The disclosed embodiments relate generally to fact databases. More particularly, the disclosed embodiments relate to identifying duplicate objects in an object collection.
Data is often organized as large collections of objects. When objects are added over time, there are often problems with data duplication. For example, a collection may include multiple objects that represent the same entity. As used herein, the term “duplicate objects” refers to objects representing the same entity. The names used to describe the represented entity are not necessarily the same among the duplicate objects.
Duplicate objects are undesirable for many reasons. They increase storage cost and take a longer time to process. They lead to inaccurate results, such as an inaccurate count of distinct objects. They also cause data inconsistency.
Conventional approaches identifying duplicate objects assume a homogeneity in the input set of objects (all books, all products, all movies, etc). Identifying duplication for objects of different type requires looking at different fields for different type. For example, when identifying duplicate objects in a set of objects representing books, traditional approaches match the ISBN value of the objects; when identifying duplicate objects in objects representing people, traditional approaches match the SSN value of the objects. One drawback of the conventional approaches is that they are only effective for specific types of objects, and tend to be ineffective when applied to a collection of objects with different types. Also, even if the objects in the collection are of the same type, these approaches tend to be ineffective when the objects include incomplete or inaccurate information.
What is needed is a method and system that identifies duplicate objects in a large number of objects having different types and/or incomplete information.
The invention is a system and method for identifying duplicate objects from a plurality of objects. For each object, the name used to describe the represented entity is normalized. A signature is generated for each object based on the normalized name. Objects are grouped into buckets based on the signature of the objects. Objects within the same bucket are compared to each other using a matcher to identify duplicate objects. The matcher can be selected from a collection of matchers.
This approach normalizes names used by objects to describe the represented entity. Therefore, objects representing the same entity share the same normalized name. As a result, this approach can identify duplicate objects even if the associated names initially are different. This approach is also computationally cost-efficient because objects are pair-wise matched only within a bucket, rather than being pair-wise matched across all buckets.
These features and benefits are not the only features and benefits of the invention. In view of the drawings, specification, and claims, many additional features and benefits will be apparent.
The figures depict various embodiments 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.
System Architecture
Document hosts 102 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 a document host 102 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. A document host 102 is implemented by a computer system, and typically includes a server adapted to communicate over the network 104 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 a host 102 are typically held in a file directory, a database, or other data repository. A host 102 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.
Janitors 110 operate to process facts extracted by importer 108. This processing can include but is not limited to, data cleansing, object merging, and fact induction. In one embodiment, there are a number of different janitors 110 that perform different types of data management operations on the facts. For example, one janitor 110 may traverse some set of facts in the repository 115 to find duplicate facts (that is, facts that convey the same factual information) and merge them. Another janitor 110 may also normalize facts into standard formats. Another janitor 110 may also remove unwanted facts from repository 115, such as facts related to pornographic content. Other types of janitors 110 may be implemented, depending on the types of data management functions desired, such as translation, compression, spelling or grammar correction, and the like.
Various janitors 110 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' birthday is “Dec. 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 Bilihdate by one janitor and then another janitor might notice that Dec. 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 engine 112 builds and manages the repository 115. Service engine 114 is an interface for querying the repository 115. Service engine 114's main function is to process queries, score matching objects, and return them to the caller but it is also used by janitor 110.
Repository 115 stores factual information extracted from a plurality of documents that are located on document hosts 102. 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.
Repository 115 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 the repository 115, but rather are defined by the set or group of facts with the same associated object ID, as described below. Further details about facts in repository 115 are described below, in relation to
It should be appreciated that in practice at least some of the components of the data processing system 106 will be distributed over multiple computers, communicating over a network. For example, repository 115 may be deployed over multiple servers. As another example, the janitors 110 may be located on any number of different computers. For convenience of explanation, however, the components of the data processing system 106 are discussed as though they were implemented on a single computer.
In another embodiment, some or all of document hosts 102 are located on data processing system 106 instead of being coupled to data processing system 106 by a network. For example, importer 108 may import facts from a database that is a part of or associated with data processing system 106.
Data Structure
As described above, each fact is associated with an object ID 209 that identifies the object that the fact describes. Thus, each fact that is associated with a same entity (such as George Washington), will have the same object ID 209. 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.
Also, while the illustration of
Each fact 204 also may include one or more metrics 218. 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.
Each fact 204 includes a list of one or more sources 220 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 in
Some embodiments include one or more specialized facts, such as a name fact 207 and a property fact 208. A name fact 207 is a fact that conveys a name for the entity or concept represented by the object ID. A name fact 207 includes an attribute 224 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.” A name fact 207, being a special instance of a general fact 204, includes the same fields as any other fact 204; it has an attribute, a value, a fact ID, metrics, sources, etc. The attribute 224 of a name fact 207 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.
A property fact 208 is a fact that conveys a statement about the entity or concept represented by the object ID. Prope liy facts are generally used for summary information about an object. A property fact 208, being a special instance of a general fact 204, also includes the same parameters (such as attribute, value, fact ID, etc.) as other facts 204. The attribute field 226 of a property fact 208 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 in
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 the data processing system 106 has no factual information and, as far as the data processing system 106 is concerned, does not exist. In some embodiments, facts of a null object may be left in the repository 115, 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 the repository 115. In some other embodiments, facts of null objects are physically removed from repository 115.
