US20120117077A1 - Annotation Framework - Google Patents

Abstract

A fact repository contains facts having attributes and values and further having associated annotations, which are used, among other things, to vet facts in the repository and which can be returned in response to a query.

Description

RELATED APPLICATIONS
• This application is a divisional application of U.S. patent application Ser. No. 11/356,728, filed Feb. 17, 2006, entitled “Annotation Framework”, which is incorporated by reference herein in its entirety.
• This application is related to the following U.S. Applications, all of which are incorporated by reference herein in their entirety:
• • U.S. application Ser. No. 11/357,748, entitled “Support for Object Search”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/342,290, entitled “Data Object Visualization”, filed on Jan. 27, 2006;
  • U.S. application Ser. No. 11/342,293, entitled “Data Object Visualization Using Maps”, filed on Jan. 27, 2006;
  • U.S. application Ser. No. 11/356,679, entitled “Query Language”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/356,837, entitled “Automatic Object Reference Identification and Linking in a Browseable Fact Repository”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/356,851, entitled “Browseable Fact Repository”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/356,842, entitled “ID Persistence Through Normalization”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/341,069, entitled “Object Categorization for Information Extraction”, filed on Jan. 27, 2006;
  • U.S. application Ser. No. 11/356,838, entitled “Modular Architecture for Entity Normalization”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/356,765, entitled “Attribute Entropy as a Signal in Object Normalization”, filed Feb. 17, 2006;
  • U.S. application Ser. No. 11/341,907, entitled “Designating Data Objects for Analysis”, filed on Jan. 27, 2006; and
  • U.S. application Ser. No. 11/342,277, entitled “Data Object Visualization Using Graphs”, filed on Jan. 27, 2006.
BACKGROUND OF THE INVENTION
• 1. Field of the Invention
• This invention pertains in general to searching collections of data and, in particular, to ways of adding detecting information in such collections of data.
• 2. Description of the Related Art
• The World Wide Web and other information storage and retrieval systems contain a great deal of information. People have devised many ways of organizing and viewing this information.
• As the retrieval and storage of information on the Internet continues to evolve, information is being stored in many different formats besides web pages. Moreover, in a very large repository of data, it becomes difficult to search the data in real-time. What is needed are new and advanced ways of accessing large collections of data from diverse sources, such as the Internet.
SUMMARY OF THE INVENTION
• The described embodiments of the present invention provide a methodology and system for storing annotations of facts stored in a repository in association with objects. The fact repository includes a large collection of facts, each of which is associated with an object, such as a person, place, book, movie, country, or any other entity of interest. Each fact comprises an attribute, which is descriptive of the type of fact (e.g., “name,” or “population”) and a value for that attribute (e.g., “George Washington”, or “1,397,264,580”). A value can also contain any amount of text—from a single term or phrase to many paragraphs or pages—such as appropriate to describe the attribute. Each object will have a name fact that is the name of the object. The value can include one or more phrases that are themselves the names of other facts.
• In one embodiment, annotations are stored corresponding to some of the facts in the repository. Some annotations, for example, represent number values in the facts, while other annotation values represent dates, geographic locations, ISBN values, names, units, and so on.
• Annotations are passed to requesting objects in response to queries and are used to determine search results Annotations are also used to decide whether facts match and whether facts contain reasonable values.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 shows a system architecture, in accordance with an embodiment of the invention.
• FIGS. 2(a)-2(d) are block diagrams illustrating a data structure for facts within a repository ofFIG. 1 in accordance with preferred embodiments of the invention.
• FIG. 2(e) is a block diagram illustrating an alternate data structure for facts and objects in accordance with preferred embodiments of the invention.
• FIG. 2(f) shows a plurality of annotations associated with facts in a repository.
• FIG. 2(g) shows an example of two facts and an associated annotation of one of the facts.
• FIGS. 3(a)-3(c) are flow charts showing methods of creating and using annotations.
• FIG. 4 is a block diagram showing annotation creation.
• FIG. 5 is a flow chart showing annotation creation.
• FIG. 6 shows an example of an XML feed sent in response to a request from on object requester and including annotations.
