`Culliss
`
`[54] METHOD FOR ORGANIZING
`INFORMATION
`
`[76] Inventor: Gary Culliss, 9737 Larsen St.,
`Overland Park, Kans. 66214
`
`I21] APP1-NO-3 08/904,795
`[22] Filed
`Aug- 1’ 1997
`
`Related US. Application Data
`
`[63] Continuation-in-part of application No. 08/840,922, Apr. 25,
`1997~
`
`[51]
`
`Int. Cl? .................................................... .. G06F 17/30
`
`[52] U S C]
`
`707/5 707/10
`
`Fleld of Search ................................ ..
`
`3, 10, 2,
`707/1
`
`[56]
`
`References Cited
`
`U_S_ PATENT DOCUMENTS
`
`5,321,833
`
`6/1994 Chang et al. .
`
`US006006222A
`[11] Patent Number:
`[45] Date of Patent:
`
`6,006,222
`Dec. 21, 1999
`
`OTHER PUBLICATIONS
`Savoy, “A NeW Probabilistic Scheme for Information
`Retrieval in Hypertext”, NeW RevieW of Hypermedia and
`Multimedia, Applications and Research, vol. 1, pp. 107—34,
`1995.
`Fuhr et. al. Probabilistic Learning Approaches for indexing
`and retrieval With the TREC—2 Collection, TREC Text
`REtrieval Conference, pp. 67—74, 1993.
`Primary Examiner—Jack M. Choules
`Attorney) Agent) 0" F i" m—GarY Culliss
`[57]
`ABSTRACT
`
`A method of organizing information in Which the search
`activity of a user is monitored and such activity is used to
`organize articles in a subsequent Search by the same or
`another user Who enters a similar search query. The inven
`tion operates by assigning scores to articles under the key
`terms in the index. As users enter search queries and select
`articles, the scores are altered. The scores are then used in
`Subsequent Searches to Organize the articles that match a
`search query. As millions of people use the Internet, type in
`millions of search queries, and display or select from the
`many articles available over the Internet, the ranks the
`information available over the Internet through an evolu
`tionary process. The invention includes additional e'rnbodi
`ments WhlCh Incorporate category key terms and rating key
`terms
`
`
`
`5,535,382 5,754,939
`
`
`
`7/1996 Ogawa ........ .. 5/1998 Herz et al. ............................. .. 455/4.2
`
`46 Claims, 1 Drawing Sheet
`
`Receive First Search Query
`10 x from First User and Identify
`Related Articles
`
`7
`Present Articles Related to
`20 x First Search Query to First
`User
`
`V
`Allow First User to Select
`30 x One or More Articles
`
`7
`Alter Scores in Index
`40 x According to Selections Made
`by First User
`
`7
`Receive Search Query from
`50 x Second User
`
`7
`Present Articles Related to
`60 ’_\__ Second Search Query to
`Second User Ranked by
`Scores in Index
`
`Petitioner Apple Inc. - Exhibit 1006, p. 1
`
`
`
`U.S. Patent
`
`Dec. 21, 1999
`
`6,006,222
`
`Receive First Search Query
`10x
`from First User and Identify
`Related Articles
`
`20x
`
`Present Articles Related to
`First Search Query to First
`User
`
`30x
`
`Allow First User to Select
`One or More Articles
`
`i
`i
`l
`i
`i
`
`Alter Scores in Index
`According to Selections Made
`by First User
`
`Receive Search Query from
`50x
`Second User
`
`Present Articles Related to
`60x
`Second Search Query to
`Second User Ranked by
`Scores in Index
`
`FIGURE 1
`
`Petitioner Apple Inc. - Exhibit 1006, p. 2
`
`
`
`1
`METHOD FOR ORGANIZING
`INFORMATION
`
`6,006,222
`
`2
`-continued
`
`U.S. Pat. No.