Overview of Methodology
Referring now to
In one embodiment, the present invention is implemented in a janitor 110 to identify duplicate objects so that the duplicate objects can be merged together. Duplicate objects are objects representing the same entity but having a different object ID. Janitor 110 examines object reference table 210, and reconstructs the objects based on the associations between object IDs and fact IDs maintained in object reference table 210. Alternatively, janitor 110 can retrieve objects by asking service engine 114 for the information stored in repository 115. Depending how object information is stored in repository 115, janitor 110 needs to reconstruct the objects based on the facts and object information retrieved.
The flowchart shown in
As shown in
Referring to
In one embodiment, normalizer 410 normalizes a name value by applying a set of normalization rules to the name value. A normalization rule can remove from the name value information unnecessary to describe the represented entity (e.g., removing the from the United States). Alternatively, a normalization rule can standardize the format of the name value (e.g., changing a person's name from a last name first order to a first name first order, such as from Washington, George to George Washington). Some of the normalization rules are language specific while others are universally applicable to name values in different languages. Some embodiments allow a name fact to indicate that the associated name value is an exception to one or more of the normalization rules. When normalizer 410 identifies such indication it will not apply the normalization rules indicated. For example, an object with a name value of J. F. K may indicate that the associated name value is an exception to a single-letter-word removal rule.
One example of the normalization rules, uppercase-to-lowercase conversion rule, converts uppercase characters in a name value to corresponding lowercase characters, such as from “America” to “america.” The name values of some duplicate objects may use capital characters to describe the represented entity while the name values of others may ignore capital characters. For example, one object representing the Apple computer iMac may have a name value of iMac, while other objects representing the same entity may have a name value of Imac, imac, or IMAC. Each of the above four distinct name values describes the same entity—the Apple computer iMac. By applying the uppercase-to-lowercase conversion rule, all four name values are standardized to be imac. Applying the uppercase-to-lowercase conversion rule to the name values of the set of objects illustrated in
Another example of the normalization rules, stop-words removal rule, removes stop words from name values. Stop words are small or frequently used words that are generally overlooked by the search engines. Common stop words are words such as the, a, an, this, and that. Stop words tend to convey no additional value in describing the represented entity, therefore the name values of some objects include stop words while the name values of others do not. For example, for two duplicate objects describing the United Nations, one may have a name value of the United Nations while the other may have a name value of United Nations. By applying the above stop-words-removal rule, the two duplicate objects' name values are standardized to be United Nations. Normalizer 410 can dynamically update the collection of words it deems as stop words.
Another example of the normalization rules, social-titles removal rule, removes social titles from name values. Social titles are identifying appellations signifying status of the entity described. Common social titles are words such as Mr., Ms., Mrs., Miss, Sir, etc. Because social titles are not essential in identifying the represented entity, the name values of some objects do not include them. For example, for two duplicate objects representing the English mathematician and physicist Isaac Newton, one may have a name value of Sir Isaac Newton while the other may have a name value of Isaac New ton. By applying the above social-titles removal rule, the two duplicate objects' name values are standardized to be Isaac Newton. Similarly, the normalization rules can include a honorific-titles removal rule which removes honorific words such as General, President, Congressman, Senator from the name values.
Applying the uppercase-to-lowercase conversion rule to the normalized name values shown in
Another example of the normalization rules, single-letter-word removal rule, removes single letter words from name values. When identifying an entity, certain non-essential words are often omitted or only shown their initial characters. One example of such non-essential words is a person's middle name. Some objects representing a person includes the person's middle name initial in the associated name value while others do not. For example, for two duplicate objects representing a John Henry, one may have a name value of John W. Henry while the other may have a name value of John Henry. By applying the above single-letter-word removal rule and a punctuation-marks removal rule as described below, the two duplicate objects' name values are standardized to be John Henry.
Another example of the normalization rules, alphabetic sort rule, sorts the words in a name value in alphabetic order. When identifying an entity, the name of the entity can be in one of several different formats. For example, China can be either called People's Republic of China or China, People's Republic. Also, a person can either be addressed in a first-name first way or in a more formal last-name first way. Correspondingly, for two duplicate objects representing a person named John Henry, the name value of one object can be John Henry while the name value of the other can be Henry, John. By applying the alphabetic sort rule and a punctuation marks removal rule as described below, the two duplicate objects' name values are standardized to be Henry John.
Yet another example of the normalization rules, punctuation-marks removal rule, removes punctuation marks from name values. Punctuation marks are used to clarify meaning by indicating separation of words into clauses and phrases. Because punctuation marks are not essential in identifying an entity, some objects omit them in the associated name values. Also, punctuation marks in a fact value may become unnecessary after normalizer 410 applies one or more normalization rules to the fact value. For example, after applying the alphabetic sorting rule to a name value of Henry, Bill, the name value becomes Bill Henry, and the comma mark becomes unnecessary. The punctuation removal rule removes the extra comma sign and standardizes the name value to be Bill Henry.