• The figures depict a preferred embodiment of the present invention for purposes of illustration only. One skilled in the art will readily recognize from the following discussion that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the invention described herein.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
• FIG. 1 shows asystem architecture100 adapted to support one embodiment of the invention.FIG. 1 shows components used to add facts into, and retrieve facts from arepository115. Thesystem architecture100 includes anetwork104, through which any number ofdocument hosts102 communicate with adata processing system106, along with any number ofobject requesters152,154.
• Document hosts102 store documents and provide access to documents. A document is comprised of any machine-readable data including any combination of text, graphics, multimedia content, etc. A document may be encoded in a markup language, such as Hypertext Markup Language (HTML), i.e., a web page, in an interpreted language (e.g., JavaScript) or in any other computer readable or executable format. A document can include one or more hyperlinks to other documents. A typical document will include one or more facts within its content. A document stored in adocument host102 may be located and/or identified by a Uniform Resource Locator (URL), or Web address, or any other appropriate form of identification and/or location. Adocument host102 is implemented by a computer system, and typically includes a server adapted to communicate over thenetwork104 via networking protocols (e.g., TCP/IP), as well as application and presentation protocols (e.g., HTTP, HTML, SOAP, D-HTML, Java). The documents stored by ahost102 are typically held in a file directory, a database, or other data repository. Ahost102 can be implemented in any computing device (e.g., from a PDA or personal computer, a workstation, mini-computer, or mainframe, to a cluster or grid of computers), as well as in any processor architecture or operating system.
• FIG. 1 shows components used to manage facts in afact repository115.Data processing system106 includes one ormore importers108, one ormore janitors110, abuild engine112, aservice engine114, and a fact repository115 (also called simply a “repository”). Each of the foregoing are implemented, in one embodiment, as software modules (or programs) executed byprocessor116.Importers108 operate to process documents received from the document hosts, read the data content of documents, and extract facts (as operationally and programmatically defined within the data processing system106) from such documents. Theimporters108 also determine the subject or subjects with which the facts are associated, and extract such facts into individual items of data, for storage in thefact repository115. In one embodiment, there are different types ofimporters108 for different types of documents, for example, dependent on the format or document type.
• Janitors110 operate to process facts extracted byimporter108. This processing can include but is not limited to, data cleansing, object merging, and fact induction. In one embodiment, there are a number ofdifferent janitors110 that perform different types of data management operations on the facts. For example, onejanitor110 may traverse some set of facts in therepository115 to find duplicate facts (that is, facts that convey the same factual information) and merge them. Anotherjanitor110 may also normalize facts into standard formats. Anotherjanitor110 may also remove unwanted facts fromrepository115, such as facts related to pornographic content. Other types ofjanitors110 may be implemented, depending on the types of data management functions desired, such as translation, compression, spelling or grammar correction, and the like.
• Various janitors110 act on facts to normalize attribute names, and values and delete duplicate and near-duplicate facts so an object does not have redundant information. For example, we might find on one page that Britney Spears's birthday is “12/2/1981” while on another page that her date of birth is “Dec. 2, 1981.” Birthday and Date of Birth might both be rewritten as Birthdate by one janitor and then another janitor might notice that 12/2/1981 and Dec. 2, 1981 are different forms of the same date. It would choose the preferred form, remove the other fact and combine the source lists for the two facts. As a result when you look at the source pages for this fact, on some you'll find an exact match of the fact and on others text that is considered to be synonymous with the fact.
• Build engine112 builds and manages therepository115.Service engine114 is an interface for querying therepository115.Service engine114's main function is to process queries, score matching objects, and return them to the caller but it is also used byjanitor110.
• Repository115 stores factual information extracted from a plurality of documents that are located on document hosts102. A document from which a particular fact may be extracted is a source document (or “source”) of that particular fact. In other words, a source of a fact includes that fact (or a synonymous fact) within its contents.
• Repository115 contains one or more facts. In one embodiment, each fact is associated with exactly one object. One implementation for this association includes in each fact an object ID that uniquely identifies the object of the association. In this manner, any number of facts may be associated with an individual object, by including the object ID for that object in the facts. In one embodiment, objects themselves are not physically stored in therepository115, but rather are defined by the set or group of facts with the same associated object ID, as described below. Further details about facts inrepository115 are described below, in relation toFIGS. 2(a)-2(d).
