Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing conversion path performance measures and reports. In one aspect, user interaction data are obtained, were the user interaction data specify user interactions for a plurality of conversions. User interactions that are associated with each conversion are selected from the user interaction data, where the associated user interactions for each conversion are user interactions with a converting user during the conversion cycle for the conversion. Using the user interaction data for the selected user interactions, a quantity of user interactions that are associated with each conversion and occurred during the conversion cycle for the conversion are determined. In turn, conversion path performance measures are computed and reports specifying the conversion path performance measures are generated.

Abstract: A media creative attribution method includes determining a response profile within an attribution time window, the response profile being a portion of a unique visitor (UV) curve associated with a website. In some cases, a shadow baseline analysis is run on every media creative that aired within an extended time window to determine whether to adjust the response profile. A total lift within the attribution time window is determined utilizing a baseline of the UV curve. A weight for each media creative that aired within the attribution time window is determined. Utilizing the weight, the total lift is allocated to individual media creatives that aired within the attribution time window. The allocated attribution can be utilized to generate performance metrics relating to the individual media creatives that aired within the attribution time window. The performance metrics such as cost per visitor can be visualized through a user interface or dashboard.

Abstract: The invention provides a method and system for a financial institution for capturing the business of a financial services customer. The system may include a data access system for obtaining and storing data; and an analysis system for processing the data. The analysis system (1) selects a study group of customers from the data; (2) observes the study group so as to produce a profile of each study group customer's financial information; (3) produces a mathematical representation of an interrelationship between study group customers' profiles and the data; (4) applies the mathematical representation to customers not within the study group in order to estimate those customers' profiles; (5) identifies a portion of the customer's monetary flow eligible to be conducted through the financial institution, the identification based in part on the customer's profile; and (6) does not require direct input from the customer.

Abstract: Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform functions comprising: generating one or more feature vectors for a user, the one or more feature vectors at least comprising transaction-based features and slot-based features; generating, using a machine learning architecture, a repurchase prediction for the user based, at least in part, on the one or more feature vectors; generating, using the machine learning architecture, a time slot prediction for the user based, at least in part, on the one or more feature vectors, the time slot prediction predicting a time slot desired by the user for an upcoming transaction; and executing a reservation function that facilitates reserving of the time slot for the user. Other embodiments are disclosed herein.

Abstract: In some implementations, a computing device can proactively determine a destination and request traffic information for routes from a starting location to the destination. In some implementations, a computing device can identify some routes between a starting location and a destination as non-recommended routes and recommend other routes. In some implementations, a computing device can rank routes between a starting location and a destination based on automatically-determined user interest. In some implementations, a computing device can determine a user is familiar with a route and adjust the information presented to the user about the route accordingly.

Abstract: A system may selectively deliver interactive portions of a piece of content, based for example on a recipient user's historical tendency to use interactive portions of prior content. Multiple levels of interactivity may be defined for a given piece of content, such as an advertisement, and different users may receive different levels of interactivity based on how much they use interactive features. Some users may receive no interactive features at all, while others may receive the interactive features, and despite the fact that all of the users' devices may be perfectly capable of processing the interactive features for the user. The omission from transmission may conserve bandwidth and transmission resources.

Abstract: A method for generating advertising content includes: during a content creation period, receiving a video advertisement from an entity; packaging the video advertisement with a pointer to the entity and a checkout portal for content linked to the video advertisement to generate an augmented video advertisement; during a live period succeeding the content creation period, for a first instance of the augmented video advertisement accessed at a first computing device associated with a first user, associating a first purchase completed within the checkout portal of the first instance of the augmented video advertisement with the entity based on the pointer; and, for a second instance of the augmented video advertisement accessed at a second computing device associated with a second user, associating a second purchase completed within the checkout portal of the second instance of the augmented video advertisement with the entity based on the pointer.

Abstract: A bidding assistance device includes a ratio calculation unit configured to calculate a ratio of an actual advertising cost index to a goal advertising cost index, the actual advertising cost index being based on an advertising delivery performance at a bidding time in an advertising period, the goal advertising cost index being requested by an advertiser in the advertising period, an adjustment factor calculation unit configured to calculate an adjustment factor from the ratio and from a relationship between the advertising period and the bidding time, the adjustment factor being used to adjust a bid price of an advertisement, and a determination unit configured to determine the bid price at the bidding time using the adjustment factor calculated by the adjustment factor calculation unit, the goal advertising cost index, and a forecast effectiveness index forecast from advertising effectiveness, the advertising effectiveness being obtained from the advertising delivery performance.

