Abstract: For a multi-party advertising exchange including advertising and publishing entities from disparate advertising networks, flexible advertising transaction accounting methods are provided for circumstances where a relatively small number of trusted participants agree to reduce onerous reporting requirements conducted at a per transaction level. Trusted participants optionally report aggregate data for transactions including aggregate fraud data measurements to determine aggregate apportioning of cost and payments associated with advertising transactions. Aggregate reporting, subject to safeguards, enables a reduction in the overhead associated with per transaction accounting, and other participants need not duplicate measurements of trusted participants who are positioned best to collect the measurements, enabling a cooperative effort to the benefit of all of the trusted participants.

Abstract: For a multi-party advertising exchange including advertising entities and publishing entities from different advertising networks, automatic apportioning of advertising transactions across inventory from different advertising channels is provided. Optimal combinations or advertising bundles of different channels are generated for the participant based on an analysis of inventory across different advertising channels for a given set of participation goals/constraints. A small portion of a participant's overall advertising expenditure can be held back for exploratory purposes to determine better combinations of advertising channels based on actual performance over time.

Abstract: For a multi-party advertising exchange, including publishing entities and advertising entities from disparate advertising networks, which facilitates transactions for publishing inventory, a value add broker is provided to aggregate information from third parties having valuable information for input to the exchange or to perform services that are valuable to transactions in the exchange. The valuable information or services further facilitate the transactions for the publishing inventory automatically generating a benefit for the third parties providing the valuable information or services commensurate with the value added to the transactions.

Abstract: For a multi-party online advertising exchange including advertising and publishing entities and one or more third party participants, the disclosed systems and methods enable third party participation in arbitrage opportunities in online advertising transactions. A plurality of underlying transaction details are abstracted and provided to the third party participants without loss of generalization and while preserving relationships in the transaction data, to enable a third party share risk in advertising transactions. Various system refinements are provided and disclosed according to a host of optional embodiments.

Abstract: The subject disclosure pertains to systems and methods for facilitating generation of item recommendations based at least in part upon pair-wise lift. Pair-wise lift is a measure of correlation between a pair of items and is generally calculated based upon past usage data. If usage data is insufficient or unavailable, pair-wise lift for a pair of items can be estimated based upon metadata associated with the items. In other aspects, pair-wise lift can be used to generate an explanation for recommended items. An explanation for an item recommendation can be based upon common metadata features associated with the item pair. The relative impact each metadata feature has on predicted pair-wise lift can be evaluated to determine the common feature(s) most likely to have caused the item to be recommended.

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.