• It should be appreciated that in practice at least some of the components of thedata processing system106 will be distributed over multiple computers, communicating over a network. For example,repository115 may be deployed over multiple servers. As another example, thejanitors110 may be located on any number of different computers. For convenience of explanation, however, the components of thedata processing system106 are discussed as though they were implemented on a single computer.
• In another embodiment, some or all of document hosts102 are located ondata processing system106 instead of being coupled todata processing system106 by a network. For example,importer108 may import facts from a database that is a part of or associated withdata processing system106.
• FIG. 1 also includes components to accessrepository115 on behalf of one ormore object requesters152,154. Object requesters are entities that request objects fromrepository115.Object requesters152,154 may be understood as clients of thesystem106, and can be implemented in any computer device or architecture. As shown inFIG. 1, afirst object requester152 is located remotely fromsystem106, while asecond object requester154 is located indata processing system106. For example, in a computer system hosting a blog, the blog may include a reference to an object whose facts are inrepository115. An object requester152, such as a browser displaying the blog will accessdata processing system106 so that the information of the facts associated with the object can be displayed as part of the blog web page. As a second example,janitor110 or other entity considered to be part ofdata processing system106 can function as object requester154, requesting the facts of objects fromrepository115.
• FIG. 1 shows thatdata processing system106 includes amemory107 and one ormore processors116.Memory107 includesimporters108,janitors110, buildengine112,service engine114, andrequester154, each of which are preferably implemented as instructions stored inmemory107 and executable byprocessor116.Memory107 also includesrepository115.Repository115 can be stored in a memory of one or more computer systems or in a type of memory such as a disk.FIG. 1 also includes a computerreadable medium118 containing, for example, at least one ofimporters108,janitors110, buildengine112,service engine114,requester154, and at least some portions ofrepository115.FIG. 1 also includes one or more input/output devices120 that allow data to be input and output to and fromdata processing system106. It will be understood thatdata processing system106 preferably also includes standard software components such as operating systems and the like and further preferably includes standard hardware components not shown in the figure for clarity of example.
• FIG. 2(a) shows an example format of a data structure for facts withinrepository115, according to some embodiments of the invention. As described above, therepository115 includesfacts204. Eachfact204 includes a unique identifier for that fact, such as afact ID210. Eachfact204 includes at least anattribute212 and avalue214. For example, a fact associated with an object representing George Washington may include an attribute of “date of birth” and a value of “Feb. 22, 1732.” In one embodiment, all facts are stored as alphanumeric characters since they are extracted from web pages. In another embodiment, facts also can store binary data values. Other embodiments, however, may store fact values as mixed types, or in encoded formats.
• As described above, each fact is associated with anobject ID209 that identifies the object that the fact describes. Thus, each fact that is associated with a same entity (such as George Washington), will have thesame object ID209. In one embodiment, objects are not stored as separate data entities in memory. In this embodiment, the facts associated with an object contain the same object ID, but no physical object exists. In another embodiment, objects are stored as data entities in memory, and include references (for example, pointers or IDs) to the facts associated with the object. The logical data structure of a fact can take various forms; in general, a fact is represented by a tuple that includes a fact ID, an attribute, a value, and an object ID. The storage implementation of a fact can be in any underlying physical data structure.
• FIG. 2(b) shows an example of facts having respective fact IDs of10,20, and30 inrepository115.Facts10 and20 are associated with an object identified by object ID “1.”Fact10 has an attribute of “Name” and a value of “China.”Fact20 has an attribute of “Category” and a value of “Country.” Thus, the object identified by object ID “1” has aname fact205 with a value of “China” and acategory fact206 with a value of “Country.”Fact30208 has an attribute of “Property” and a value of “Bill Clinton was the 42nd President of the United States from 1993 to 2001.” Thus, the object identified by object ID “2” has a property fact with a fact ID of30 and a value of “Bill Clinton was the 42nd President of the United States from 1993 to 2001.”In the illustrated embodiment, each fact has one attribute and one value. The number of facts associated with an object is not limited; thus while only two facts are shown for the “China” object, in practice there may be dozens, even hundreds of facts associated with a given object. Also, the value fields of a fact need not be limited in size or content. For example, a fact about the economy of “China” with an attribute of “Economy” would have a value including several paragraphs of text, numbers, perhaps even tables of figures. This content can be formatted, for example, in a markup language. For example, a fact having an attribute “original html” might have a value of the original html text taken from the source web page.
