi-tasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.
- A number of implementations have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the disclosure. Accordingly, other implementations are within the scope of the following claims. For example, the actions recited in the claims can be performed in a different order and still achieve desirable results.
Claims (20)
1. A method comprising:
at a mobile computing device including a touchscreen display, a non-transitory memory that stores a plurality of native applications that are installed at the mobile computing device, and one or more processors:
displaying, on the touchscreen display, a screen that includes icons for the native applications installed on the mobile computing device, wherein the mobile computing device is configured to detect a user selection of a particular icon and launch the native application corresponding with the selected icon;
determining that a first set of native applications are launched at the mobile computing device at or near a first time of day;
determining whether a current time of day matches the first time of day; and
upon determining that the current time of day matches the first time of day, rearranging the icons on the screen into a first arrangement of icons, wherein the first arrangement of icons displays the icons for the first set of native applications above a line.
2. The method of claim 1 , wherein the first time of day comprises a time that occurs during the morning of the day.
3. The method of claim 1 , wherein the first set of native applications includes an email application and a news application; and
wherein rearranging the icons into the first arrangement comprises displaying an email icon for the email application and a news icon for the news application above the line.
4. The method of claim 1 , further comprising:
determining that a second set of native applications are launched at the mobile computing device at or near a second time of day;
determining whether a current time of day matches the second time of day; and
upon determining that the current time of day matches the second time of day, rearranging the icons on the screen into a second arrangement of icons, wherein the second arrangement of icons displays the icons for the second set of native applications above the line.
5. The method of claim 4 , wherein the second time of day comprises a time that occurs during the evening of the day.
6. The method of claim 4 , wherein the second set of native applications includes a messaging application; and
wherein rearranging the icons into the second arrangement comprises displaying a messaging icon for the messaging application above the line.
7. The method of claim 1 , further comprising:
receiving a user selection of one of the icons displayed above the line; and
launching the native application corresponding with the selected icon.
8. A mobile computing device comprising:
a touchscreen display;
a non-transitory memory that stores a plurality of native applications that are installed at the mobile computing device; and
one or more processors configured to:
display, on the touchscreen display, a screen that includes icons for the native applications installed on the mobile computing device, wherein the mobile computing device is configured to detect a user selection of a particular icon and launch the native application corresponding with the selected icon;
determine that a first set of native applications are launched at the mobile computing device at or near a first time of day;
determine whether a current time of day matches the first time of day; and
upon determining that the current time of day matches the first time of day, rearrange the icons on the screen into a first arrangement of icons, wherein the first arrangement of icons displays the icons for the first set of native applications above a line.
9. The mobile computing device of claim 8 , wherein the first time of day comprises a time that occurs during the morning of the day.
10. The mobile computing device of claim 8 , wherein the first set of native applications includes an email application and a news application; and
wherein rearranging the icons into the first arrangement comprises displaying an email icon for the email application and a news icon for the news application above the line.
11. The mobile computing device of claim 1 , wherein the one or more processors are further configured to:
determine that a second set of native applications are launched at the mobile computing device at or near a second time of day;
determine whether a current time of day matches the second time of day; and
upon determining that the current time of day matches the second time of day, rearrange the icons on the screen into a second arrangement of icons, wherein the second arrangement of icons displays the icons for the second set of native applications above the line.
12. The mobile computing device of claim 11 , wherein the second time of day comprises a time that occurs during the evening of the day.
13. The mobile computing device of claim 11 , wherein the second set of native applications includes a messaging application; and
wherein rearranging the icons into the second arrangement comprises displaying a messaging icon for the messaging application above the line.
14. The mobile computing device of claim 8 , wherein the one or more processors are further configured to:
receive a user selection of one of the icons displayed above the line; and
launch the native application corresponding with the selected icon.