Applying the single-letter-word removal rule, the alphabetic sort rule, and the punctuation-marks removal rule to the normalized name values shown in
Referring to
The purpose of generating a signature is to optimize the object normalization process. In general, normalizing a repository of objects requires comparing all possible pairs of objects in the repository, which is computationally impractical for a large collection of objects. As an optimization, it's desirable to design the signature generator 440 to always create the same signature for duplicate objects. As a result, only objects sharing the same signature need to be compared to identify duplicate objects and normalize the repository of objects. In order for the signature generator 440 to always create the same signature for duplicate objects, it needs to be inclusive and ignore minor differences among the objects.
In one embodiment, signature generator 440 generates 320 signatures 450 based solely on the name values of objects 430. For example, signature generator 440 can generate 320 the signature 450 by removing any white space in the name value of object 430. Janitor 110 then places object 430 into a bucket 460 in accordance with signature 450.
It is noted that signatures 450 generated by signature generator 440 can be a null signature, a signature with an empty value. In one embodiment, janitor 110 does not place an object 430 with a null signature into any bucket 460. As a result, objects with null signatures are neither compared nor merged with other objects. Signature generator 440 can generate a null signature because the object does not have a name fact. Signature generator 440 can also generate a null signature because the normalized name value of the object is empty (e.g., the original name value consists only of stop words, and the normalizer 410 removes all the stop words from the name value). Alternatively, the signature generator 440 can purposefully generate a null signature for certain objects to prevent the objects from being considered for merger.
Alternatively, signature generator 440 can generate 320 signatures 450 based on a combination of name values and other fact values of the associated objects 430. In one example, the signature generator 440 applies some normalization rules similar to the ones described above to the other fact values before generating 320 the signature 450.
By normalizing 310 the name values of each object, janitor 110 can detect duplicate objects with different name values describing the same represented entity. Objects 430 created from different data sources may not share the same name value, even if they represent the same entity. For example, an object 430 representing George Washington created based on a webpage devoted to his childhood may have a name value of George Washington, while another object 430 also representing George Washington created based on a webpage dedicated to his years of presidency probably would have a different name value of President George Washington. By normalizing the name values of each object, janitor 110 can standardize the name values such that objects representing the same entity share the same normalized name value. For example, the normalized name value of both of the above objects are george washington.
Because signature 450 is based on the normalized name value, signature generator 440 generates the same signature 450 for duplicate objects. Because janitor 110 groups objects 430 based on their associated signatures, duplicate objects tend to be grouped 330 into the same bucket 460. For example, as illustrated in
After all objects are grouped 430 into buckets 460, for every bucket 460 created, janitor 110 applies 340 a matcher 420 to every two objects in bucket 460, and identifies 350 the matching objects 470 as duplicate objects. Matcher 420 is designed to match duplicate objects based on the similarity of facts with the same attribute associated with the two objects (also called simply common facts). Similarity between two corresponding facts can be determined in a number of ways. For example, two facts are determined to be similar when the fact values are identical. In another example, two facts can be determined to be similar when the fact values are lexically similar, such as “U.S.A.” and “United States.” Alternatively, two facts are determined' to be similar when the fact values are proximately similar, such as “176 pounds” and “176.1 pounds.” In another example, two facts are determined to be similar when the fact values are similar based on string similarity measure (e.g., edit distance, Hamming Distance, Levenshtein Distance, Smith-Waterman Distance, Gotoh Distance, Jaro Distance Metric, Dice's Coefficient, Jaccard Coefficient to name a few).
For example, matcher 420 determines whether two objects match based on the number of common facts with similar values (also called simply similar common facts) and the number of common facts with values that are not similar (also called simply dissimilar common facts). In one such matcher 420, two objects are deemed to match when there is more similar common fact than dissimilar common facts. Because the name values are used to generate 320 the signatures of each object 430, matcher 420 does not consider name fact when determining whether two objects match.
When the above matcher 420 is applied to the buckets shown in
In another embodiment, janitor 110 does not first apply matcher 420 to every two objects in bucket 460 and then identify matching objects 470 as duplicate objects. Instead, janitor 110 applies matcher 420 to two objects in bucket 460. If matcher 420 indicates the two objects to be matching objects 470, janitor 110 merges them, keeps the merged object in bucket 460, and removes the other object(s) out of bucket 460. Then, janitor 110 restarts the process by applying matcher 420 to two objects in bucket 460 that have not been matched before. This process continues until matcher 420 has been applied to every pair of objects in bucket 460.
Janitor 110 can merge two objects in several different ways. For example, janitor 110 can choose one of the two objects as the merged object, add facts only present in the other object to the merged object, and optionally reconcile the dissimilar common facts of the merged object. Alternatively, janitor 110 can create a new object as the merged object, and add facts from the two matching objects to the merged object.