`
`Title
`
`RELATED APPLICATION
`
`5,377,355
`
`METHOD AND APPARATUS FOR
`AUTOMATED PROCEDURE INIrIATION
`IN A DATA PROCESSING SYSTEM
`INCLUDING SOLICITING AN
`EvALUATION VOTE FROM USERS
`AUTOMATICALLY DETERMINED IN
`RESPONSE TO IDENTIFICATION OF A
`FUNCTIONAL AREA ASSOCIATED WITH
`A DOCUMENT;
`APPARATUS AND METHOD FOR
`FINDING RECORDS IN A DATABASE
`BY FORMULATING A QUERY USING
`EQUIVALENT TERMS WHICH
`CORRESPOND TO TERMS IN THE
`INPUT QUERY;
`HISTORICAL DATABASE TRAINING
`METHOD FOR NEURAL NETWORKS;
`USER INTERFACE SYSTEM AND
`METHOD FOR TRAvERSING A
`DATABASE;
`APPARATUS AND METHOD FOR
`DETERMINING DATA OF COMPOUND
`WORDS;
`METHOD AND APPARATUS FOR DATA
`MERGING/SORTING AND SEARCHING
`USING A PLURALITY OF BIr-SLICED
`PROCESSING UNIrS;
`METHOD AND APPARATUS FOR
`PROCESSING DATA BASE;
`
`10
`
`15
`
`5,404,507
`
`5,408,586
`
`5,408,655
`
`20
`
`5,371,676
`
`5,185,888
`
`4,967,341
`
`This patent application is a continuation-in-part of
`co-pending patent application, Ser. No. 08/840,922, ?led
`Apr. 25, 1997, also entitled “Method for Organizing Infor
`mation.”
`
`BACKGROUND OF THE INVENTION
`
`1. Related Disclosures
`This patent application contains subject matter disclosed
`in Disclosure Document Numbers 411,887; 417,369 and
`417,458.
`2. Field of the Invention
`The present invention relates to search engines, and more
`particularly pertains to a method for organiZing information
`by monitoring the search activity of users.
`3. Description of the Prior Art
`The Internet is an extensive netWork of computer systems
`Which alloWs a user to connect With various computer
`servers or systems. The Internet permits users to send and
`receive data betWeen computers connected to this netWork.
`This data may include Web sites, home pages, databases, text
`collections, audio, video or any other type of information
`made available over the Internet (collectively referred to as
`“articles”) from a computer server connected to the Internet.
`The articles may also include key terms representing
`selected portions of the information contained in the article.
`These key terms are available over the Internet to other
`computers and permit these other computers to locate the
`article.
`
`To locate articles on the Internet, a user of a remote
`computer searches for the key terms using a search program
`knoWn as a search engine. Examples of search engines
`currently available on the Internet include “Yahoo!” (TM),
`“Excite” (TM), and “Alta Vista” (TM). These programs
`alloW the remote user to type in one or more search terms,
`typically as a combination of English Words. The search
`terms may be connected by Boolean logic operators or may
`be truncated and combined With Wild card terms to form a
`search query. The search engine then compares the search
`query With the key terms from the articles and retrieves at
`least a portion of the articles having key terms Which match
`the search query. The search engine Will then display to the
`user the portion of the article such as the title. The user can
`then scroll through these retrieved portions of the articles
`and select a desired article.
`Conventional key Word searching and various prior art
`methods of accomplishing such key Word searching are
`disclosed in at least the folloWing patents:
`
`U.S. Pat. No.