Abstract: In a general aspect, a cloud-based control platform remotely controls variable settings on a remotely-controlled system. In some aspects, the platform maintains a database of data objects for assets, and the data objects include static and dynamic hyperparameters associated with the asset and a template that specifies values of a variable associated with the asset for future time points. The platform updates a template for a data object by calculating target values for the variable for the future time points based on a target criterion, communicating with remote computer systems to determine current values of the dynamic hyperparameters, calculating scaled values for the future time points by applying a determined ratio to the target values, adjusting the scaled values based on the static hyperparameters, applying an override value to the adjusted scaled values. The remotely-controlled system is updated according to the updated template at each of the future time points.

Abstract: System architecture and methods for enabling an online real-time auction (either server-side or client-side) of advertising inventory that works in conjunction with ad serving technologies. In some implementations, the system architecture facilitates multiple parallel requests from an end-user's browser client to multiple real-time bidders who respond with a bid and advertisement that are sent to an ad serving system to be compared with other statically priced advertisements and exchange demand to determine the winning advertisements that will be displayed to the end-user. Data is aggregated for each bid and price limits are set based on the aggregations. In other implementations, all the bids from the multiple bidders are passed to the publisher for selection of a bid by the publisher.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for managing a relationship between content and an environment for provisioning the content. In one aspect, a method includes receiving a request for a content item; and in response to receiving the request: selecting a creative from a plurality of creatives, the creative including a reference to a profile associated with one or more elements; retrieving content data from one or more content feeds bound to the elements; and delivering the creative and the content data to a user device.

Abstract: A device detection system is configured to determine whether a device is located within a venue. The device detection system provides user identifiers associated with a device to an anonymization server. The anonymization server provides anonymous identifiers to the device detection system corresponding to the user identifiers. A content distribution system provides content items for a campaign associated with the venue to users. The content distribution system provides user identifiers to the anonymization server and receives anonymous identifiers from the anonymization server. The content distribution system provides the anonymous identifiers to the device detection system. The device detection system compares the sets of anonymous identifiers to identify users that received a content item and subsequently visited the venue.

Abstract: Methods, systems, and computer readable media related to generating a combined search query based on search parameters of a current search query of a user and search parameters of one or more previously submitted search quer(ies) of the user that are determined to be of the same line of inquiry as the current search query. Two or more search queries may be determined to share a line of inquiry when it is determined that they are within a threshold level of semantic similarity to one another. Once a shared line of inquiry has been identified and a combined search query generated, users may interact with the search parameters and/or the search results to update the search parameters of the combined search query.

Abstract: A system and methods are disclosed for creating and using an audience-responsiveness analytics index for television advertising, in particular, advertising for connected television viewing. The audience-responsiveness analytics index may be configured to provide data obtained or acquired by measuring audience responsiveness to video advertising placed in streaming content viewed over connected televisions. The audience-responsiveness analytics index may be a graph used to optimize advertising with households assigned scores for multiple categories of advertising.

Abstract: A computer-implemented process includes providing a request to access content associated with a website; receiving web page information associated with the website, and data for displaying a notification that indicates availability of a native computer application that is associated with the website; providing a request to access information associated with the native computer application; receiving information associated with the native computer application that describes interactions between a plurality of users of an application marketplace and a component of the application marketplace that provides and receives information related to the native computer application; and displaying the descriptive information for viewing by a user, including displaying information that indicates the interactions between the plurality of users and the component of the application marketplace.