• Also, while the illustration ofFIG. 2(b) shows the explicit coding of object ID, fact ID, attribute, and value, in practice the content of the fact can be implicitly coded as well (e.g., the first field being the object ID, the second field being the fact ID, the third field being the attribute, and the fourth field being the value). Other fields include but are not limited to: the language used to state the fact (English, etc.), how important the fact is, the source of the fact, a confidence value for the fact, and so on.
• FIG. 2(c) shows an example object reference table210 that is used in some embodiments. Not all embodiments include an object reference table. The object reference table210 functions to efficiently maintain the associations between object IDs and fact IDs. In the absence of an object reference table210, it is also possible to find all facts for a given object ID by querying the repository to find all facts with a particular object ID. WhileFIGS. 2(b) and2(c) illustrate the object reference table210 with explicit coding of object and fact IDs, the table also may contain just the ID values themselves in column or pair-wise arrangements.
• FIG. 2(d) shows an example of a data structure for facts withinrepository115, according to some embodiments of the invention showing an extended format of facts. In this example, the fields include anobject reference link216 to another object. Theobject reference link216 can be an object ID of another object in therepository115, or a reference to the location (e.g., table row) for the object in the object reference table210. Theobject reference link216 allows facts to have as values other objects. For example, for an object “United States,” there may be a fact with the attribute of “president” and the value of “George W. Bush,” with “George W. Bush” being an object having its own facts inrepository115. In some embodiments, thevalue field214 stores the name of the linked object and thelink216 stores the object identifier of the linked object. Thus, this “president” fact would include thevalue214 of “George W. Bush”, and objectreference link216 that contains the object ID for the for “George W. Bush” object. In some other embodiments,facts204 do not include alink field216 because thevalue214 of afact204 may store a link to another object.
• Eachfact204 also may include one ormore metrics218. A metric provides an indication of the some quality of the fact. In some embodiments, the metrics include a confidence level and an importance level. The confidence level indicates the likelihood that the fact is correct. The importance level indicates the relevance of the fact to the object, compared to other facts for the same object. The importance level may optionally be viewed as a measure of how vital a fact is to an understanding of the entity or concept represented by the object.
• Eachfact204 includes a list of one ormore sources220 that include the fact and from which the fact was extracted. Each source may be identified by a Uniform Resource Locator (URL), or Web address, or any other appropriate form of identification and/or location, such as a unique document identifier.
• The facts illustrated inFIG. 2(d) include anagent field222 that identifies theimporter108 that extracted the fact. For example, theimporter108 may be a specialized importer that extracts facts from a specific source (e.g., the pages of a particular web site, or family of web sites) or type of source (e.g., web pages that present factual information in tabular form), or animporter108 that extracts facts from free text in documents throughout the Web, and so forth.
• Some embodiments include one or more specialized facts, such as aname fact207 and aproperty fact208. Aname fact207 is a fact that conveys a name for the entity or concept represented by the object ID. Aname fact207 includes anattribute224 of “name” and a value, which is the name of the object. For example, for an object representing the country Spain, a name fact would have the value “Spain.” Aname fact207, being a special instance of ageneral fact204, includes the same fields as anyother fact204; it has an attribute, a value, a fact ID, metrics, sources, etc. Theattribute224 of aname fact207 indicates that the fact is a name fact, and the value is the actual name. The name may be a string of characters. An object ID may have one or more associated name facts, as many entities or concepts can have more than one name. For example, an object ID representing Spain may have associated name facts conveying the country's common name “Spain” and the official name “Kingdom of Spain.” As another example, an object ID representing the U.S. Patent and Trademark Office may have associated name facts conveying the agency's acronyms “PTO” and “USPTO” as well as the official name “United States Patent and Trademark Office.” If an object does have more than one associated name fact, one of the name facts may be designated as a primary name and other name facts may be designated as secondary names, either implicitly or explicitly.