15. A computer program product encoded on a non-transitory computer readable storage medium comprising instructions that when executed by a mobile computing device, cause the mobile computing device to perform operations comprising:
displaying a screen that includes icons for native applications installed on the mobile computing device, wherein the mobile computing device is configured to detect a user selection of a particular icon and launch the native application corresponding with the selected icon;
determining that a first set of native applications are launched at the mobile computing device at or near a first time of day;
determining whether a current time of day matches the first time of day; and
upon determining that the current time of day matches the first time of day, rearranging the icons on the screen into a first arrangement of icons, wherein the first arrangement of icons displays the icons for the first set of native applications above a line.
16. The computer program product of claim 15 , wherein the first time of day comprises a time that occurs during the morning of the day.
17. The computer program product of claim 15 , wherein the first set of native applications includes an email application and a news application; and
wherein rearranging the icons into the first arrangement comprises displaying an email icon for the email application and a news icon for the news application above the line.
18. The computer program product of claim 15 , wherein the operations further comprise:
determining that a second set of native applications are launched at the mobile computing device at or near a second time of day;
determining whether a current time of day matches the second time of day; and
upon determining that the current time of day matches the second time of day, rearranging the icons on the screen into a second arrangement of icons, wherein the second arrangement of icons displays the icons for the second set of native applications above the line.
19. The computer program product of claim 18 , wherein the second time of day comprises a time that occurs during the evening of the day.
20. The computer program product of claim 18 , wherein the second set of native applications includes a messaging application; and
wherein rearranging the icons into the second arrangement comprises displaying a messaging icon for the messaging application above the line.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/180,022 US20160291840A1 (en) | 2013-12-27 | 2016-06-11 | Method and Apparatus for Automated Time-Adaptive Rearrangement of Visual Elements in Mobile Computing Device |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201361921245P | 2013-12-27 | 2013-12-27 | |
US14/446,991 US9405838B2 (en) | 2013-12-27 | 2014-07-30 | Determining an active persona of a user device |
US15/180,022 US20160291840A1 (en) | 2013-12-27 | 2016-06-11 | Method and Apparatus for Automated Time-Adaptive Rearrangement of Visual Elements in Mobile Computing Device |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/446,991 Continuation US9405838B2 (en) | 2013-12-27 | 2014-07-30 | Determining an active persona of a user device |
Publications (1)
Publication Number | Publication Date |
---|---|
US20160291840A1 true US20160291840A1 (en) | 2016-10-06 |
Family
ID=53479489
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/446,991 Expired - Fee Related US9405838B2 (en) | 2013-12-27 | 2014-07-30 | Determining an active persona of a user device |
US15/180,022 Abandoned US20160291840A1 (en) | 2013-12-27 | 2016-06-11 | Method and Apparatus for Automated Time-Adaptive Rearrangement of Visual Elements in Mobile Computing Device |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/446,991 Expired - Fee Related US9405838B2 (en) | 2013-12-27 | 2014-07-30 | Determining an active persona of a user device |
Country Status (5)
Country | Link |
---|---|
US (2) | US9405838B2 (en) |
EP (1) | EP3087501A4 (en) |
KR (1) | KR101959368B1 (en) |
CN (1) | CN106104510A (en) |
WO (1) | WO2015099893A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018118547A1 (en) * | 2016-12-21 | 2018-06-28 | Microsoft Technology Licensing, Llc | Generation of application allowed lists for machines |
US11443112B2 (en) * | 2019-09-06 | 2022-09-13 | International Business Machines Corporation | Outcome of a natural language interaction |