`
`Title
`
`5,588,060
`
`5,546,390
`
`5,528,757
`
`METHOD AND APPARATUS FOR A
`KEY-MANAGEMENT SCHEME FOR
`INTERNET PROTOCOLS;
`METHOD AND APPARATUS FOR RADIX
`DECISION PACKET PROCESSING;
`ROUTING SYSTEM FOR RETRIEVING
`REQUESTED PROGRAM BY DISCARDING
`RECEIVED PROGRAM IDENTICAL WITH
`STORED PROGRAMS AND
`TRANSFERRING THE RECEIVED
`PROGRAM NOT IDENTICAL WITH
`STORED PROGRAMS;
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`A person Who places an article on the Internet typically
`intends for it to be available to all people Who type in search
`terms that are even remotely related to the subject matter of
`the article. This increases the exposure of the article to the
`public searching the Internet. Such increased exposure can
`potentially increase product sales or advertising revenue for
`the oWner of the article, especially if advertising revenue is
`related to the number of visits to the article or Web site. Thus,
`there is an incentive to use as many search terms as are
`possibly related to the article. In fact, some articles or search
`engines use every Word in the article as key terms. As a
`result, search engines Will retrieve many articles Which are
`unrelated to the subject matter Which the user desires to ?nd
`through a combination of search terms.
`Further, some Internet users are not skilled in selecting
`and connecting key Word search queries. These users Will
`often type in a single Word or tWo Words connected by an
`“AND” operator. These searches are extremely broad and
`often retrieve thousands of articles Which the user must
`manually search through by examining the title or other brief
`information about each article to ?nd the desired informa
`tion or advertisement.
`As the total number of articles posted on the Internet
`continues to increase, there is an increasing number of
`articles retrieved With each search query. To ?nd an article,
`the user must manually search through the retrieved articles
`and/or repeatedly modify the search query.
`
`OBJECTS AND SUMMARY OF THE
`INVENTION
`
`Accordingly, it is an object of the invention to organiZe
`articles available on the Internet.
`It is another object of the present invention to monitor
`searching activity to organiZe articles in accordance With the
`searching activity of one or more users.
`To accomplish these and other objects, the present inven
`tion generally comprises a method of organiZing information
`in Which the search activity of a user is monitored and such
`
`Petitioner Apple Inc. - Exhibit 1006, p. 3
`
`
`
`3
`activity is used to organize articles displayed in the search
`results. The invention operates by assigning scores to key
`terms and categories for articles. As users enter search
`queries and select articles, the scores are altered. The scores
`are then used in subsequent searches to organiZe the articles
`that match a search query. The method alloWs the search
`activity of a large number of Internet users to organiZe the
`information available over the Internet through an evolu
`tionary process.
`This brief description sets forth rather broadly the more
`important features of the present invention in order that the
`detailed description thereof that folloWs may be better
`understood, and in order that the present contributions to the
`art may be better appreciated. There are, of course, addi
`tional features of the invention that Will be described here
`inafter and Which Will be for the subject matter of the claims
`appended hereto.
`In this respect, before explaining a preferred embodiment
`of the invention in detail, it is understood that the invention
`is not limited in its application to the details of the method
`set forth in the folloWing description. The invention is
`capable of other embodiments and of being practiced and
`carried out in various Ways. Also, it is to be understood, that
`the phraseology and terminology employed herein are for
`the purpose of description and should not be regarded as
`limiting.
`As such, those skilled in the art Will appreciate that the
`conception, upon Which disclosure is based, may readily be
`utiliZed as a basis for designing other methods and systems
`for carrying out the objects and purposes of the present
`invention. It is important, therefore, that the claims be
`regarded as including such equivalent constructions insofar
`as they do not depart from the spirit and scope of the present
`invention.
`
`DESCRIPTION OF THE DRAWING
`
`FIG. 1 illustrates in How diagram form the operational
`steps taken by the system.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENTS
`
`The Internet is an extensive netWork of computer systems
`Which alloWs a user to connect With various computer
`servers or systems. The Internet permits users to send and
`receive data betWeen the computers connected to this net
`Work. The data can be read, vieWed or listened to on a
`broWser or other softWare program from over the Internet on
`a remote user’s computer. This data may comprise articles,
`databases, data collections, Web sites, Web pages, graphics,
`encryption, audio, video or any other type of information
`collectively referred to as articles and designated herein by
`the generic labels A1, A2, A3, etc.
`The present invention maintains an index of key Words,
`terms, data or identi?ers in English or other languages,
`computer code, or encryption Which are collectively referred
`to as key terms and represented herein by the generic labels
`Alpha, Beta, Gamma, Delta, Epsilon, etc.