• Aproperty fact208 is a fact that conveys a statement about the entity or concept represented by the object ID. Property facts are generally used for summary information about an object. Aproperty fact208, being a special instance of ageneral fact204, also includes the same parameters (such as attribute, value, fact ID, etc.) asother facts204. Theattribute field226 of aproperty fact208 indicates that the fact is a property fact (e.g., attribute is “property”) and the value is a string of text that conveys the statement of interest. For example, for the object ID representing Bill Clinton, the value of a property fact may be the text string “Bill Clinton was the 42nd President of the United States from 1993 to 2001.” Some object IDs may have one or more associated property facts while other objects may have no associated property facts. It should be appreciated that the data structures shown inFIGS. 2(a)-2(d) and described above are merely exemplary. The data structure of therepository115 may take on other forms. Other fields may be included in facts and some of the fields described above may be omitted. Additionally, each object ID may have additional special facts aside from name facts and property facts, such as facts conveying a type or category (for example, person, place, movie, actor, organization, etc.) for categorizing the entity or concept represented by the object ID. In some embodiments, an object's name(s) and/or properties may be represented by special records that have a different format than the general facts records204.
• As described previously, a collection of facts is associated with an object ID of an object. An object may become a null or empty object when facts are disassociated from the object. A null object can arise in a number of different ways. One type of null object is an object that has had all of its facts (including name facts) removed, leaving no facts associated with its object ID. Another type of null object is an object that has all of its associated facts other than name facts removed, leaving only its name fact(s). Alternatively, the object may be a null object only if all of its associated name facts are removed. A null object represents an entity or concept for which thedata processing system106 has no factual information and, as far as thedata processing system106 is concerned, does not exist. In some embodiments, facts of a null object may be left in therepository115, but have their object ID values cleared (or have their importance to a negative value). However, the facts of the null object are treated as if they were removed from therepository115. In some other embodiments, facts of null objects are physically removed fromrepository115.
• FIG. 2(e) is a block diagram illustrating analternate data structure290 for facts and objects in accordance with preferred embodiments of the invention. In this data structure, anobject290 contains anobject ID292 and references or points tofacts294. Each fact includes afact ID295, anattribute297, and avalue299. In this embodiment, anobject290 actually exists inmemory107.
• Referring again toFIG. 1, the content of the facts in therepository115 is also indexed inindex127. Theindex127 maintains a term index, which maps terms to {object, fact, field, token} tuples, where “field” is, e.g., attribute or value. Theservice engine114 is adapted to receive keyword queries from clients such as object requestors, and communicates with theindex127 to retrieve the facts that are relevant to user's search query. For a generic query containing one or more terms, theservice engine114 assumes the scope is at the object level. Thus, any object with one or more of the query terms somewhere (not necessarily on the same fact) will match the query for purposes of being ranked in the search results.
• In one embodiment the ranking (score) of an object is a linear combination of relevance scores for each of the facts. The relevance score for each fact is based on whether the fact includes one or more query terms (a hit) in, for example, one of the attribute, value, or source portion of the fact. Each hit is scored based on the frequency of the term that is hit, with more common terms getting lower scores, and rarer terms getting higher scores (e.g., using a TF-IDF based term weighting model). The fact score is then adjusted based on additional factors. These factors include the appearance of consecutive query terms in a fact, the appearance of consecutive query terms in a fact in the order in which they appear in the query, the appearance of an exact match for the entire query, the appearance of the query terms in the name fact (or other designated fact, e.g., property or category), and the percentage of facts of the object containing at least one query term. Each fact's score is also adjusted by its associated confidence measure and by its importance measure. Since each fact is independently scored, the facts most relevant and important to any individual query can be determined, and selected. In one embodiment, a selected number (e.g., 5) of the top scoring facts is selected for display in response to a query.
• A user interface for browsing thefact repository115 is discussed in co-pending U.S. application Ser. No. 11/356,851, entitled “Browseable Fact Repository,” filed Feb. 17, 2006, which is incorporated by reference herein in its entirety.
• In a preferred embodiment of the invention, facts can have associated “annotations.” Annotations are subsets of snippets of the fact's value that are used to provide additional information about facts. Often, an annotation has a high confidence level, especially if it has been entered by a human being or somehow otherwise corroborated. For example, the string “New York, N.Y.” can be used to generate a map plot, but an annotation can associate a latitude and longitude with that string and the latitude/longitude can be plotted on a map. As another example, the Britney Spears object has a large number of facts relating to Britney Spears. Annotations exist for such facts as her birthday, wedding anniversary, and birthplace. It is possible to compare annotations of various facts to determine whether a fact has reasonable values. Certain types of information such as numbers, dates, and locations are good candidates for becoming annotations, as discussed in detail below.