In another embodiment, a matcher 420 can be a function or a module. The system selects matcher 420 from a collection of matcher functions/modules. The collection of matcher functions/modules includes functions/modules provided by a third party and functions/modules previously created. By providing the ability to select a matcher 420, janitor 110 can reuse the existing well-tested functions/modules, and select matcher 420 based on the specific needs.
There are many ways for janitor 110 to select a matcher function/module. For example, janitor 110 can select matcher 420 based on system configuration data. Alternatively, the selection can be determined at run time based on information such as grouper 410 selected. For example, if the resulting buckets of grouper 410 include many objects, janitor 110 selects a matcher function/module requiring a higher entropy threshold.
After identifying 350 the matching objects as duplicate objects, janitor 110 can merge the duplicate objects into a merged object, so that each entity is represented by no more than one object and each fact that is associated with a same entity will have the same object ID. 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.
This application is a continuation of U.S. patent application Ser. No. 14/229,774, entitled “Entity Normalization Via Name Normalization,” by Jonathan T. Betz, filed on Mar. 28, 2014, which is a continuation of U.S. patent application Ser. No. 11/394,508, now U.S. Pat. No. 8,700,568, entitled “Entity Normalization Via Name Normalization,” by Jonathan T. Betz, filed on Mar. 31, 2006, which is a continuation-in-part of U.S. patent application Ser. No. 11/356,838, now U.S. Pat. No. 7,672,971, entitled “Modular Architecture For Entity Normalization,” by Jonathan T. Betz and Farhan Shamsi, filed on Feb. 17, 2006. All above-identified patents and/or patent applications are hereby incorporated by reference in its entirety. This application potentially relates to the following U.S. Applications, all of which are incorporated by reference herein: U.S. application Ser. No. 11/366,162, entitled “Generating Structured Information,” filed Mar. 1, 2006, by Egon Pasztor and Daniel Egnor; U.S. application Ser. No. 11/357,748, entitled “Support for Object Search,” filed Feb. 17, 2006, by Alex Kehlenbeck, Andrew W. Hogue; U.S. application Ser. No. 11/342,290, entitled “Data Object Visualization,” filed on Jan. 27, 2006, by Andrew W. Hogue, David Vespe, Alex Kehlenbeck, Mike Gordon, Jeffrey C. Reynar, David Alpert; U.S. application Ser. No. 11/342,293, entitled “Data Object Visualization Using Maps,” filed on Jan. 27, 2006, by Andrew W. Hogue, David Vespe, Alex Kehlenbeck, Mike Gordon, Jeffrey C. Reynar, David Alpert; U.S. application Ser. No. 11/356,679, entitled “Query Language,” filed Feb. 17, 2006, by Andrew W. Hogue, Doug Rohde; U.S. application Ser. No. 11/356,837, entitled “Automatic Object Reference Identification and Linking in a Browseable Fact Repository,” filed Feb. 17, 2006, by Andrew W. Hogue; U.S. application Ser. No. 11/356,851, entitled “Browseable Fact Repository,” filed Feb. 17, 2006, by Andrew W. Hogue, Jonathan T. Betz; U.S. application Ser. No. 11/356,842, entitled “ID Persistence Through Normalization,” filed Feb. 17, 2006, by Jonathan T. Betz, Andrew W. Hogue; U.S. application Ser. No. 11/356,728, entitled “Annotation Framework,” filed Feb. 17, 2006, by Tom Richford, Jonathan T. Betz; U.S. application Ser. No. 11/341,069, entitled “Object Categorization for Information Extraction,” filed on Jan. 27, 2006, by Jonathan T. Betz; U.S. application Ser. No. 11/356,838, entitled “Modular Architecture for Entity Normalization,” filed Feb. 17, 2006, by Jonathan T. Betz, Farhan Shamsi; U.S. application Ser. No. 11/356,765, entitled “Attribute Entropy as a Signal in Object Normalization,” filed Feb. 17, 2006, by Jonathan T. Betz, Vivek Menezes; U.S. application Ser. No. 11/341,907, entitled “Designating Data Objects for Analysis,” filed on Jan. 27, 2006, by Andrew W. Hogue, David Vespe, Alex Kehlenbeck, Mike Gordon, Jeffrey C. Reynar, David Alpert; U.S. application Ser. No. 11/342,277, entitled “Data Object Visualization Using Graphs,” filed on Jan. 27, 2006, by Andrew W. Hogue, David Vespe, Alex Kehlenbeck, Mike Gordon, Jeffrey C. Reynar, David Alpert; U.S. application Ser. No. 11/394,610, entitled “Determining Document Subject by Using Title and Anchor Text of Related Documents,” filed on Mar. 31, 2006, by Shubin Zhao; U.S. application Ser. No. 11/394,552, entitled “Anchor Text Summarization for Corroboration,” filed on Mar. 31, 2006, by Jonathan T. Betz and Shubin Zhao; and U.S. application Ser. No. 11/394,414, entitled “Unsupervised Extraction of Facts,” filed on Mar. 31, 2006, by Jonathan T. Betz and Shubin Zhao.