`The articles are each associated With one or more of these
`key terms by any conceivable method of association, such as
`through indexing all Words or through meta-tag headers
`containing key Words selected by the author or editor.
`Further, a key term score is associated With each article for
`each of the key terms. For example, an initial index setting
`may look like this:
`
`6,006,222
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A1-1
`A2-1
`A3 - 1
`
`A1-1
`
`A1-1
`A3-1
`
`A2-1
`A3-1
`
`A1-1
`A3-1
`
`The invention Will accept a search query from a user and
`a search engine Will identify key terms Which match the
`search query. These key terms Which match the search query
`are called matched key terms. The search engine then
`identi?es in any conceivable manner the articles Which are
`associated With the matched key terms. This can be done by
`comparing all or part of the search query, or terms equivalent
`to those in the search query With the key terms in the index
`to identify the key terms Which match the search query. The
`search engine may account for Boolean logic operators in
`the search query.
`In the example above, and as illustrted at 10 in FIG. 1, a
`search query of “Alpha AND Gamma” Would identify
`articles A1 and A3 because they are both associated With the
`matched key terms Alpha and Gamma. Because articles A1
`and A3 are associated With the matched key terms, these
`articles are called matched articles.
`As shoWn in FIG. 1 at 20, the search engine Will then
`display a squib of each of the matched articles. The squib
`may comprise any portion, hypertext link to or representa
`tion of the matched article, such as the title, headings, ?rst
`feW lines of text, audio, video or any other type of infor
`mation. As shoWn in FIG. 1 at 30, the user can then scroll
`through the squibs of the articles and select a desired one of
`the matched articles by opening, retrieving, reading,
`viewing, listening to or otherWise closely inspecting the
`article from over the Internet or from any other storage area.
`The matched article selected by the user is called the
`selected matched article.
`Once the user has selected a matched article, and as
`shoWn in FIG. 1 at 40, the index can be altered such that the
`key term scores for the selected matched article under the
`matched key terms are altered relative to other key term
`scores. This indicates that the user believes that the matched
`key terms for that selected matched article are properly
`associated With the selected matched article. To alter the key
`term scores, a positive score such as (+1) can be added to the
`key term scores, for example, although any other possible
`indication can be used and the key term scores can be altered
`by any possible type of operation, mathematical or
`otherWise, to alter the key term scores for the selected
`matched article under the matched key terms relative to
`other key term scores.
`Thus, after executing the search query “Alpha AND
`Gamma,” the search engine Would display the squib of
`matched articles A1 and A3. If the user selected only article
`A3, the index could be altered such that the key term scores
`for the selected matched article A3 under the matched key
`terms Alpha and Gamma are altered relative to the other key
`term scores. The index Would then look like this:
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A1-1
`A2-1
`A3 - 2
`
`A1-1
`
`A1-1
`A3-2
`
`A2-1
`A3-1
`
`A1-1
`A3-1
`
`For the next search by either the same or a different user,
`the invention could then rank the matched articles by using
`
`15
`
`25
`
`35
`
`45
`
`55
`
`65
`
`Petitioner Apple Inc. - Exhibit 1006, p. 4
`
`
`
`6,006,222
`
`5
`the key term scores, as shown in FIG. 1 at 50 and 60. To this
`end, the key term scores of each matched article under each
`of the matched key terms of the neW search could then be
`associated in any possible manner to create a comparison
`score for each matched article. For example, the key term
`scores could be added, multiplied together or averaged to
`create the comparison score for that matched article. The
`matched articles can then be displayed to the user in order
`of comparison score superiority, such as by displaying the
`matched article With the highest comparison score ?rst.
`For example, the search query “Alpha AND Epsilon”
`Would again identify matched articles A1 and A3. By
`multiplying the key term scores of each matched article
`under each matched key term together to get the comparison
`score, the comparison score for article A1 Would equal 1,
`Whereas the comparison score for article A3 Would be 2. The
`invention Would then display the matched article A3 to the
`user in a superior position to matched article A1.