Abstract: Methods and systems for providing an advanced hardware system for a self-service kiosk are disclosed herein. In specific embodiments of the invention, a payment interface device is provided. The device includes an integrated touch display, a means for receiving payment information, and a casing holding the integrated touch display and the means for receiving payment information. The device also includes an applications microprocessor and a secure microprocessor located within the casing. The applications microprocessor instantiates an operating system and stores instructions for execution by the operating system for operative communication with a peripheral bar code scanner and a peripheral display. The secure microprocessor stores instructions to encrypt payment information for the system and is in operative communication with the means for receiving payment information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting content items for presentation along with publisher resources. In one aspect, a method includes receiving a request for a content item for presentation at a user device with a publisher resource; receiving, from a content item provider a first selection value for each of a plurality of content items provided by the content item provider; determining, for each of the content items and using accessed distribution parameters, a second selection value distinct from the first selection value; and determining, for each of the content items and independent of a bid corresponding to the content item, a combined selection value based on the first selection value for the content item and the second selection value for the content item.

Abstract: A deal allocation platform is disclosed. The deal allocation platform may be configured to be used in multiple contexts, such as, for example, a batch processing context and a real-time context. Further, the deal allocation platform may be configured to reallocate users assigned to deals based on an approximation of a global optimization problem. In addition, the deal allocation platform may enable configuration of one or multiple experiments.

Abstract: User attribute data associated with a user may be used to resolve unresolved entities in connection with the performance of computer-based actions, e.g., actions resulting from action queries generated based at least in part on content being accessed on a computing device.

Abstract: A computer-implemented method, a computer program product, and a computer system for event-based dynamic prediction in location sharing on mobile devices. The computer system captures biometric data of a user, in response to starting location sharing requested by the user. The computer system collects action-based context of behavior of the user during the location sharing, determines activities of the user and one or more sharing users during the location sharing, determines context of the location sharing, tracks activities of the user and the one or more sharing users after the location sharing is ended, and tracks social network relationships between the user and the one or more sharing users. The computer system creates a knowledge corpus for training a machine learning model for prediction in a future location sharing event, using obtained information.

Abstract: A method and a system for using differential privacy techniques to provide axe obfuscation with respect to information included in an inventory axe list of available securities is provided. The method includes: obtaining first information to be included in a first inventory axe list to be published on a particular day; retrieving second information included in a second inventory axe list that was published on the previous day and/or several previous days; obfuscating the obtained first information by applying an algorithm based on the difference between the first information and the second information; and publishing the first inventory axe list by transmitting the obfuscated first information to a plurality of intended recipients. The quality of obfuscation may be measured and controlled as a function of desired privacy level and potential cost.

Abstract: An online system calculates bids for content items to display to users based on the value of a product described in the content item and the likelihood of a viewing user purchasing the product. The online system identifies an impression opportunity for an ad request and computes an expected value of the conversion and a likelihood of the conversion. The online system computes a bid amount based on the expected conversion value and the likelihood of the conversion. Bids based on the value of the conversion allow a third party system offering the product to optimize for the value of each conversion instead of the conversion rate.

Abstract: Highly user-specific data is used to calculate user intent to make a purchase and the value of such a purchase. User activity and information is aggregated, per user, for a set window of time and real-time data on recent site behavior is obtained. Aggregated and/or real-time data is considered by a predictive intent model (calculating the probability that the user will make a purchase) and a predictive value model (calculating the expected revenue such a purchase may generate). Weights, specific to each model, are assigned to predictor features tracked in the aggregated and/or real-time user data. The most highly-weighted features of the intent model relate to users' viewing history, and the most highly-weighted features of the value model relate to price and market. By these means, a user conversion value can be obtained, guiding the application of user acquisition strategies for different home sharing markets.

Abstract: Embodiments of the present disclosure provide methods, systems, apparatuses, and computer program products for predictive dynamic bidding rules generation for digital content objects.

Abstract: Prospective deals for a deal vendor's deal inventory may be scored according to one or more population-specific expected-yield scores, such as expected revenue yield, expected new customers yield, and the like. A prospective deal is categorized and characterized according to a number of yield-related characteristics. Some or all of the prospective deal's characteristics are mapped to yield components derived from completed deals that have been offered to a particular population. An expected-yield score for a prospective deal with regard to that population is obtained by combining the mapped yield components.