Number | Name | Date | Kind |
---|---|---|---|
5010478 | Deran | Apr 1991 | A |
5133075 | Risch | Jul 1992 | A |
5347653 | Flynn et al. | Sep 1994 | A |
5440730 | Elmasri et al. | Aug 1995 | A |
5475819 | Miller et al. | Dec 1995 | A |
5519608 | Kupiec | May 1996 | A |
5560005 | Hoover et al. | Sep 1996 | A |
5574898 | Leblang et al. | Nov 1996 | A |
5680622 | Even | Oct 1997 | A |
5694590 | Thuraisingham et al. | Dec 1997 | A |
5701470 | Joy et al. | Dec 1997 | A |
5717911 | Madrid et al. | Feb 1998 | A |
5717951 | Yabumoto | Feb 1998 | A |
5778378 | Rubin | Jul 1998 | A |
5787413 | Kauffman et al. | Jul 1998 | A |
5793966 | Amstein et al. | Aug 1998 | A |
5802299 | Logan et al. | Sep 1998 | A |
5815415 | Bentley et al. | Sep 1998 | A |
5819210 | Maxwell et al. | Oct 1998 | A |
5819265 | Ravin et al. | Oct 1998 | A |
5822743 | Gupta et al. | Oct 1998 | A |
5826258 | Gupta et al. | Oct 1998 | A |
5838979 | Hart et al. | Nov 1998 | A |
5909689 | Van | Jun 1999 | A |
5920859 | Li | Jul 1999 | A |
5943670 | Prager | Aug 1999 | A |
5956718 | Prasad et al. | Sep 1999 | A |
5974254 | Hsu | Oct 1999 | A |
5987460 | Niwa et al. | Nov 1999 | A |
6018741 | Howland et al. | Jan 2000 | A |
6044366 | Graffe et al. | Mar 2000 | A |
6052693 | Smith et al. | Apr 2000 | A |
6078918 | Allen et al. | Jun 2000 | A |
6112203 | Bharat et al. | Aug 2000 | A |
6112210 | Nori et al. | Aug 2000 | A |
6122647 | Horowitz et al. | Sep 2000 | A |
6134555 | Chadha et al. | Oct 2000 | A |
6138270 | Hsu | Oct 2000 | A |
6173275 | Caid et al. | Jan 2001 | B1 |
6212526 | Chaudhuri et al. | Apr 2001 | B1 |
6240546 | Lee et al. | May 2001 | B1 |
6263328 | Coden et al. | Jul 2001 | B1 |
6285999 | Page | Sep 2001 | B1 |
6289338 | Wood et al. | Sep 2001 | B1 |
6311194 | Sheth et al. | Oct 2001 | B1 |
6327574 | Kramer et al. | Dec 2001 | B1 |
6349275 | Schumacher et al. | Feb 2002 | B1 |
6377943 | Jakobsson et al. | Apr 2002 | B1 |
6397228 | Lamburt et al. | May 2002 | B1 |
6473898 | Waugh et al. | Oct 2002 | B1 |
6502102 | Haswell et al. | Dec 2002 | B1 |
6556991 | Borkovsky et al. | Apr 2003 | B1 |
6567936 | Yang et al. | May 2003 | B1 |
6572661 | Stern et al. | Jun 2003 | B1 |
6584464 | Warthen et al. | Jun 2003 | B1 |
6606625 | Muslea et al. | Aug 2003 | B1 |
6606659 | Hegli et al. | Aug 2003 | B1 |
6609123 | Cazemier et al. | Aug 2003 | B1 |
6656991 | Staccione et al. | Dec 2003 | B2 |
6665659 | Logan et al. | Dec 2003 | B1 |
6665666 | Brown et al. | Dec 2003 | B1 |
6665837 | Dean et al. | Dec 2003 | B1 |
6745189 | Schreiber et al. | Jun 2004 | B2 |
6754873 | Law et al. | Jun 2004 | B1 |
6799176 | Page et al. | Sep 2004 | B1 |
6804667 | Martin et al. | Oct 2004 | B1 |
6820081 | McDonald et al. | Nov 2004 | B1 |
6820093 | De et al. | Nov 2004 | B2 |
6823495 | Vedula et al. | Nov 2004 | B1 |
6850896 | Kelman et al. | Feb 2005 | B1 |
6886005 | Davis et al. | Apr 2005 | B2 |
6901403 | Bata | May 2005 | B1 |
6957213 | Yuret et al. | Oct 2005 | B1 |
6963880 | Pingte et al. | Nov 2005 | B1 |
7003522 | Reynar et al. | Feb 2006 | B1 |
7003719 | Rosenoff et al. | Feb 2006 | B1 |
7020662 | Boreham et al. | Mar 2006 | B2 |
7051023 | Kapur et al. | May 2006 | B2 |
7080073 | Jiang et al. | Jul 2006 | B1 |
7080085 | Raghavan et al. | Jul 2006 | B1 |
7143099 | Dunham et al. | Nov 2006 | B2 |
7146536 | Bingham et al. | Dec 2006 | B2 |
7162499 | Lees et al. | Jan 2007 | B2 |
7194380 | Barrow et al. | Mar 2007 | B2 |
7197449 | Hu et al. | Mar 2007 | B2 |
7277879 | Varadarajan et al. | Oct 2007 | B2 |
7305380 | Hoelzle et al. | Dec 2007 | B1 |
7363312 | Goldsack et al. | Apr 2008 | B2 |
7472182 | Young et al. | Dec 2008 | B1 |
7483829 | Murakami et al. | Jan 2009 | B2 |
7493317 | Geva | Feb 2009 | B2 |
7610382 | Siegel | Oct 2009 | B1 |
7672971 | Betz et al. | Mar 2010 | B2 |
7747571 | Boggs | Jun 2010 | B2 |
7797282 | Kirshenbaum et al. | Sep 2010 | B1 |
