Abstract: An improved system and method for predicting user navigation within sponsored search advertisements is provided. A list of sponsored advertisements for display on a web page of search results may be received. A click prediction classifier may be applied to predict a click probability of each sponsored advertisement and a dwell time prediction classifier may be applied to predict a dwell time probability on web pages of a website of each sponsored advertisement. A probability of user navigation may be predicted for each sponsored advertisement using a probability of a click on each sponsored advertisement and a probability of a dwell time on web pages of a website of each sponsored advertisement. The list of the sponsored advertisements may be ranked in part by the probability of user navigation and served to a web browser executing on a client device for display on a web page of search results.

Abstract: An improved system and method for aggregating a list of top ranked objects from ranked combination lists using an early termination algorithm is provided. Ranked lists of individual object attributes may be aggregated into ranked lists of combination object attributes. The ranked lists of object attributes, including ranked lists of individual object attributes as well as ranked lists of combination object attributes, may be scanned in parallel. A fixed number of top scoring objects may be stored in a results list of top ranked objects. An upper bound of best possible aggregation scores of unseen object in the ranked lists of object attributes may be computed to incorporate the extra information given by the combination lists of attributes. If the upper bound computed is less than the score of top scoring objects in the results list, then the top scoring objects in the results list may be output.

Abstract: A system and method for increasing the speed of flow of a quantity of data or of a plurality of data types through a communication system. The communication system comprises a plurality of switches including a source and a sink, each of the switches connected to a neighboring switch by a communication link having a capacity, each of the links having a pair of queue buffers, and one queue buffer of each pair of queue buffers is located at each switch connected by a link. For a single data type, a flow is computed for each link as a function of a last previous amount of flow across the same link. In the multi-data type case, a weighted queue difference is computed as a function of a last previous amount of flow between the same pair of queues and an amount of flow fi of said data type i is routed across each said link such that &Sgr;1≦i≦Kfi (&Dgr;′i(e)−fi) is maximized.

Abstract: I/O-efficient methods and apparatus are provided for the d-dimensional join problem in one, two, and three dimensions, and are also generalized for arbitrary higher dimensions. Let N be the total number of rectangles in the two sets to be joined, M the total amount of memory available, B the disk block size, and T the total number of pairs in the output of the join. Define n=N/B, m=M/B, and t=T/B. For one and two dimensions, I/O-optimal join methods are provided that run in O(nlogmn+t) I/O operations and have utility to temporal and spatial database systems. For dimensions d≧3, methods are provided that run in O(nlogm(d−1) n+t) I/O operations, which is within a logm(d−2)n factor of the currently known lower bounds.

Abstract: A method of using an algorithm to partition a multi-dimensional data set into a minimum number of rectangular-shaped partitions that can be processed more quickly than the non-partitioned data set while satisfying certain specified performance constraints.

Abstract: A novel and unique method, apparatus and programmed storage medium for approximating large data distributions of a database in order to allow a user to accurately analyze the entire data distribution using a limited amount of memory space in a reasonable amount of time. The approximation is based on partitioning the data domain into a number of regions, approximating each region using any well-known technique, and composing the errors of approximation to optimize suitable criteria for approximating the entire data distribution.

Abstract: A computer method, apparatus and programmed medium for optimizing the number of buckets, and thus minimizing the necessary amount of memory space, needed to construct a histogram of a data distribution contained within a computer database with a cumulative error bounded by a specified threshold. The method according to the present invention allows a user to determine the near-minimal memory space necessary to store an approximation of a database with a maximum error measure at most three times that specified by the user to allow the user to maximize computer resources.