US11861463B2 (en) * | 2019-09-06 | 2024-01-02 | International Business Machines Corporation | Identifying related messages in a natural language interaction |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10300393B2 (en) * | 2014-10-21 | 2019-05-28 | Activision Publishing, Inc. | Music based video game with components |
US11363460B1 (en) * | 2015-03-03 | 2022-06-14 | Amazon Technologies, Inc. | Device-based identification for automated user detection |
US10198498B2 (en) * | 2015-05-13 | 2019-02-05 | Rovi Guides, Inc. | Methods and systems for updating database tags for media content |
US10726438B2 (en) * | 2015-09-03 | 2020-07-28 | Microsoft Technology Licensing, Llc | Personalized contextual coupon engine |
WO2017042733A2 (en) * | 2015-09-09 | 2017-03-16 | Cell Buddy Network Ltd. | Alias communication |
US20170097827A1 (en) * | 2015-10-06 | 2017-04-06 | Microsoft Technology Licensing, Llc | Role-specific device behavior |
US10601953B2 (en) * | 2016-09-22 | 2020-03-24 | Adobe Inc. | Decomposing media content accounts for persona-based experience individualization |
US10512846B2 (en) | 2017-03-07 | 2019-12-24 | Sony Interactive Entertainment LLC | Emulating player behavior after player departure |
US10946280B2 (en) | 2017-03-08 | 2021-03-16 | Sony Interactive Entertainment LLC | In-game reactions to interruptions |
CN107463660B (en) * | 2017-07-31 | 2020-10-16 | 广州虎牙信息科技有限公司 | Method for measuring and calculating data of active users of products and computer equipment |
KR102138761B1 (en) * | 2017-09-29 | 2020-07-28 | 이다커뮤니케이션즈(주) | System for providing persona design guideline based on customer pattern and method using the same |
KR102119238B1 (en) * | 2018-09-28 | 2020-06-26 | 이다커뮤니케이션즈(주) | Personalized persona management server and mangment method thereof |
US11222061B2 (en) * | 2019-03-28 | 2022-01-11 | Facebook, Inc. | Generating digital media clusters corresponding to predicted distribution classes from a repository of digital media based on network distribution history |
GB2583718A (en) * | 2019-05-01 | 2020-11-11 | Samsung Electronics Co Ltd | Method, apparatus and computer program for updating a cluster probability model |
US20210034945A1 (en) * | 2019-07-31 | 2021-02-04 | Walmart Apollo, Llc | Personalized complimentary item recommendations using sequential and triplet neural architecture |
US11321406B2 (en) | 2019-07-31 | 2022-05-03 | Walmart Apollo, Llc | Personalized ranking using deep attribute extraction and attentive user interest embeddings |
US10803065B1 (en) * | 2019-09-12 | 2020-10-13 | Business Objects Software Ltd | Microcubes |
CN113449471B (en) * | 2021-06-25 | 2022-08-30 | 东北电力大学 | Wind power output simulation generation method for continuously improving MC (multi-channel) by utilizing AP (access point) clustering-skipping |
US20230161890A1 (en) * | 2021-11-24 | 2023-05-25 | X Development Llc | Persona prediction for access to resources |
KR20230120383A (en) * | 2022-02-09 | 2023-08-17 | 삼성전자주식회사 | Method and device for synchronizing device settings |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140059599A1 (en) * | 2012-08-17 | 2014-02-27 | Flextronics Ap, Llc | Dynamic arrangment of an application center based on usage |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030191753A1 (en) * | 2002-04-08 | 2003-10-09 | Michael Hoch | Filtering contents using a learning mechanism |
US7594121B2 (en) * | 2004-01-22 | 2009-09-22 | Sony Corporation | Methods and apparatus for determining an identity of a user |
US20060074883A1 (en) | 2004-10-05 | 2006-04-06 | Microsoft Corporation | Systems, methods, and interfaces for providing personalized search and information access |
CN102667839A (en) * | 2009-12-15 | 2012-09-12 | 英特尔公司 | Systems, apparatus and methods using probabilistic techniques in trending and profiling and template-based predictions of user behavior in order to offer recommendations |
US10074094B2 (en) | 2010-03-09 | 2018-09-11 | Excalibur Ip, Llc | Generating a user profile based on self disclosed public status information |
CN101826114B (en) * | 2010-05-26 | 2012-05-09 | 南京大学 | Multi Markov chain-based content recommendation method |
US9536000B2 (en) * | 2010-10-29 | 2017-01-03 | Excalibur Ip, Llc | Profile prediction for shared computers |
-
2014
- 2014-07-30 US US14/446,991 patent/US9405838B2/en not_active Expired - Fee Related
