Abstract: A method for text compression comprises recognizing a prefix string of one or more text characters preceding a target string of a plurality of text characters to be compressed. The prefix string is provided to a natural language generation (NLG) model configured to output one or more predicted continuations each having an associated rank. If the one or more predicted continuations include a matching predicted continuation relative to the next one or more text characters of the target string, the next one or more text characters are compressed as an NLG-type compressed representation. If no predicted continuations match the next one or more text characters of the target string, a longest matching entry in a compression dictionary is identified. The next one or more text characters of the target string are compressed as a dictionary-type compressed representation that includes the dictionary index value of the longest matching entry.

Abstract: A method for text compression comprises recognizing a prefix string of one or more text characters preceding a target string of a plurality of text characters to be compressed. The prefix string is provided to a natural language generation (NLG) model configured to output one or more predicted continuations each having an associated rank. If the one or more predicted continuations include a matching predicted continuation relative to the next one or more text characters of the target string, the next one or more text characters are compressed as an NLG-type compressed representation. If no predicted continuations match the next one or more text characters of the target string, a longest matching entry in a compression dictionary is identified. The next one or more text characters of the target string are compressed as a dictionary-type compressed representation that includes the dictionary index value of the longest matching entry.

Abstract: A method for text compression comprises recognizing a prefix string of one or more text characters preceding a target string of a plurality of text characters to be compressed. The prefix string is provided to a natural language generation (NLG) model configured to output one or more predicted continuations each having an associated rank. If the one or more predicted continuations include a matching predicted continuation relative to the next one or more text characters of the target string, the next one or more text characters are compressed as an NLG-type compressed representation. If no predicted continuations match the next one or more text characters of the target string, a longest matching entry in a compression dictionary is identified. The next one or more text characters of the target string are compressed as a dictionary-type compressed representation that includes the dictionary index value of the longest matching entry.

Abstract: A personalized content recommendation provisioning method and system are described, according to various implementations. A token value allocation engine optimizes the allocation of content consumption revenue to publisher systems and recommendation source systems (e.g., the systems that provide the recommended content for presentation with the publisher system's native content) based on a model calculating and applying a life time or long term value (LTV) metric. The LTV metric for revenue is based on a detection of a how often a user system engages with content derived from or associated with a publisher system.

Abstract: Method, system, and programs for estimating interests of a plurality of users with respect to a new piece of information are disclosed. In one example, historical interests of the plurality of users are obtained with respect to one or more existing pieces of information. One or more users are selected from the plurality of users. Historical interests of the one or more users can minimize an objective function over the plurality of users. Interests of the one or more users are obtained with respect to the new piece of information. Estimated interests of the plurality of users are generated with respect to the new piece of information based on the obtained interests of the one or more users.

Abstract: Disclosure includes a system, method and architecture to monitor media content streams, e.g., streams of digital encoded media data such as without limitation audio and/or video data, from a number of media content streaming providers and identify media content items from those media content items being streamed by the media content streaming providers for inclusion in a personalized media content stream for a user. In a direct streaming approach, the media content items contained in a stream are available while each item is being streamed by the content provider. In a centralized streaming approach, media content items contained in each media content stream that is currently being streamed by content provider are available for a limited window of time, which can be dependent on buffer length.

Abstract: A personalized content recommendation provisioning method and system are described, according to various implementations. A token value allocation engine optimizes the allocation of content consumption revenue to publisher systems and recommendation source systems (e.g., the systems that provide the recommended content for presentation with the publisher system's native content) based on a model calculating and applying a life time or long term value (LTV) metric. The LTV metric for revenue is based on a detection of a how often a user system engages with content derived from or associated with a publisher system.

Abstract: Disclosed is a method and system for situational substitutions in fantasy sports. A server maintains personnel packages received from a client device operated by a user owning a fantasy team that includes a default personnel package and zero or more situational personnel packages of fantasy players. Each situational personnel package includes fantasy players and is associated with a specific situation that might occur within a real life sports game. The server detects a specific real life sports game situation matching the specific situation associated with a situational personnel package. Based on this detection, the server substitutes the fantasy player(s) in the situational personnel package for one or more of the fantasy players in the default package. The server determines a score for the fantasy team based on points accrued by fantasy players in the default personnel package and the substituted one or more fantasy players in the personnel package.

Abstract: A method for estimating model parameters. The method comprises receiving a data set related to a plurality of users and associated content, partitioning the data set into a plurality of sub data sets in accordance with the users so that data associated with each user are not partitioned into more than one sub data set, storing each of the sub data sets in a separate one of a plurality of user data storages, each of said data storages being coupled with a separate one of a plurality of estimators, storing content associated with the plurality of users in a content storage, where the content storage is coupled to the plurality of estimators so that the content in the content storage is shared by the estimators, and estimating, asynchronously by each estimator, one or more parameters associated with a model based on data from one of the sub data sets.

Abstract: A method for adjusting one or more parameters associated with a model. The method comprises obtaining, from a first source, first information related to activity of a user. The method further comprises adjusting one or more parameters associated with a model based on the first information collected within a first length of time, and obtaining, from a second source, second information related to activity of the user. The method further comprises adjusting the one or more parameters associated with the model based on the second information collected within a second length of time and a measure indicative of performance of the model, wherein the second length of time is larger than the first length of time.

Abstract: Method, system, and programs for estimating interests of a plurality of users with respect to a new piece of information are disclosed. In one example, historical interests of the plurality of users are obtained with respect to one or more existing pieces of information. One or more users are selected from the plurality of users. Historical interests of the one or more users can minimize an objective function over the plurality of users. Interests of the one or more users are obtained with respect to the new piece of information. Estimated interests of the plurality of users are generated with respect to the new piece of information based on the obtained interests of the one or more users.

Abstract: A system and method for aggregating emotions of users for a media program. A server stores reference audio signal fingerprints, each associated with a reference audio signal of the program. For each user, the server computer: receives a first audio signal fingerprint from a client device operated by a user, the first audio signal fingerprint associated with a first audio signal comprising ambient sound associated with the user and an audio signal of the program; searches the stored reference audio signal fingerprints to determine one that is related to the first audio signal fingerprint; determines an ambient sound signal by obtaining a difference between the stored reference audio signal fingerprint and the first audio signal fingerprint; and determines, using the ambient sound signal, an emotion of the user for a program segment. The server computer aggregates the emotions to determine a representative emotion of the users for the segment.

Abstract: The present invention is directed towards systems and methods for providing a plurality of facets in response to a search query. The method according to one embodiment of the present invention comprises receiving a search query from a user and retrieving a search result set. The method then identifies an optimum number of facets and selects a plurality of facets with a maximum incremental value. Finally, the method provides the search result set and the selected facets to a user.

Abstract: One embodiment accesses two or more photos taken by one or more travelers at one or more destinations and one or more points-of-interest located within the destinations; constructs one or more photo streams for each unique traveler-destination combination, wherein each one of the photo streams comprises two or more of the photos taken by the corresponding traveler at the corresponding destination; maps each one of the photos to one of the points-of-interest; constructs one or more timed paths for each unique traveler-destination combination based on the photo streams and the mapping between the photos and the points-of-interest, wherein each one of the timed paths comprises one or more of the points-of-interest located within the corresponding destination and visited by the corresponding travel; and constructs an itinerary based on a start point-of-interest, an end point-of-interest, a time constraint, and the timed paths.

Abstract: A method and system for providing targeted applications within an electronic message interface. The method comprises retrieving content from a user's electronic message in the electronic message interface, parsing and interpreting the user's context from the content, and accepting bidding on the user's interpreted context from one or more applications, the one or more applications including triggering conditions related to the user's interpreted context. The one or more applications are ranked based on the bidding. The method further comprises loading the one or more applications into the electronic message interface based on the ranking, and providing one or more follow-up transactions associated with the one or more loaded applications based on the user's interpreted context.

Abstract: The present invention provides a method and system for providing targeted applications within a search engine results page. The method and system includes receiving a search query from a user and interpreting the search query. The method and system then first maps the interpreted query to one or more action templates, wherein mapping the interpreted query to one or more action templates comprises selecting one or more actions associated with the interpreted query. The method and system then maps the selected one or more actions associated with the interpreted query to a plurality of applications and selecting one or more applications associated with the one or more actions. Finally, the method and system displays the one or more applications within a search results page.

Abstract: A crowdsourcing approach is used to collect from contributors, e.g., a large group of consumers, item pricing offered by the sellers from which the contributors have purchased the items. A contributor may provide item pricing information associated with a given store by uploading a receipt from the store, which receipt identifies the item(s) purchased by the contributor and a price for each item purchased, store information, e.g., store name, location, telephone number, etc. A database or other data store may be used to maintain contributor, store and item information. The stored information may identify which store is/are selling which item(s) and at what price(s). A shopping list may be generated, which includes information identifying one or more items and, for each item, the store at which the item(s) may be purchased at the lowest available price relative to other stores. By analyzing consumers' shopping habits, personalized target of ads and/or promotions can be achieved.

Abstract: The present invention provides a method and system for providing targeted applications within a search engine results page. The method and system includes receiving a search query from a user and interpreting the search query. The method and system then first maps the interpreted query to one or more action templates, wherein mapping the interpreted query to one or more action templates comprises selecting one or more actions associated with the interpreted query. The method and system then maps the selected one or more actions associated with the interpreted query to a plurality of applications and selecting one or more applications associated with the one or more actions. Finally, the method and system displays the one or more applications within a search results page.

Abstract: Disclosed is a method and system for situational substitutions in fantasy sports. A server maintains personnel packages received from a client device operated by a user owning a fantasy team that includes a default personnel package and zero or more situational personnel packages of fantasy players. Each situational personnel package includes fantasy players and is associated with a specific situation that might occur within a real life sports game. The server detects a specific real life sports game situation matching the specific situation associated with a situational personnel package. Based on this detection, the server substitutes the fantasy player(s) in the situational personnel package for one or more of the fantasy players in the default package. The server determines a score for the fantasy team based on points accrued by fantasy players in the default personnel package and the substituted one or more fantasy players in the personnel package.

Abstract: A method for estimating model parameters. The method comprises receiving a data set related to a plurality of users and associated content, partitioning the data set into a plurality of sub data sets in accordance with the users so that data associated with each user are not partitioned into more than one sub data set, storing each of the sub data sets in a separate one of a plurality of user data storages, each of said data storages being coupled with a separate one of a plurality of estimators, storing content associated with the plurality of users in a content storage, where the content storage is coupled to the plurality of estimators so that the content in the content storage is shared by the estimators, and estimating, asynchronously by each estimator, one or more parameters associated with a model based on data from one of the sub data sets.