• FIG. 2(f) shows a format of a plurality of annotations associated with facts inrepository115 in an embodiment of the invention. The Figure shows three types of annotations (although there can be other types of annotations):Number annotations296,date annotations297, and GeoPoint (GeoPt)annotations298. Each annotation has at least a fact ID, a beginning index in the value of the fact with the fact ID, and a length within the value of the fact value. The fact ID identifies which fact the annotation is associated with. The beginning index is an index into the fact's value where the annotation value begins. The length is a length of the annotation value within the fact's value.
• FIG. 2(g) shows an example of two facts and an associated annotation of one of the facts. As an example, if afact10 has a value of “Britney Spears was born Dec. 2, 1981” a date annotation associated with that fact might have thefact ID282 offact10, an index of284 (because the date starts on character 25, indexed from 1), and alength286 of 16 (because the date is 16 characters long). (Other embodiments index from 0.)
• In one embodiment, each type of fact is stored together (i.e., all the number facts together, etc). In other embodiments, annotation types are not necessarily stored together. In one embodiment, all fact values, including numeric values dates, latitude and longitude are stored inrepository115 as characters. Some embodiments also store a copy of the annotation value (for example “Dec. 2, 1981”) with the annotation. Other embodiments store a pointer or reference into the appropriate fact to retrieve the annotation value. Other embodiments store the annotations as part of the facts instead of separately. Some embodiments index annotations inindex127, causing them to be searchable.
• As shown inFIG. 3(a), annotations can be used when “vetting” facts stored inrepository115. Such an operation might be performed byjanitor110 or by any appropriate entity. For a fact received for review instep322,janitor110 determines324 whether the fact is duplicated by other facts in the repository. For example, two facts may contain the same information in different formats. For example, a plurality of facts might have annotations that indicate or suggest that string values of “10 million”, “10,000,000”, and “10000000.000” all represent the same value. As another example, a first fact may contain a distance value in metric format, while a second fact might contain the same distance in Imperial format. In some embodiments, comparison of annotations alone is sufficient to determine duplicates. In other embodiments, comparison of annotations is an initial step andjanitors110 can go on to make more accurate comparisons using the candidates identified by annotations. Thus, annotations can also be used as an initial screening device to find duplicate facts.
• For a given fact received instep324,janitor110 can also use annotations to determine326 whether the fact has a reasonable value. For example, if there exists an annotation for Britney Spears's birthplace stating that it is “Louisiana”, thenjanitor110 may use that information to vet the large numbers of facts associated with the Britney Spears object. The Britney Spears's birthplace annotation may have been automatically generated or it may have been hand-entered and vetted by a human being (the latter presumably being more trustworthy). For example, because Louisiana is abbreviated as “LA,” some facts associated with the Britney Spears object may mistakenly contain the value that Britney Spears was born in Los Angeles, which also is sometimes abbreviated as “LA.”Janitor110 can determine that the facts with a birthplace of Los Angeles” do not match the annotation and choose to remove (or not to store) those facts from therepository115.
• Similarly, the service engine can use annotations to determine which facts to send in response to a query. Facts having values in accordance with corresponding annotation values are given preference.
• FIGS. 3(b) and3(c) are flow charts showing additional methods of using annotations.FIG. 3(b) shows an method where a query is received352 (for example by object service114). The object service determines354 a plurality of facts that match the query. Then the object service determines356 one or more facts that correspond to the determined facts and returns358 the facts and the annotations. Thus, if the query is “Britney Spears”object service114 would determine that the facts ofFIG. 2(g) match and would also determine that the annotation ofFIG. 2(g) matches since it corresponds to facts that match the query. In this method, annotations are returned in addition to facts because they provide a fast way to locate pertinent data in the facts without a real-time search of the fact values.