Abstract: Methods, apparatus, systems to determine a conditional probability based on audience member probability distributions for media audience measurement are disclosed. Disclosed example methods for media audience measurement include determining a first audience probability distribution for a first member of a household and determining a second audience probability distribution for a second member of the household. Disclosed example methods also include calculating probabilities for audience combinations of the first member and the second member of the household based on the first audience probability distribution and the second audience probability distribution. Disclosed example methods further include determining a household audience characteristic probability based on the calculated probabilities of the audience combinations of the household. The household audience characteristic indicates likelihoods of different possible audience compositions of the household for a media event.

Abstract: An electronic apparatus performs a method of generating a three-dimensional (3D) head deformation model that includes: receiving a two-dimensional (2D) facial image; identifying a first set of keypoints in the 2D facial image based on some artificial intelligence (AI) models; mapping the first set of keypoints to a second set of keypoints based on a set of user-provided keypoint annotations located on a plurality of vertices of a mesh of a 3D head template model; performing deformation to the mesh of the 3D head template model to obtain a deformed 3D head mesh model by reducing the differences between the first set of keypoints and the second set of keypoints; and applying a blendshape method to the deformed 3D head mesh model to obtain a personalized head model according to the 2D facial image.

Abstract: Methods, systems, and computer programs are presented for presenting return-on-investment (ROI) information, for budgeted services that resulted in a successful service delivery, on a user interface for setting the budget for a service request. One method includes an operation for identifying daily budgets for budgeted services that resulted in a successful service delivery (BSSSD). Each daily budget indicates an amount for spending in promotion of the BSSSD in an online service. The method further includes receiving a request, in a graphical user interface (GUI) of the online service, for posting a daily budget for a first budgeted service. Further, a performance value, associated with the daily budgets of the BSSSD that are similar to the first budgeted service, is selected. Further, the method includes causing presentation, by the one or more processors, of the performance value in the GUI.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to estimate demographics of users employing social media. An example method disclosed herein includes (1) identifying a social media message regarding an asset, the social media message associated with a user identifier associated with the user, (2) determining demographics associated with a group of people exposed to the asset, (3) associating the user identifier with the asset, and (4) repeating (1) to (3). The example method also includes (5) combining demographics associated with two or more different groups of people with which the user identifier is associated to estimate a demographic profile for the user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for managing a relationship between content and an environment for provisioning the content. In one aspect, a method includes receiving a request for a content item; and in response to receiving the request: selecting a creative from a plurality of creatives, the creative including a reference to a profile associated with one or more elements; retrieving content data from one or more content feeds bound to the elements; and delivering the creative and the content data to a user device.

Abstract: The timing for displaying an overlay a video is optimized using a system that includes an engagement component that receives information regarding frames of a video at which respective users of a plurality of users stop engaging with the video and an analysis component that calculates user engagement scores for the frames based on the information, wherein the user engagement scores reflect a percentage of the plurality of users that stop engaging with the video at the respective frames, and wherein the higher the percentage the lower the user engagement score. The system further includes an identification component that identifies a subset of the frames associated with relatively lower user engagement scores in comparison to other frames of the video and an overlay component that configures an overlay for displaying over the video near at least one of the frames included in the subset.

Abstract: A method for image processing includes: receiving a third two-dimensional image and a depth image corresponding to the third two-dimensional image, wherein the third two-dimensional image and the depth image include a face; establishing a three-dimensional model of the face according to the depth image; rotating the three-dimensional model of the face by a first angle; projecting the three-dimensional model of the face rotated by the first angle to an image coordinate system of the third two-dimensional image; and building a three-dimensional model of a background region of the third two-dimensional image, processing a background region of an image projected to the image coordinate system of the third two-dimensional image to obtain a fourth image.

Abstract: A file transfer display control method and apparatus, and a corresponding terminal are provided. In the method the terminal may generate a floating window to display transfer progress information during a file transfer executed by a content aggregation application in the terminal, where the transfer progress information indicates a progress of the file transfer. The terminal acquires position information of the floating window in a display interface of the content aggregation application. The terminal display the transfer progress information in the floating window based on the position information in the display interface.

Abstract: A method and system provides a data management system that provides data management services and products to users. The method and system provides a predictive model that generates probability scores indicating the likelihood that current users of the data management system would select promotional messages if the promotional messages are presented to the current users.

