Abstract: Computer-readable media for determining whether to accept a candidate order from a content provider, or advertiser, to display a particular number of advertisements within a specified time segment are provided. Initially, the content provider may include placement criteria that, among other things, identify a leaf node at which impressions of the advertisement are expected to be rendered. Generally, the leaf node refers to a location within a topic graph that describes inventory that is permissible to allocate to satisfy the candidate order. To perform the determination, the inventory of impressions available for accommodating the candidate order and a log of booked orders scheduled to be placed within the time segment are identified. Linear programs are then utilized to determine whether the estimated inventory that satisfies the placement criteria is available by predictively placing the booked orders at the estimated inventory. If estimated inventory remains available, the candidate order is accepted.

Abstract: A system and method of refining context-free grammars (CFGs). The method includes deriving back-off grammar (BOG) rules from an initially developed CFG and utilizing the initial CFG and the derived BOG rules to recognize user utterances. Based on a response of the initial CFG and the derived BOG rules to the user utterances, at least a portion of the derived BOG rules are utilized to modify the initial CFG and thereby produce a refined CFG. The above method can carried out iterativey, with each new iteration utilizing a refined CFG from preceding iterations.

Abstract: A method for generating a Bayesian network in a parallel manner is based on an initial model having a plurality of nodes. Each node corresponds to a variable of a data set and has a local distribution associated therewith. The method includes assigning a plurality of subsets of the nodes to a respective plurality of constructors. The plurality of constructors is operated in a parallel manner to identify edges to add between nodes in the initial model. The identified edges are added to the initial model to generate the Bayesian network. The edges indicate dependency between nodes connected by the edges.

Abstract: Systems, methods, and computer storage media having computer-executable instructions embodied thereon that facilitate evaluation of digital content preferences are provided. A user is presented with items of digital content and permitted to manipulate the arrangement of the digital content items in the context of a layout area. Based on the user's manipulation of the digital content items, a user preference regarding an arrangement of digital content, such as a location preference, a position preference, and/or a usage preference, is identified. In embodiments, such a user preference can be utilized to later display digital content to a user in accordance therewith.

Abstract: A “Music Mapper” automatically constructs a set coordinate vectors for use in inferring similarity between various pieces of music. In particular, given a music similarity graph expressed as links between various artists, albums, songs, etc., the Music Mapper applies a recursive embedding process to embed each of the graphs music entries into a multi-dimensional space. This recursive embedding process also embeds new music items added to the music similarity graph without reembedding existing entries so long a convergent embedding solution is achieved. Given this embedding, coordinate vectors are then computed for each of the embedded musical items. The similarity between any two musical items is then determined as either a function of the distance between the two corresponding vectors. In various embodiments, this similarity is then used in constructing music playlists given one or more random or user selected seed songs or in a statistical music clustering process.

Abstract: Computer-readable media for determining whether to accept a candidate order from a content provider, or advertiser, to display a particular number of advertisements within a specified time segment are provided. Initially, the content provider may include placement criteria that, among other things, identify a leaf node at which impressions of the advertisement are expected to be rendered. Generally, the leaf node refers to a location within a topic graph that describes inventory that is permissible to allocate to satisfy the candidate order. To perform the determination, the inventory of impressions available for accommodating the candidate order and a log of booked orders scheduled to be placed within the time segment are identified. Linear programs are then utilized to determine whether the estimated inventory that satisfies the placement criteria is available by predictively placing the booked orders at the estimated inventory. If estimated inventory remains available, the candidate order is accepted.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant specifies tax rate(s), such as import tax and export tax, that apply to at least one other entity in the exchange. Since tax rate(s) can be expressed in different transactional terms by different parties, each tax rate is reduced to a common tax rate expression within the exchange for comparison. Intelligent tax rate selection and support can be provided to dynamically set tax rates that achieve utilitarian goals for the individual participants taking into account the tax rates expressed by other participants and their respective advertising goals, and dynamically adjusting tax rates over time in response to condition changes. Various refinements are provided and disclosed according to a host of optional implementations.

Abstract: A method and system for generating a price landscape for an advertiser for bids placed by the advertiser for advertisement space is provided. A price landscape system generates a price landscape based on information provided by an advertisement placement service that may include overall price estimation data and advertiser-specific performance data. The price landscape system generates price landscape data for an advertiser that combines the overall price estimation data and the advertiser-specific performance data to provide a more accurate assessment of the advertiser's expected performance than can be determined from the overall price estimation data or the advertiser-specific performance data alone.

Abstract: A method for generating a Bayesian network in a parallel manner is based on an initial model having a plurality of nodes. Each node corresponds to a variable of a data set and has a local distribution associated therewith. The method includes assigning a plurality of subsets of the nodes to a respective plurality of constructors. The plurality of constructors is operated in a parallel manner to identify edges to add between nodes in the initial model. The identified edges are added to the initial model to generate the Bayesian network. The edges indicate dependency between nodes connected by the edges.