• FIG. 3(c) is a flow chart showing use of annotations to help select facts. A query is received362 (for example by object service114). The object service determines366 a plurality of annotations that are relevant to the query. Relevance of annotations can be determined, for example, by accessing theindex127 if annotations are indexed, by accessing a separate annotation index (not shown) if annotations are separately indexed, by searching the values of the annotations, or by any other appropriate method. Then the object service determines366 one or more facts that correspond to the determined annotations and returns368 the facts. In some embodiments, the annotations are also returned. Thus, if the query is “Britney Spears's birthday”object service114 would determine that the annotation ofFIG. 2(g) matches and would also determine thatfact10 is associated with that annotation. Some embodiments would return onlyfact10. Other embodiments would returnfact10 and other facts of the “Britney Spears” object that is associated withfact10. Some embodiments may also return the annotation itself. In this method, annotations are used because they provide a fast way to locate pertinent data in the facts without a real-time search through the fact values.
• Annotations can also be used byservice engine114 to identify erroneous facts that do not match annotation values.
• FIG. 4 is a block diagram showing one method of annotation creation.FIG. 5 is a flow chart showing annotation creation.
• Arecognizer402 is preferably implemented as a software program or hardware device. It examines504 facts inrepository115 to extract annotation values. Recognizers may recognize, for example, dates, numbers, locations (which are used to create GeoPt annotations), distances, directions, ISBN numbers, names of persons, measurements, height, weight, mass, volume, and so on. As another example, measurements can be annotated to include units such as “10 meters”, “100 degrees Fahrenheit, and so on. Extracted annotations are encoded in one or more appropriately sized annotation blocks506 as shown inFIG. 2(f) with a fact ID, index, and length. In some embodiments, the annotation blocks are compacted508 before being stored510 inrepository115 or other appropriate storage. In other embodiments, annotations are added by human beings. In other embodiments, some annotations are added by human beings and some are automatically generated. In some embodiments, annotations added by human beings are given a higher confidence value than annotations that are automatically created.
• As discussed above, annotations can be returned in response to a query. This query can be, for example, a browser-based query, in which case, the annotations are returned with the query results. The annotations can be mixed in with the query results or can be displayed separately. For example, annotations can be displayed separately at the top of a results web page.
• Queries can also take the form of requests for XML feeds or other formats of data feeds.FIG. 6 shows an example of anpartial entry600 in an XML feed306 sent in response to a request from object requester152 (for example, the request: http://google.com/ref/api/q=china). This example is presented partly to show examples of some kinds of annotations and their values.
• In the example, an entry corresponds to an object and each atom corresponds to a fact in the object. An object can have more than one associated fact. This example uses the Atom format, although any appropriate feed format can be used.
• In the described embodiment, each object has an associated object ID601 (in this example “1a27a12f0de4b506a93022496b370c40”). The feed includes a plurality offacts602,604,606.Fact602 shows an example of a fact having an attribute and an attribute value, along with a link representing a source of the fact.
• Note that, in certain embodiments, for facts associated with an object to be returned in response to a query, only one fact needs to match the query. Thus, in the example, the q parameter is “q=China”.Fact602 has a fact with a value containing the string “China”.Facts604 and606 do not contain the query string “China,” but are returned because they are facts associated with the same object asfact602.
• Some embodiments return partial facts as well as or in addition to complete facts. For example, a query for the date of birth attribute for object number 2038472038 might just return “Jun. 17, 1974” and no other facts for that object.
• Fact604 also includes an attribute and a value, and further includes an annotation. In one embodiment, an annotation is associated with a fact inrepository115.
• InFIG. 6, the feed forfact604 has an attribute and a value:
• • attribute=“Population” value=“1306313812 (July 2005 est.)”
• The feed forfact604 also has a source link:
• • <atom:link href=“http://www.cia.gov”/>
• The feed forfact604 also has two associated annotations:

• 	<g:annotation begin=“0” length=“10”>
  	<g:number value=“1306313812” />
  	</g:annotation>
  	<g:annotation begin=“12” length=“9”>
  	<g:date value=“July 2005” />
  	</g:annotation>

• In this embodiment, each annotation has a beginning index, a length, and a value. One of the annotations is a number annotation with a value of “1306313812”. The other annotation is a date annotation with a value of “July 2005”. In some embodiments, a date annotation has separate fields for month, day, and year.
