Microsoft Technology Licensing, LLC

Staples, Inc.

Tchavdar Dangaltchev

Timothy Wee

Karthik Kumara

Courosh Mehanian

Jing Pan

Gang Wu

Technology for predicting online user shopping behavior, such as whether a user will purchase a product, is described. An example method includes receiving current session data describing a current session for a current user, extracting a current clickstream from the current session data classifying the current clickstream as a purchase clickstream or a non-purchase clickstream by processing the current clickstream using one or more sets of Hidden Markov Model parameters produced by one or more Hidden Markov Models, and computing, using the one or more computing devices, a purchase probability that the current user will purchase a product during the current session based on the classifying.

Clickstream purchase prediction using Hidden Markov Models

Technology for determining the order of the search results to maximize a financial goal is described. In an example embodiment, a method, implemented using the one or more computing devices, such as client and/or server devices, may receive a product search request from a user device associated with a user and retrieve a set of products from a product database based on the product search request. Based on a purchase probability and one or more of a margin and a price for that product, the method determines an expected financial gain for each of the products of the set and sorts the set of products into an ordered set of products having an order based on the expected financial gain associated with each of the products. The method may then provide the ordered set of products for display to the user on the user device.

Ordering search results to maximize financial gain

Computer-readable media and a computer system for correcting bid estimates that are calculated from stored data encompassing an incumbent customer's participation in a keyword auction are provided. Initially, input criteria is received, which includes customer-history data and a candidate position, within a ranking of incumbent customers competing to display an advertisement, that is attractive to the incumbent customer. A corrected rank model of the competing incumbent customers' ranking is generated, which effectively discounts the stored data related to the incumbent content provider. The corrected rank model is utilized to predict an adjusted average position of the incumbent content provider, within the ranking of the competing incumbent customers, without physically extracting stored data associated therewith. A corrected price model that ignores the influence of the incumbent customer's participation in the advertising auction is then constructed. This corrected price model and the candidate position facilitate predicting the corrected bid estimate.

Estimating advertising prices for an incumbent content provider

Methods, systems, and computer storage media having computer-executable instructions embodied thereon that, when executed, perform methods for determining bid amounts necessary to ensure advertisements or other content items will be presented in designated positions other than the most prominent (or first) position when bid information associated with a lesser included match type is not requested, are provided. Bids are received that are associated with a given keyword, a content item and a desired placement position. Based thereupon, a placement position in which the bid will cause associated content item is determined and compared to the desired placement position. If the determined placement position is not within a given error threshold of the desired placement position, the bid is iteratively adjusted in such a way that an adjusted bid amount is determined which will ensure placement in the desired position.

Growing community of inventors

Bellevue, WA, United States of America

Tchavdar Dangaltchev