A task-based advertisement system and method are provided. The system employs high-order concepts (e.g., booking a flight, checking stock quotes etc.) embodied in “task(s)” which can then be bid upon by advertisers. The task(s) employed by the system are based upon a semantic solution
A task-based advertisement system and method are provided. The system employs high-order concepts (e.g., booking a flight, checking stock quotes etc.) embodied in “task(s)” which can then be bid upon by advertisers. The task(s) employed by the system are based upon a semantic solution to a natural-language query. The system includes a search engine that is capable of serving content in response to user query(ies). The system further includes a task server that can include hardware and/or software to retrieve task(s) in response to user query(ies). The task(s) retrieved by the task server can be presented to advertiser(s) who can bid on the task(s).
대표청구항▼
What is claimed is: 1. A task-based advertisement system comprising a computer that comprises a processor that executes computer-executable instructions that cause the computer to implement a task server that: selects, based upon the tokens in a query from a user, a task from among a plurality of t
What is claimed is: 1. A task-based advertisement system comprising a computer that comprises a processor that executes computer-executable instructions that cause the computer to implement a task server that: selects, based upon the tokens in a query from a user, a task from among a plurality of tasks, each task in the plurality of tasks being an action relevant to users of a search engine, each task in the plurality of tasks specifying multiple slots that define parameters of the task into which tokens in queries can be filled; after selecting the task, generates a semantic solution that represents a mapping between the slots of the task and tokens in the query that fill the slots of the task, wherein the task server generates the semantic solution at least in part by: identifying an annotation token among the tokens in the query, the annotation token being a token that indicates a significance of a subject token, the subject token being another one of the tokens in the query; and determining, based at least in part on the significance of the subject token, whether the subject token should be mapped in a given one of the slots of the task; and after generating the semantic solution, provides data representing the task, along with the semantic solution, to search engine, wherein the search engine provides the task to an advertiser computer that determines whether to make a bid on the task based at least in part on the semantic solution. 2. The task-based advertisement system of claim 1, wherein the task server selects a selected group of tasks from among the plurality of tasks based upon the tokens in the query, wherein each task in the selected group of tasks is respectively associated with one or more keywords, wherein each of the keywords associated with the tasks in the selected group of tasks are weighted, and wherein the task server ranks the tasks in the selected group of tasks based on the weights of the keywords associated with the tasks in the selected group of tasks, the task selected by the task server being the task in the selected group of tasks having the highest rank. 3. The task-based advertisement system of claim 1, wherein the query is a natural language query. 4. The task-based advertisement system of claim 1, wherein the search engine further provides, in response to the query, a targeted advertisement to the user based in part on information received from the advertiser. 5. The task-based advertisement system of claim 1, wherein the bid represents a monetary amount to be paid in the event the search engine provides an advertisement associated with the advertiser to the user. 6. The task-based advertisement system of claim 1, wherein the task comprises a name, a title, a description, and a keyword. 7. The task-based advertisement system of claim 5, wherein the task server selects the task using a query classifier model; and wherein the task server employs click-through information to update the query classifier model, the click-through information regarding a selection, by the user, of the advertisement. 8. The task-based advertisement system of claim 1, wherein the task server generates the semantic solution using a slot-filling model; and wherein the task server updates the slot-filling model based on information regarding a user web action. 9. The task-based advertisement system of claim 1, wherein the advertiser provides advertisement information to the search engine, the advertisement information regarding an advertisement that is to be provided to the user in the event that the advertiser is a successful bidder on the task. 10. The task-based advertisement system of claim 9, the advertisement information comprising an identifier employed by the search engine to locate the advertisement, the advertisement accessible by the search engine. 11. The task-based advertisement system of claim 9, the advertiser dynamically providing the advertisement information to the search engine. 12. The task-based advertisement system of claim 1, the task comprising an XML packet which is sent to the advertiser by the search engine. 13. The task-based advertisement system of claim 1, the bid comprising an HTML data packet. 