• In the feed forfact606 has an associated geopoint annotation:

• 	</g:annotation>
  	<g:annotation begin=“28” length=“39”>
  	<g:geoPt lat=“38.936283” lon=2“−77.05994” />
  	</g: annotation>

• A geopoint annotation represents a latitude and longitude associated with a fact. Geopoint annotations are noteworthy at least because they contain a two-part value (latitude and longitude). Other types of annotations can have two-or-more-part values as well. Note that in this example, the latitude and longitude values themselves do not appear in the fact. Instead, the value “Washington, District of Columbia, 20008” appears. At some point, an annotation was created for the latitude and longitude of Washington, D.C. and stored in association with that fact (and most likely with all other facts containing a mention of “Washington, D.C.”).
• It will be understood that other types of annotations can be used in the present invention, including but not limited to: name annotations, isbn annotations, units (to indicate a unit for numerical values).
• As discussed above, the feed received by an object requester can be used in a number of ways, including but not limited to display and calculation.
• Some embodiments also support an “alt” parameter, which allows the user to specify the format of the results. It has three possible values: “atom” (the default), “rss”, “osrss”.
EXAMPLES
• /alt=atom/q=albert+einstein
• results for the query “albert einstein” returned in Atom format
• /alt=rss/q=albert+einstein
• results for the query “albert einstein” returned in RSS format
• /alt=osrss/q=albert+einstein
• results for the query “albert einstein” returned in OpenSearch format
• The alt parameter can be combined with the various other parameters:
• /max-facts=5/alt=rss/max-sources=4/q=albert+einstein?start-index=10 five facts per object, four sources per fact, starting at the 10th result, in RSS format.
• The present invention has been described in particular detail with respect to one possible embodiment. Those of skill in the art will appreciate that the invention may be practiced in other embodiments. First, the particular naming of the components, capitalization of terms, the attributes, data structures, or any other programming or structural aspect is not mandatory or significant, and the mechanisms that implement the invention or its features may have different names, formats, or protocols. Further, the system may be implemented via a combination of hardware and software, as described, or entirely in hardware elements. Also, the particular division of functionality between the various system components described herein is merely exemplary, and not mandatory; functions performed by a single system component may instead be performed by multiple components, and functions performed by multiple components may instead performed by a single component.
• Some portions of above description present the features of the present invention in terms of algorithms and symbolic representations of operations on information. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. These operations, while described functionally or logically, are understood to be implemented by computer programs. Furthermore, it has also proven convenient at times, to refer to these arrangements of operations as modules or by functional names, without loss of generality.
• Unless specifically stated otherwise as apparent from the above discussion, it is appreciated that throughout the description, discussions utilizing terms such as “processing” or “computing” or “calculating” or “determining” or “displaying” or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system memories or registers or other such information storage, transmission or display devices.
• Certain aspects of the present invention include process steps and instructions described herein in the form of an algorithm. It should be noted that the process steps and instructions of the present invention could be embodied in software, firmware or hardware, and when embodied in software, could be downloaded to reside on and be operated from different platforms used by real time network operating systems.
• The present invention also relates to an apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored on a computer readable medium that can be accessed by the computer. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, application specific integrated circuits (ASICs), or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus. Furthermore, the computers referred to in the specification may include a single processor or may be architectures employing multiple processor designs for increased computing capability.
• The algorithms and operations presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may also be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems will be apparent to those of skill in the, along with equivalent variations. In addition, the present invention is not described with reference to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any references to specific languages are provided for disclosure of enablement and best mode of the present invention.
• The present invention is well suited to a wide variety of computer network systems over numerous topologies. Within this field, the configuration and management of large networks comprise storage devices and computers that are communicatively coupled to dissimilar computers and storage devices over a network, such as the Internet.
• Finally, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, the disclosure of the present invention is intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.

Claims (1)

1. A method of creating annotations for facts in a fact repository, performed at a computer system having one or more processors and memory storing one or more programs for execution by the one or more processors, comprising:

processing a fact in a fact repository, the fact having a fact attribute and a fact value, to determine a corresponding value for an annotation, wherein the annotation includes additional information about the fact, wherein the annotation is directed to a subset of the fact value, wherein the fact repository includes a plurality of facts associated with objects, wherein a respective fact in the fact repository includes a respective fact attribute and a respective fact value, wherein the respective fact value is a text string, and wherein the plurality of facts is extracted from a plurality of documents; and

storing the corresponding value as an annotation value in the fact repository, wherein the annotation value is indexed to a substring of the fact value.

Images