7953720 | Rohde et al. | May 2011 | B1 |
20020038307 | Obradovic et al. | Mar 2002 | A1 |
20020042707 | Zhao et al. | Apr 2002 | A1 |
20020065845 | Naito et al. | May 2002 | A1 |
20020073115 | Davis et al. | Jun 2002 | A1 |
20020083039 | Ferrari et al. | Jun 2002 | A1 |
20020087567 | Spiegler et al. | Jul 2002 | A1 |
20020147738 | Reader et al. | Oct 2002 | A1 |
20020169730 | Lazaridis et al. | Nov 2002 | A1 |
20020169770 | Kim et al. | Nov 2002 | A1 |
20020178488 | Harris et al. | Dec 2002 | A1 |
20020194172 | Schreiber et al. | Dec 2002 | A1 |
20030018652 | Heckerman et al. | Jan 2003 | A1 |
20030058706 | Okamoto et al. | Mar 2003 | A1 |
20030078902 | Leong et al. | Apr 2003 | A1 |
20030120644 | Shirota et al. | Jun 2003 | A1 |
20030120670 | Nareddy et al. | Jun 2003 | A1 |
20030120675 | Stauber et al. | Jun 2003 | A1 |
20030126102 | Borthwick et al. | Jul 2003 | A1 |
20030149567 | Schmitz et al. | Aug 2003 | A1 |
20030154071 | Shreve et al. | Aug 2003 | A1 |
20030177110 | Okamoto et al. | Sep 2003 | A1 |
20030182310 | Charnock et al. | Sep 2003 | A1 |
20030195877 | Ford et al. | Oct 2003 | A1 |
20030196052 | Bolik et al. | Oct 2003 | A1 |
20040003067 | Ferrin et al. | Jan 2004 | A1 |
20040024739 | Copperman et al. | Feb 2004 | A1 |
20040049499 | Nomoto et al. | Mar 2004 | A1 |
20040064447 | Simske et al. | Apr 2004 | A1 |
20040088292 | Dettinger et al. | May 2004 | A1 |
20040107125 | Guheen et al. | Jun 2004 | A1 |
20040107205 | Burdick et al. | Jun 2004 | A1 |
20040122844 | Malloy et al. | Jun 2004 | A1 |
20040123240 | Gerstl et al. | Jun 2004 | A1 |
20040128624 | Arellano et al. | Jul 2004 | A1 |
20040153456 | Charnock et al. | Aug 2004 | A1 |
20040167870 | Wakefield et al. | Aug 2004 | A1 |
20040177015 | Galai et al. | Sep 2004 | A1 |
20040240542 | Yeredor et al. | Dec 2004 | A1 |
20040255237 | Tong et al. | Dec 2004 | A1 |
20040260979 | Kumai et al. | Dec 2004 | A1 |
20040267700 | Dumais et al. | Dec 2004 | A1 |
20050055399 | Savchuk et al. | Mar 2005 | A1 |
20050076012 | Manber et al. | Apr 2005 | A1 |
20050086211 | Mayer et al. | Apr 2005 | A1 |
20050086222 | Wang et al. | Apr 2005 | A1 |
20050089223 | Krumm et al. | Apr 2005 | A1 |
20050097150 | McKeon et al. | May 2005 | A1 |
20050125311 | Chidiac et al. | Jun 2005 | A1 |
20050149576 | Marmaros et al. | Jul 2005 | A1 |
20050149851 | Mittal et al. | Jul 2005 | A1 |
20050187923 | Cipollone et al. | Aug 2005 | A1 |
20050240615 | Barsness et al. | Oct 2005 | A1 |
20050256825 | Dettinger et al. | Nov 2005 | A1 |
20050288982 | Konno et al. | Dec 2005 | A1 |
20060026122 | Hurwood et al. | Feb 2006 | A1 |
20060036504 | Allocca et al. | Feb 2006 | A1 |
20060036686 | Shimogori et al. | Feb 2006 | A1 |
20060041597 | Conrad et al. | Feb 2006 | A1 |
20060047838 | Chauhan et al. | Mar 2006 | A1 |
20060053171 | Eldridge et al. | Mar 2006 | A1 |
20060053175 | Gardner et al. | Mar 2006 | A1 |
20060074824 | Li et al. | Apr 2006 | A1 |
20060074910 | Yun et al. | Apr 2006 | A1 |
20060085465 | Nori et al. | Apr 2006 | A1 |
20060085755 | Arai et al. | Apr 2006 | A1 |
20060143227 | Helm et al. | Jun 2006 | A1 |
20060143603 | Kalthoff et al. | Jun 2006 | A1 |
20060152755 | Curtis et al. | Jul 2006 | A1 |
20060167991 | Heikes et al. | Jul 2006 | A1 |
20060178848 | Ilzuka et al. | Aug 2006 | A1 |
20060224582 | Hogue et al. | Oct 2006 | A1 |
20060238919 | Bradley et al. | Oct 2006 | A1 |
20060248045 | Toledano et al. | Nov 2006 | A1 |
20060248456 | Bender et al. | Nov 2006 | A1 |
20060253418 | Charnock et al. | Nov 2006 | A1 |
20060288268 | Srinivasan et al. | Dec 2006 | A1 |
20060293879 | Zhao et al. | Dec 2006 | A1 |
20070005593 | Self et al. | Jan 2007 | A1 |
20070005639 | Gaussier et al. | Jan 2007 | A1 |
20070016890 | Brunner et al. | Jan 2007 | A1 |
20070038610 | Omoigui et al. | Feb 2007 | A1 |
20070073768 | Goradia et al. | Mar 2007 | A1 |