14. A search engine method comprising: receiving, by a computer, a user query from a user of a search engine; providing, by the computer, a task query and contextual information to a task server, the task query and the contextual information based at least in part on the user query; receiving, by the computer, data representing a task from the task server, the task being an action relevant to the user of the search engine, the task including a semantic solution that comprises mappings between slots of the task and tokens in the user query the tokens including at least an annotation token and a subject token, the slots including a given slot, the semantic solution comprising a mapping between the subject token and the given slot when the annotation token indicates that the subject token is likely relevant to the given slot, each of the slots defining a parameter of the task into which one or more tokens in user queries can be filled; providing, by the computer, the task to advertisers; receiving, by the computer, bids on the task from at least some of the advertisers; and providing, by the computer, a page to the user as a response to the user query, the page containing search results and an advertisement, the search results responsive to the user query, the advertisement associated with a successful one of the bids. 15. The method of claim 14, further comprising: after receiving the bids, retrieving, by the computer, information regarding the advertisement from one of the advertisers. 16. The method of claim 14, further comprising: obtaining, by the computer, click-through information from the user; logging, by the computer, the click-through information; and providing, by the computer, the click-through information to the task server, the task server using the click-through information to select tasks in response to receiving task queries. 17. The method of claim 14, further comprising: receiving, by the computer, information regarding a user web action; logging, by the computer, the user web action; and providing, by the computer, information regarding the user web action to the task server, the task server using the information regarding the user web action to generate one or more potential mappings between tokens in queries and slots of tasks. 18. An advertiser method comprising: receiving, by a computer, an XML packet from a search engine, the XML packet representing a task, the task being an action relevant to a user who submitted a user query to the search engine, the XML packet comprising a title of the task and a description of the task, the task including a semantic solution that represents a mapping between slots of the task and tokens in the user query that fill the slots of the task, each of the slots defining a parameter of the task into which one or more tokens in user queries can be filled, each of the slots having a slot name that identifies the slot and a slot type that identifies a type of value in the slot; based at least in part on the semantic solution, determining, by the computer, whether to make a bid on the task; when a determination is made to make the bid on the task, providing, by the computer, bid information to the search engine, the bid information indicating the bid; dynamically generating, by the computer, an HTML data packet based at least in part on a given token, the HTML data packet representing an advertisement to be displayed to the user, the given token being one of the tokens mapped to the slots of the task; embedding, by the computer, a parameter value in a link to a site, the link contained in the HTML data packet, the parameter value corresponding to the given token; providing, by the computer, the HTML data packet to the search engine; receiving, by the computer, a request for the site from the user, the request including the parameter value; in response to receiving the request for the site: automatically populating, by the computer, a field in the site with the parameter value; and after populating the field, sending, by the computer, the site to the user. 19. The method of claim 18, further comprising: providing information associated with a user web action to the search engine.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (41)
Amirghodsi Siamak (Prairie View IL) Daneshbodi Farnoud (Prairie View IL), Adaptive natural language computer interface system.
Cameron Paul Scott ; Nash John Charles ; Bloomer Robert Christopher ; Wollan Robert Edward ; Kreutter Kelly Marie ; Ahler Olmstead Melinda Ann ; Renner Dale Harry ; Bourne Ryan Douglas ; Carnish Keit, Computerized source searching system and method for use in an order entry system.
Cha, Keon-Hoe; Chung, Eui-Sok; Lim, Soo-Jung; Kang, Hyun-Kyu, Information generation and retrieval method based on standardized format of sentence structure and semantic structure and system using the same.
Ashbey James A. (162 Trevelyah Road London SW17 9LW GBX), Interactive video system having frame recall dependent upon user input and current displayed image.
Brown Peter F. (New York NY) Cocke John (Bedford NY) Della Pietra Stephen A. (Pearl River NY) Della Pietra Vincent J. (Blauvelt NY) Jelinek Frederick (Briarcliff Manor NY) Lai Jennifer C. (Garrison N, Method and system for natural language translation.
Wang, Hai-Feng; Huang, Chang-Ning; Lee, Kai-Fu; Di, Shuo; Gao, Jianfeng; Cai, Dong-Feng; Chien, Lee-Feng, System and iterative method for lexicon, segmentation and language model joint optimization.
Corston, Simon H.; Dolan, William B.; Vanderwende, Lucy H.; Braden-Harder, Lisa, System for processing textual inputs using natural language processing techniques.
Arnold, James F.; Israel, David J.; Tyson, W. Mabry; Bear, John S.; Voss, Loren L., System, method and article of manufacture for concept based information searching.
Ginter Karl L. ; Shear Victor H. ; Sibert W. Olin ; Spahn Francis J. ; Van Wie David M., Systems and methods for secure transaction management and electronic rights protection.
Carpenter Claudia ; Hansen George A. ; Lam Beatrice M. Y. ; Lozares Lawrence ; Mako Krisztina ; Mernyk Paul A. ; Robnett Robert Chris, User interface with individually configurable panel interfaces for use in a computer system.
Macbeth, Steven William; O'Connor, Michael William; Jiang, Zhaowei Charlie; Van Snellenberg, Michael; Jiang, Li; Sadhukha, Shajib; Kumar, Amit, User task completion via open market of actions and/or providers.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.