Abstract: A service is described that makes it easy and cost-effective to configure network traffic flowing to multiple endpoints. Enterprises and developers can easily manage the way that end-users are routed to applications running in different geographic regions. Customers can route traffic based on multiple criteria in a customer policy. Customers can manage large-scale, multi-region applications, increase application availability, and minimize latency to end-users. The service can consider the state of end-users, such as their geography, their network, and their latency to different geographic regions. The service can also consider the state of the customer's applications, such as the load on the application, the application's health, the cost of using that application (e.g., resource costs), resource states (e.g., disk throughput), resource location, customer-provided endpoint cost parameters, etc.

Abstract: A system, method and computer-readable medium for request routing based on prioritization and load information are provided. A client request processing a resource identifier for requested content transmits a DNS query to a content delivery network service provider. Based on a content provider identifier included in the DNS query, a receiving DNS nameserver determines a selection from a hierarchical organization of POPs and the relative capacity for a selected POP to process a subsequent request for resources. Based on the processing of the prioritization information and load information, the receiving DNS nameserver resolves the DNS query.

Abstract: Systems and methods for managing requesting routing functionality associated with resource requests for one or more resources associated with a content provider are provided. The request routing functionality can correspond to the processing of domain name service (“DNS”) requests for resources by computing devices and the resolution of the DNS requests by the identification of a network address of a computing device that will provide the requested resources. Based on the processing resource requests initiated by a client computing device, a CDN service provider can obtain performance information from client computing devices and correlate the performance information into a prioritized list for classes of client computing devices. The prioritized list can be utilized in resolving subsequent DNS queries.

Abstract: Systems and methods for managing requesting routing functionality associated with resource requests for one or more resources associated with a content provider are provided. The request routing functionality can correspond to the processing of domain name service (“DNS”) requests for resources by computing devices and the resolution of the DNS requests by the identification of a network address of a computing device that will provide the requested resources. Based on the processing resource requests initiated by a client computing device, a CDN service provider can obtain performance information from client computing devices and correlate the performance information into a prioritized list for classes of client computing devices. The prioritized list can be utilized in resolving subsequent DNS queries.

Abstract: Systems, methods, and computer storage media having computer-executable instructions embodied thereon that utilize images to generate identifiers of entities and to determine various relationships between entities and web pages are described. A collection of class images that represent various people and non-people entities deemed relevant for tracking user behavior is generated. Given a class image that represents an entity, an image identifier comprising unique feature vectors for that class image is generated. Given an arbitrary web page, all the images embedded on the web page are analyzed. Each embedded image is compared to the image identifiers of the collection of class images to determine whether or not any entities of interest appear in the embedded images of the web page. If relevant entities do appear on the web page, various associations can be made between the entities found on the page, and between the uniform resource locator (URL) of the web page where the entities appear.

Abstract: Methods and computer-storage media having computer-executable instructions embodied thereon that facilitate classifying user intent with respect to an entity using intent-strength scores. A user query indicating a particular entity is received. The user's intent with respect to the particular entity is determined by assigning an intent-strength score to the user. The user's intent-strength score is determined using intent-strength scores assigned to seed patterns identified for entities in a category, as well as the received user query. In embodiments, a user's intent-strength score may be updated based on a subsequent query, or may be changed according to a function. A list of users having particular intent-strength scores for particular entities may be also be generated.

Abstract: Systems, methods, and computer storage media having computer-executable instructions embodied thereon that utilize images to generate identifiers of entities and to determine various relationships between entities and web pages are described. A collection of class images that represent various people and non-people entities deemed relevant for tracking user behavior is generated. Given a class image that represents an entity, an image identifier comprising unique feature vectors for that class image is generated. Given an arbitrary web page, all the images embedded on the web page are analyzed. Each embedded image is compared to the image identifiers of the collection of class images to determine whether or not any entities of interest appear in the embedded images of the web page. If relevant entities do appear on the web page, various associations can be made between the entities found on the page, and between the uniform resource locator (URL) of the web page where the entities appear.

Abstract: Systems, methods, and computer storage media having computer-executable instructions embodied thereon that facilitate advertisement selection using advertising units. An entity that is a sequence of two or more words is referenced. The entity includes substrings comprising a portion of the entity. Search data in association with the entity is compared to corresponding search data in association with the substrings of the entity. Based on the comparison, it is determined that the entity comprises an advertisable unit that functions as a unit for purposes of selecting an advertisement for display. The advertisable unit is used to select an advertisement to be presented to the user.

Abstract: Methods and computer-readable media are provided for defining user intent so that user intent can be determined and advertisements and other information can be provided to a user based on that user's intent. A topical expression is identified and is associated with attributes and actions. Actions indicate steps that can be performed to achieve a task associated with the topical expression. An intent structure is then generated. The intent structure illustrates a relationship between the identified topical expression and other topical expressions. User data may then be received and mapped to the intent structures to determine present and future user intent.

Abstract: Embodiments of the present invention relate to systems, methods, and computer-storage media for providing a method of presenting advertisements to a target group of potential consumers generated from a group of seed consumers. In one embodiment, a target list request is received from an advertiser. A set of seed consumers associated with the advertiser is then determined. A set of potential consumers is then determined based on the set of seed consumers. Further, a subset of the set of potential consumers is determined, each potential consumer of the subset having a common characteristic. A target list of potential consumers is then generated comprising at least a portion of the subset of potential consumers. The target list of potential consumers is then provided to the advertiser.

Abstract: Systems, methods, and computer storage media having computer-executable instructions embodied thereon that facilitate advertisement selection using advertising units. An entity that is a sequence of two or more words is referenced. The entity includes substrings comprising a portion of the entity. Search data in association with the entity is compared to corresponding search data in association with the substrings of the entity. Based on the comparison, it is determined that the entity comprises an advertisable unit that functions as a unit for purposes of selecting an advertisement for display. The advertisable unit is used to select an advertisement to be presented to the user.