Abstract: A system and method of refining context-free grammars (CFGs). The method includes deriving back-off grammar (BOG) rules from an initially developed CFG and utilizing the initial CFG and the derived BOG rules to recognize user utterances. Based on a response of the initial CFG and the derived BOG rules to the user utterances, at least a portion of the derived BOG rules are utilized to modify the initial CFG and thereby produce a refined CFG. The above method can carried out iterativey, with each new iteration utilizing a refined CFG from preceding iterations.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each publishing entity can expose user attribute space(s) for its publishing inventory to the exchange, which are normalized to a common vocabulary within the exchange. Similarly, each advertising entity can specify one or more preferences for user attributes as requirements or definitions for advertising transactions in the exchange which are also normalized to the common vocabulary. The common vocabulary enables the comparison of a first set of user attributes to a second set of user attributes within the exchange. Supplemental user attribute data may also be received from audience data brokers, and the performance of advertising as a function of sets and subsets of user attributes over time may be recorded so that optimal display strategies can be determined.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant can specify one or more display properties as requirements or definitions for advertising transactions in the exchange. The disparate display properties of multiple participants are normalized within the advertising exchange to a common vocabulary by translating the display properties to a common set of representations within the exchange enabling the comparison of a first set of display properties to a second set of display properties. The invention also optionally records the performance of advertising as a function of display properties over time so that optimal display strategies can be determined by participants with access to the performance information. Various system refinements are provided and disclosed according to a host of optional embodiments.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant can specify one or more utility functions that are invertible with respect to a common measure, such as revenue. In one non-limiting embodiment, each utility function is invertible with respect to expected revenue per standard advertising unit, e.g., expected cost per impression. The disparate utility functions of multiple participants are also normalized within the advertising exchange by converting the utility functions to the common measure enabling the comparison or translation of a first set of utility functions to a second set of utility functions in quantifiable terms. Various system refinements are provided and disclosed according to a host of optional embodiments.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant specifies tax rate(s), such as import tax and export tax, that apply to at least one other entity in the exchange. Since tax rate(s) can be expressed in different transactional terms by different parties, each tax rate is reduced to a common tax rate expression within the exchange for comparison. Intelligent tax rate selection and support can be provided to dynamically set tax rates that achieve utilitarian goals for the individual participants taking into account the tax rates expressed by other participants and their respective advertising goals, and dynamically adjusting tax rates over time in response to condition changes. Various refinements are provided and disclosed according to a host of optional implementations.

Abstract: An optimization tool is provided for individual participants who participate in multiple online advertising environments wherein the tool interfaces with the multiple environments in an optimal manner on behalf of the individual participants. A participant using the tool can specify goals and/or constraints for participating in one or more of the networks, and then the tool automatically optimizes advertising expenditure for advertising transactions across the different networks. The tool also tracks the performance of the participant across the different networks and dynamically tunes the participant's advertising expenditure based on such performance as well as the changing conditions of the marketplace.

Abstract: For a multi-party advertising exchange including advertising and publishing entities from disparate advertising networks, a framework is provided for applying insurance terms to advertising transactions in the exchange, enabling risk sensitive participants to ease their participation in a potentially volatile online advertising marketplace. Participants can use the insurance mechanism of the advertising exchange to reduce variance in expected outcome for transactions in the advertising exchange.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant can specify one or more utility functions that are invertible with respect to a common measure, such as revenue. In one non-limiting embodiment, each utility function is invertible with respect to expected revenue per standard advertising unit, e.g., expected cost per impression. The disparate utility functions of multiple participants are also normalized within the advertising exchange by converting the utility functions to the common measure enabling the comparison or translation of a first set of utility functions to a second set of utility functions in quantifiable terms. Various system refinements are provided and disclosed according to a host of optional embodiments.

Abstract: For a multi-party advertising exchange including advertising and publishing entities from different advertising networks, the invention provides architectures for an online advertising marketplace that range from lightweight to heavyweight implementations. A lightweight client side implementation of an interface includes centralized processing and storage of federated advertising marketplace data by centralized servers or services. A heavyweight client side implementation of an interface for advertising entities includes providing a peer instance of a federated advertising exchange application or set of processes is provided to each advertising entity as an interface for advertising entities where processing and storage are performed locally to each peer instance. Distributed advertising data can be replicated or synchronized with other peer instances.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, automatic value sharing is provided for distributed users of the advertising exchange that provide valuable information to the exchange. Users provide information that has quantifiable value to the advertising exchange and/or its participants. In exchange for the information, predicated on any function or cost model for quantifying the value of the information, a reward or advantage, such as revenue, is given to the user by the advertising exchange and/or participants of the advertising exchange in order to incentivize users' valuable contributions. As a result, the advertising exchange becomes a more efficient marketplace, increasing predictability and the participant(s) benefit in correspondence with the value of any information received by those participant(s). Various refinements are provided and disclosed according to a host of optional embodiments.

Abstract: For a multi-party advertising exchange including advertising and publishing entities, each participant specifies tax rate(s), such as import tax and export tax, that apply to at least one other entity in the exchange. Since tax rate(s) can be expressed in different transactional terms by different parties, each tax rate is reduced to a common tax rate expression within the exchange for comparison. Intelligent tax rate selection and support can be provided to dynamically set tax rates that achieve utilitarian goals for the individual participants taking into account the tax rates expressed by other participants and their respective advertising goals, and dynamically adjusting tax rates over time in response to condition changes. Various refinements are provided and disclosed according to a host of optional implementations.