Abstract: An example system includes a memory store of aggregate definitions. Each aggregate definition specifies a key value, an output store, a feature, a half-life value, and an aggregate operation metric to apply to a cross of the feature and the half-life value to generate aggregate metrics. The system also includes an aggregation engine that generates aggregate feature records from the input source based on the aggregate definitions and stores the aggregate feature records in the output store. An aggregate feature record includes an aggregate of the metric for the feature decayed over time using the half-life. The system also includes a query service that identifies, using the aggregate definitions, responsive aggregate feature records that satisfy parameters of a received request, applies the half-life to the responsive feature records, and provides the responsive feature records to a requester, the requester using the responsive feature records as input for a neural network.

Abstract: This document describes the collection, generation, distribution, and management of online web content. The devices, systems, and methods described herein can be used to collect and generate online web content and communications. Specifically, the disclosed devices and systems may be employed to produce one or more marketing and/or advertising campaigns, as well as for monitoring, managing, defining the efficiency, effectiveness, and workability of the campaign with respect to generating user engagement, thereby accurately determining the cost benefits of the campaign. The analytic results provided may then be used to guide the generation of original web content, such as for the purposes of enhancing customer or follower experience, driving business, and for driving advertising campaigns. Alternatively, web content that is in the public domain, and determined to perform well, can be reproduced, referenced, or otherwise referred to, in the context of promoting or presenting the user's web content.

Abstract: Methods, apparatus, systems to determine a conditional probability based on audience member probability distributions for media audience measurement are disclosed. Disclosed example methods for media audience measurement include determining a first audience probability distribution for a first member of a household and determining a second audience probability distribution for a second member of the household. Disclosed example methods also include calculating probabilities for audience combinations of the first member and the second member of the household based on the first audience probability distribution and the second audience probability distribution. Disclosed example methods further include determining a household audience characteristic probability based on the calculated probabilities of the audience combinations of the household. The household audience characteristic indicates likelihoods of different possible audience compositions of the household for a media event.

Abstract: Aspects of subject technology provide systems and methods for simultaneously spell-correcting and completing partial search queries being entered by a user on the user's electronic device. An apparatus such as a computing device may receive partial search queries from the user's electronic device as each character of the partial search query is entered by the user. The apparatus may utilize a machine-learning model to generate suggested queries that include spelling-corrected versions of the received partial query, query completion suggestions for the partial query, and/or spelling-corrected completion suggestions for the partial query.

Abstract: Methods, systems, and computer-readable storage media for computer-implemented simulation of a digital advertising system to generate data. Actions include executing a plurality of actors, each actor being provided as computer-executable code including functionality of a respective component of a digital advertising system, and including one or more configuration parameters, during a simulated real-time bidding (RTB) process, transmitting, by each actor, at least one message to another actor, each message representing a transaction between actors in the digital advertising system, and being at least partially based on the one or more configuration parameters of a respective actor, and logging data representative of transactions occurring during the simulated RTB process.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting an audience for content are disclosed. In one aspect, a method includes receiving, from a content item provider, a request to distribute content items to users that have been deemed interested in a particular entity. First users that have expressed an interest in the particular entity are identified. Expansion entities for the particular entity are identified in a knowledge graph. At least one of the expansion entities can be connected to the particular entity by a relationship path. Second users are identified. The second users are deemed interested in the particular entity based on the second users having expressed an interest in an expansion entity. The content items are provided to at least a portion of the first users and at least a portion of the second users.

Abstract: The disclosed systems and methods aid advertisers in improving the calculation of the return on investment (ROI) for online marketing. The systems and methods allow advertisers to calculate the ROI of the online marketing of one or a plurality of advertisements by tracking the user identifier of the user devices used to interact with online marketing materials (e.g., online advertisement) and then, once a purchase is made, reconciling which marketing materials the customer interacted leading up to the purchase.

Abstract: An example method involves a computing device displaying a first action-outcome node corresponding to a first marketing-campaign action, displaying a plurality of outcomes of the first marketing-campaign action, positioned around the first action-outcome node, wherein each outcome corresponds to a potential result of the first marketing-campaign action, displaying a second action-outcome node corresponding to a second marketing-campaign action, displaying a plurality of outcomes of the second marketing-campaign action, positioned around the second action-outcome node, wherein each outcome corresponds to a potential result of the second marketing-campaign action, and displaying a graphical link connecting an outcome of the first action-outcome node to the second action-outcome node.

Abstract: Methods, systems and computer program products for submitting a network proxy bid and replacing the network proxy bid with an actual bid associated with advertisement placement are described. In some implementations, instead of submitting a static bid, an advertiser can submit a network proxy bid to an advertising exchange. A network proxy bid can indicate to the advertising exchange that the advertiser wishes to receive a call back from the advertising exchange when an impression matching targeting criteria specified by the advertiser becomes available. The advertiser then can calculate and submit an actual static bid to replace the network proxy bid in response to the call back.

Abstract: A system for advertising is disclosed, the system comprising a communication devices; an advertising mediums for displaying an advertisement; a vendors; an advertising agency; and a computer-readable medium coupled to the communication device having instructions stored thereon which when executed by the communication device, cause the communication device to perform operations.

Abstract: Methods and systems are described for providing programmatic creation and modification of advertising services including advertising campaigns. In one embodiment, a system includes an advertising campaign framework having one or more application programming interfaces (APIs) and software components that include one or more software programs and software code libraries. A storage medium stores instructions of the one or more APIs and software components. Processing logic is coupled to the storage medium. The processing logic is configured to execute the instructions of the one or more APIs and software components to receive a function call from an ad campaign software of an advertising entity, to provide a set of functionality for programmatically creating an ad campaign and programmatically modifying an ad campaign, and to generate a response that is sent to the ad campaign software of the advertising entity.

Abstract: Methods, apparatus, systems to determine a conditional probability based on audience member probability distributions for media audience measurement are disclosed. Disclosed example methods for media audience measurement include determining a first audience probability distribution for a first member of a household and determining a second audience probability distribution for a second member of the household. Disclosed example methods also include calculating probabilities for audience combinations of the first member and the second member of the household based on the first audience probability distribution and the second audience probability distribution. Disclosed example methods further include determining a household audience characteristic probability based on the calculated probabilities of the audience combinations of the household. The household audience characteristic indicates likelihoods of different possible audience compositions of the household for a media event.

Abstract: Systems, methods, and computer-readable media for controlling the timing of customer interactions (e.g., targeted product or service offerings to customers) in order to determine optimally efficient usage of a seller's available inventory are disclosed. A customer selection problem (CSP) may be formulated and solved to implement a customer selection policy for controlling the timing with which customers from a customer pool will be targeted with offerings. The customer selection policy may include dynamic optimization to account for intertemporal effects of customer decisions.

Abstract: A system and method includes an automatic content recognition (ACR) headend to perform ACR recognition on a video stream that is also being streamed to a smart TV, to detect timing of an upcoming ad spot to be delivered within the video stream. An addressable TV (ATV) engine can be in communication with the ACR head end and to: determine, from data of an ad campaign, whether a first ad to be delivered to the ad spot is off target for the ad spot; notify the ad replacer client to replace the first ad with an addressable ad when the first ad is off target for the ad spot; and provide, to the ad replacer client, criteria with which to constrain an ad request to obtain a second ad that is on target for the ad spot.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating and providing stacked attribution distributions within a stacked attribution user interface. For example, the disclosed systems can utilize attribution models to generate stacked attribution distributions as breakdowns of other attribution distributions. The disclosed systems can further provide a stacked attribution user interface that includes selectable elements for identifying event categories, dimensions, and attribution models for generating stacked attribution distributions. Based on user interaction with these selectable elements, the disclosed systems can dynamically generate, provide, and modify attribution breakdowns via the stacked attribution interface.

Abstract: Systems and methods of dynamically varying the intensity of providing content items in a remarketing campaign based on tracking client device interactions are provided. The system can assign an account identifier to a first segment for a pre-conversion model, responsive to receiving a first interaction associated with a content provider from a client device. The system can assign the account identifier to a second segment for the pre-conversion model, responsive to receiving a second interaction. The system can assign the account identifier to a third segment, responsive to receiving a third interaction. The third interaction can include a conversion event. The system can generate a post-conversion model based on the third segment and the pre-conversion model. The system can determine an intent index for the account identifier based on the post-conversion model. The system can store the account identifier into an interest cluster based on the intent index.