20070094246 | Dill et al. | Apr 2007 | A1 |
20070130123 | Majumder et al. | Jun 2007 | A1 |
20070150800 | Betz et al. | Jun 2007 | A1 |
20080071739 | Kumar et al. | Mar 2008 | A1 |
20080127211 | Belsey et al. | May 2008 | A1 |
20090006359 | Liao et al. | Jan 2009 | A1 |
Number | Date | Country |
---|---|---|
5-174020 | Jul 1993 | JP |
0127713 | Apr 2001 | WO |
2004114163 | Dec 2004 | WO |
2006104951 | Oct 2006 | WO |
Entry |
---|
Richardson, et al., “Beyond page rank: machine learning for static ranking”, International World Wide Web Conference Committee, 2006, 9 pages. |
Richardson, et al., “The intelligent surfer: probabilistic combination of link and content information in page rank”, Advance in Neural Information Processing System, vol. 14, MIT Press, Cambridge, MA, 2002, 8 pages. |
Riloff, et al., “Learning dictionaries for information extraction by multi-level bootstrapping”, American Association for Artificial Intelligence, 1999, 6 pages. |
Shannon, et al., “A mathematical theory of communication”, The Bell System Technical Journal, vol. 27, Oct. 1948, 55 pages. |
Sun Microsystems, “Attribute Names”, http://java.sun.com/products/jndi/tutorial/basics/direct01:y/attrnames.html, Feb. 2004, 2 pages. |
Wang , et al., “C4-2: Combining link and contents in clustering web search results to improve information interpretation”, The University of Tokyo, 2002, 9 pages. |
Wirzenius , “C preprocessor trick for implementing similar data types”, Jan. 2000, 9 pages. |
Cover, Entropy, relative entropy and mutual information, Chapter 2 Elements of Information Theory, 1991, 13 pages. |
“Andritsos: Information-theoretic tools for mining database structure from large data sets”, ACM SIGMOD, Jun. 13-18, 2004, 12 pages. |
“Information entropy”, Wikipedia, May 3, 2006, 9 pages. |
“Information theory”, Wikipedia, May 3, 2006, 12 pages. |
Merriam Webster dictionary defines “normalize” as “to make conform to or reduce to a norm or standard”, 1865, 2 pages. |
Merriam Webster dictionary defines “value” as “a numerical quantity that is assigned or is determined by calculation or measurement”, 1300, 2 pages. |
Microsoft Computer Dictionary defines “normalize” as “adjust number within specific range”, retrieved on Sep. 8, 2008 from http://proquest.safaribooksonline.com/0735614954, 4 pages. |
Microsoft Computer Dictionary defines “quantity” as “a number”, retrieved on Sep. 8, 2008 from http://proquest.safaribooksonline.com/0735614954, 4 pages. |
Microsoft Computer Dictionary defines “value” as “a quantity”, retrieved on Sep. 8, 2008 from http://proquest.safaribooksonline.com/0735614954, 4 pages. |
Agichtein , “Snowball Extracting relations from large plain-text collections”, Dec. 1999, 13 pages. |
Brill, et al., “An analysis of the ask MSR question-answering system”, Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), Jul. 2002, pp. 257-264. |
Brin, “Extracting patterns and relations from the world wide web”, 1999, 12 pages. |
Brin, “The anatomy of a large-scale hypertextual search engine”, Apr. 14-18, 1998, 26 pages. |
Bunescu, et al., “Using encyclopedia knowledge for named entity disambiguation”, Department of Computer Science, University of Texas, retrieved from the internet, Dec. 28, 2006, 8 pages. |
Chang, et al., “IEPAD—information extraction Based on Pattern Discovery”, Tenth International World Wide Web Conference, 2001, pp. 681-688. |
Chaudhuri, “Robust and Efficient Fuzzy Match for Online Data Cleaning”, SIGMOD 2003, San Diego, CA, Jun. 12, 2003, 12 pages. |
Chen, “A scheme for inference problems using rough sets and entropy”, Lecture notes in Computer Science, vol. 3642/2005, Aug. 31-Sep. 3, 2005, 10 pages. |
Chen, “Exploiting Relationships for Object Consolidation”, IQIS'05, Baltimore, MD, Jun. 2005, 12 pages. |
Cheng, “Entropy-based subspace clustering for mining numerical data”, 1999. |
