METHODS AND ARCHITECTURE FOR PERFORMING CLIENT-SIDE DIRECTED MARKETING WITH CACHING AND LOCAL ANALYTICS FOR ENHANCED PRIVACY AND MINIMAL DISRUPTION
Methods and architectures are disclosed for performing directed marketing in client applications. Operating systems and applications such as computer games, word processors, etc., are used as vehicles for presentation of advertisements. Techniques are included that maximize the effectiveness of impressions while maintaining privacy and minimizing disruption by performing local analysis of content and behavior. Local analysis can consider useful details of personal content and activities, yet this information is kept private, on the user's machine. The information is used by local learning, reasoning, and matching methods to select impressions from spanning advertising content cached on the local machine. Signals about usage or activity can be returned with user confirmation and used to design future advertisement caches sent as updates.
1 . A computer-implemented system that facilitates client-side advertising in a client, the client comprising:
an advertisement component for receiving and processing advertisement content;
an application component for inserting the advertisement content into a client application for presentation to a user; and
a probabilistic learning and reasoning component for generating a model that applies a learning and reasoning process to client processes of the advertisement component and the application component.
2 . The system of claim 1 , wherein the application component facilitates client-side analysis that includes sensitive user and client information, and which sensitive user and client information is prohibited from being communicated externally from the client.
3 . The system of claim 1 , further comprising a context component that senses, collects and stores context information related to the client's geographic location and patterns of locations, and which facilitates insertion of the advertisement content at an appropriate time.
4 . The system of claim 1 , wherein the advertisement content inserted into the application by the application component is targeted to the user.
5 . The system of claim 1 , wherein the application component facilitates insertion of targeted advertising content into the client application that receives streaming content, the targeted advertising content timed for presentation during a commercial break of the streaming content.
6 . The system of claim 1 , wherein the application component facilitates external communication of client-side user and/or system information only after user confirmation that allows the communication.
7 . The system of claim 1 , wherein the probabilistic learning and reasoning component generates an enhancement model for processing information that enhances user interest in the advertisement content.
8 . The system of claim 1 , wherein the probabilistic learning and reasoning component generates an enhancement model for continually updating client-side advertisement content based on user privacy data.
9 . The system of claim 1 , wherein the probabilistic learning and reasoning component generates a timing model for processing and presenting the advertisement content into the client application at appropriate times.
10 . The system of claim 1 , wherein the probabilistic learning and reasoning component generates a timing model for managing caching operations of the advertisement content in the client.
11 . The system of claim 10 , wherein the timing model facilitates execution of client processes so as to minimize user disruption.
12 . The system of claim 10 , wherein the timing model facilitates execution of client processes so as to minimize client system disruption.
13 . The system of claim 1 , wherein the probabilistic learning and reasoning component generates a frustration model for analyzing and processing client-side data related to user frustration behavior.
14 . A computer-implemented method of managing client-side content processing and presentation, comprising:
receiving advertisement content at a client from a vendor site for presentation in a client application;
monitoring user and system activity data in response to presentation of the advertisement content in a launched client application;
developing a probabilistic model locally based on the user and system activity data; and
processing the model to effect user behavior and system processes.
15 . The method of claim 14 , further comprising modeling context information in the probabilistic model related to geographic location of the client.
16 . The method of claim 14 , further comprising pushing the advertisement content to subjects who share out personal data and activities to measure and model relationships between content, activities, interests, and behavioral responses, such as showing signs of disinterest or signs of interest such as dwells and clickthroughs on advertisements so as to better understand the expected behaviors of users who do not share private information.
17 . The method of claim 14 , further comprising modeling disruption data associated with disrupting user activity by monitoring user task completion and application transitioning.
18 . The method of claim 14 , further comprising updating the advertisement content in the client application based on a change in geographic location of the client.
19 . The method of claim 14 , further comprising updating cache content based on a change in geographic location of the client.
20 . A computer-executable system for inserting advertisements in a client application, comprising:
computer-implemented means for receiving and processing advertisement data for insertion into a client application;
computer-implemented means for selecting the advertisement data based on at least one of user state and user preferences;
computer-implemented means for personalizing the advertisement data;
computer-implemented means for inserting the advertisement data into the client application for presentation to a client user; and
computer-implemented means for probabilistically modeling user and system behavior based on perception of the personalized advertisement data by the client user in the client application.