- 2014-11-12 EP EP14875621.6A patent/EP3087501A4/en not_active Withdrawn
- 2014-11-12 KR KR1020167020673A patent/KR101959368B1/en active IP Right Grant
- 2014-11-12 WO PCT/US2014/065184 patent/WO2015099893A1/en active Application Filing
- 2014-11-12 CN CN201480076392.8A patent/CN106104510A/en active Pending
-
2016
- 2016-06-11 US US15/180,022 patent/US20160291840A1/en not_active Abandoned
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140059599A1 (en) * | 2012-08-17 | 2014-02-27 | Flextronics Ap, Llc | Dynamic arrangment of an application center based on usage |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018118547A1 (en) * | 2016-12-21 | 2018-06-28 | Microsoft Technology Licensing, Llc | Generation of application allowed lists for machines |
US10757110B2 (en) | 2016-12-21 | 2020-08-25 | Microsoft Technology Licensing, Llc | Generation of application allowed lists for machines |
US11443112B2 (en) * | 2019-09-06 | 2022-09-13 | International Business Machines Corporation | Outcome of a natural language interaction |
US11861463B2 (en) * | 2019-09-06 | 2024-01-02 | International Business Machines Corporation | Identifying related messages in a natural language interaction |
Also Published As
Publication number | Publication date |
---|---|
KR101959368B1 (en) | 2019-03-18 |
US20150186535A1 (en) | 2015-07-02 |
CN106104510A (en) | 2016-11-09 |
US9405838B2 (en) | 2016-08-02 |
WO2015099893A1 (en) | 2015-07-02 |
EP3087501A4 (en) | 2017-08-30 |
KR20160124087A (en) | 2016-10-26 |
EP3087501A1 (en) | 2016-11-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9405838B2 (en) | Determining an active persona of a user device | |
US11810576B2 (en) | Personalization of experiences with digital assistants in communal settings through voice and query processing | |
US10614142B2 (en) | Personalized search results | |
US20240104126A1 (en) | Methods, systems, and media for modifying the presentation of contextually relevant documents in browser windows of a browsing application | |
US11341153B2 (en) | Computerized system and method for determining applications on a device for serving media | |
CN107660284B (en) | Search improvement based on machine learning | |
US10885076B2 (en) | Computerized system and method for search query auto-completion | |
US9589149B2 (en) | Combining personalization and privacy locally on devices | |
US8930294B2 (en) | Predicting user activity based on usage data received from client devices | |
US20180253469A1 (en) | Techniques for reformulating search queries | |
US20180091468A1 (en) | Indicating Live Videos for Trending Topics on Online Social Networks | |
US20140379746A1 (en) | System and method for providing topic cluster based updates | |
US20140282493A1 (en) | System for replicating apps from an existing device to a new device | |
WO2019140703A1 (en) | Method and device for generating user profile picture | |
US20140129331A1 (en) | System and method for predicting momentum of activities of a targeted audience for automatically optimizing placement of promotional items or content in a network environment | |
US20120278164A1 (en) | Systems and methods for recommending advertisement placement based on in network and cross network online activity analysis | |
US20130145024A1 (en) | Method and apparatus for transfer of radio resource allocation | |
US20110257960A1 (en) | Method and apparatus for context-indexed network resource sections | |
JP6543774B2 (en) | Information ranking based on computing device properties | |
US10986408B1 (en) | Systems and methods for recommending linear content | |
US20150186940A1 (en) | Techniques For Generating Advertisements | |
US20170249325A1 (en) | Proactive favorite leisure interest identification for personalized experiences | |
CA2864127A1 (en) | Systems and methods for recommending advertisement placement based on in network and cross network online activity analysis | |
US10657165B1 (en) | Inferring membership in a group | |
JP2022528212A (en) | Systems and methods for generating and selecting local content to improve the security and utilization of network resources |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: SAMSUNG ELECTRONICS CO., LTD., KOREA, REPUBLIC OF Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:QUIXEY, INC.;REEL/FRAME:043972/0735 Effective date: 20171019 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |