摘要:
Techniques are provided for ranking the entities that are identified in a document based on an estimated likelihood that a user will actually make use of the annotations. According to one disclosed approach, usage data that indicates how users interact with annotations contained in documents presented to the users is collected. Based on the usage data, weights are generated for features of a feature vector. The weights are then used to modify feature scores of entities, and the modified feature scores are used to determine how to annotate documents. Specifically, a set of entities are identified within a document. A ranking for the identified entities is determined based, at least in part, on (a) feature vector scores for each of the identified entities, and (b) the weights generated for the features of the feature vector. The document is then annotated based, at least in part, on the ranking.
摘要:
Disclosed is a system and program for processing anchor text for information retrieval. A set of anchors that point to a target document is formed. Anchors with same anchor text are grouped together. Information is computed for each group. Context information is generated for the target document based on the computed information.
摘要:
Systems and methods allowing publishers to define relationships between their own content items (e.g., World Wide Web pages) and content items created or maintained by other publishers. Publishers insert referral labels into their content items to indicate a referral relationship between or among publishers. In one embodiment, advertisements to be displayed while the user is viewing a content item are selected based at least in part on matching a referral label included in the target content item to a referral label in the advertisement. The referral label can also be used to manage referral-based revenue sharing among publishers.
摘要:
Publishers define relationships between their own content items (e.g., World Wide Web pages) and content items created or maintained by other publishers. A central namespace manager provides grouping labels that publishers insert into their content items. Content items having matching grouping labels are identified as related. Different classes of grouping labels can be used to define different types of relationships, and multiple classes of grouping labels can co-exist. Grouping labels are made available to an information server such as a search server, which uses the labels in identifying content items related to a currently displayed content item.
摘要:
Sales associated with service providers is increased using an opportunistic approach based on an unexpected change in a schedule or services (time, location). Windows of opportunity (time, physical location) are automatically identified, and once such a window of opportunity is detected, the people affected directly from the change of a schedule are identified. This information is then used by services providers to act quickly, in order to enhance sales. Heuristics to determine a window of opportunity are based only on schedule changes (which are associated with a rule database), and without knowing who the affected people are or what their preferences (profile data) are.
摘要:
A system and method for communication of information that has particular significance to a specific location only to those individuals that are at or near that geo-spatial location. In particular, providing safety factor data and alerts is a preferred use of the invention. The invention allows information to be tailored to the individual at that location by selectively filtering what is communicated and what is not. The system includes a transmitting proximity event server, a GPS client wireless component that can be a personal wireless communication device (such as Palm Pilot, cellular digital phones, etc.) or personal computer configured for use within a global position satellite network.
摘要:
Context-specific transaction proposals are automatically generated and presented to a user who expresses interest in a particular topic. A user viewing a World Wide Web page or other content item activates an interface to indicate that he or she is interested in additional information related to the subject of the page. A context vector or other representation of the content of the page being viewed is transmitted to an information server, which identifies possible transactions related to the content and proposes one or more of these transactions to the user. Transaction proposals can be presented together with a contextual search interface that allows the user to submit zero or more search terms together with the context vector as a search query.
摘要:
A distributed comparison shopping system is based on a decentralized, distributed architecture utilizing a peer-to-peer network. The system creates an active marketplace with real-time price comparisons, with the peer-to-peer nodes coordinating connectivity with other peers and building a dynamic network. Each message includes a fixed component and an adaptive update component. The adaptive update component contains search criteria and search status fields that are selectively modified as the message travels through the peer-peer network. A node that receives a message will interpret the search criteria and apply those criteria to a local search result. The node then either forwards the unmodified or the modified message to other nodes in its neighborhood, or, alternatively, requests an authorization to modify the message prior to rebroadcasting, from the source node.
摘要:
The present invention provides for an integrated matching service and calendaring system. Calendar events are utilized as a bridge between an electronic calendaring system and a matching service. A calendar event represents an activity, e.g., job opening, tennis match, bicycle race, etc., the requirements to match the activity, the entity attributes, and any match results. An entity defines criteria and information for a matching activity which is stored as a calendar event in the electronic calendar system. Portions of the criteria and information are stored as attachments to the calendar event. The calendar events representing a matching activity and associated attachments are provided to a matching server which locates suitable matches for the activity based upon the criteria and information of the activity. If a suitable match is located, the matching server notifies the entities involved by listing the corresponding entities as attendees associated with the calendar event.
摘要:
Disclosed is a technique for building an index in which global analysis computations and index creation are pipelined, wherein the global analysis computations share intermediate results.