Chu-Carroll, et al., “A multi-strategy and multi-source approach to question answering”, 2006, 8 pages. |
Craswell, et al., “Effective site finding using link anchor information”, SIGIR'01, Sep. 2001, 8 pages. |
Dean, et al., “MapReduce: Simplified Data Processing on Large Clusters”, 6th Symposium on Operating Systems Design and Implementation, Dec. 2004, 13 pages. |
Dey, “A distance-based approach to entity reconciliation in heterogeneous databases”, May-Jun. 2002. |
Dong, et al., “Reference reconciliation in complex information spaces”, SIGACM-SIGMOD, 2005, 12 pages. |
Downey, “Leaming text patterns for web information extraction and assessment”, American Association for Artificial Intelligence, 2002, 6 pages. |
Etzioni, et al., “Web-scale information extraction in knowitall (preliminary results)”, WWW04, AMC, May 2004, 11 pages. |
Freitag, et al., “Boosted wrapped induction”, American Association of Artificial Intelligence, 2000, 7 pages. |
Gao, et al., “Learning information extraction patterns from tabular web pages without manual labelling”, Proceedings of the IEEE/WIC Int'l Conf. on Web Intelligence (WI'03), Oct. 2009, 4 pages. |
Gray, “Entropy and information theory”, Springer-Verlag, NY, 1990, 30 pages. |
Guha, et al., “Disambiguating people in search”, WWW04, AMC, May 2004, 9 pages. |
Guha, “Object co-identification on the semantic web”, WWW04, AMC, May 2004, 9 pages. |
Haveliwala, “Topic-sensitive pagerank”, Proceedings of the 11th international conference on World Wide Web, WWW '02, May 2002, 23 pages. |
Hogue, “Tree pattern inference and matching for wrapper induction on the world wide web”, Jun. 2004, 106 pages. |
Jeh, “Scaling Personalized Web Search”, May 20-24, 2003, 24 pages. |
Ji, “Re-ranking algorithms for name tagging”, Jun. 2006, 8 pags. |
Jones, “Bootstrapping for text learning tasks”, 1999, 12 pages. |
Koeller, “Approximate matching of textual domain attributes for information source integration”, Jun. 17, 2005, 10 pages. |
Kolodner, “Indexing and retrieval strategies for natural language fact retrieval”, ACM Trans. Database System 8.3, Sep. 1983, 31 pages. |
Kosseim, “Answer formulation for question-answering”, Oct. 1, 2007, 11 pages. |
Liu, “Mining data records in web pages”, Conference 00, ACM, 2000, 10 pages. |
Mackay, “Information theory, inference and learning algorithms”, Cambridge University Press, 2003, pp. 22-33, 138-140. |
Mann, et al., “Unsupervised personal name disambiguation”, Proceedings of the Seventh Conference on Natural Language Learning at HLT NAACL, 2003, 8 pages. |
Mccallum, et al., “Object consolidation by graph partitioning with a conditionally-trained distance metric”, SIGKDD 03, ACM, Aug. 2003, 6 pages. |
Mihalcea, et al., “PageRank on semantic networks with application to word sense disambiguation”, Aug. 2004, 7 pages. |
Mihalcea, et al., “TextRank: bringing order into texts”, Jul. 2004, 8 pages. |
Page, “The pagerank citation ranking: bringing order to the web”, Stanford Digital Libraries Working Paper, 1998, 17 pages. |
Pawson, “Sorting and grouping”, www.dpawson.uk.xsl/sect2/N6280.html, Feb. 7, 2004, 19 pages. |
Prager, et al., “IBM's piquant”, TREC2003, Nov. 2003, 10 pages. |
Prager, et al., “Question answering using constraint satisfaction: QA-by-dossier-withconstraints”, 2004, 8 pages. |
Ramakrishnan, et al., “Is question answering an acquired skill?”, WWW04, ACM, May 2004, 10 pages. |
U.S. Appl. No. 11/356,838, filed Feb. 17, 2006, Issued. |
U.S. Appl. No. 11/394,508, filed Mar. 31, 2006, Issued. |
U.S. Appl. No. 14/229,774, filed Mar. 28, 2014, Allowed. |
Number | Date | Country | |
---|---|---|---|
20170300524 A1 | Oct 2017 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 14229774 | Mar 2014 | US |
Child | 15637438 | US | |
Parent | 11394508 | Mar 2006 | US |
Child | 14229774 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 11356838 | Feb 2006 | US |
Child | 11394508 | US |