`
`DISPARATE SEARCH ACTIVITY
`To compensate for disparate search activity for certain
`articles relative to other articles, the invention may include
`a key term total score for each key term score of each article
`under each key term. After each search query is entered or
`after any other event occurs, the index could then be altered
`such that the key term total score of each matched article
`under each matched key term is altered relative to other key
`term total scores. The index could be altered in this manner
`after each search query is entered or after any other event,
`such as after the user has selected one or more articles or has
`read a matched article for a predetermined length of time.
`For example, the index could have an initial setting such
`as is shoWn here Where the key term scores are separated
`from the key term total scores by a backslash
`and given
`an initial value of one.
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A1 - 1/1
`
`A1 - 1/1
`
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1
`
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1
`
`As illustrated above, if the user selected only article A3
`after executing the search query “Alpha AND Gamma,” the
`key term score for article A3 under the matched key terms
`Alpha and Gamma Would be altered relative to other key
`term scores. Further, the key term total scores for both article
`A1 and article A3 under the matched key terms could also
`be altered. If the positive score is added to the key term
`scores for the selected matched article A3 under the matched
`key terms Alpha and Gamma, and the positive score is added
`to the key term total scores for the matched articles A1 and
`A3 (regardless of Whether they Were selected or not) under
`the matched key terms, the index Would then look like this:
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A1 - 1/2
`
`A1 - 1/1
`
`A2 - 1/1
`
`A3 - 2/2
`
`A1 - 1/2
`
`A3 - 2/2
`
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1
`
`For the next search, the invention could then organiZe or
`rank the articles by ?rst comparing the key term scores as
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`55
`
`60
`
`65
`
`6
`related to the respective key term total scores for each of the
`matched articles under each matched key term to create a
`key term probability score for that matched article under
`each respective matched key term from the neW search. To
`this end, the key term scores of each matched article under
`each of the matched key terms could be compared or
`associated With their respective key term total score in any
`knoWn manner to create the key term probability scores for
`that matched article under the respective matched key terms.
`For example, the key term scores of each matched article
`under the matched key terms could be divided by their
`respective key term total score to create the key term
`probability score of that matched article under that matched
`key term.
`The key term probability scores of each matched article
`under each of the matched key terms could then be associ
`ated in any knoWn manner to create the comparison score for
`each matched article identi?ed in the search. For example,
`the key term probability scores could be added, multiplied or
`averaged together to create the comparison score for that
`matched article. The articles can then be displayed to the
`user in order of superiority, such as by displaying the article
`With the highest comparison score ?rst.
`For example, the neW and different search query “Alpha
`AND Epsilon” Would again identify matched articles A1 and
`A3. The key term probability score for matched article A1
`under the key term Alpha Would equal the key term score of
`1 divided by the key term total score of 2, for a key term
`probability score of 0.5. Similarly, the key term probability
`score for matched article A2 under the key term Alpha Would
`equal the key term score of 2 divided by the key term total
`score of 2, for a key term probability score of 1. Under the
`key term Epsilon, the key term probability score for matched
`article A1 Would be 1, and the key term probability score for
`matched article A3 Would also be 1, as neither of these
`scores has been altered from the initial setting in this
`example.
`By multiplying the key term probability scores of each
`matched article under each matched key term together to get
`the comparison score, for example, the comparison score for
`article A1 Would equal 0.5, Whereas the comparison score
`for article A3 Would be 1. The invention could then display
`the article A3 to the user in a superior position to article A1
`because the comparison score for matched article A3 is
`higher.
`Increased Resolution:
`To provide for increased resolution in search ranking, the
`index may include matching associations of tWo or more key
`terms. For example, in the index illustrated beloW, each key
`term is grouped With one or more other key terms in a matrix
`format. Single key terms can be represented by a grouping
`of identical terms. Using the same initial settings from
`above, articles are listed in the boxes formed at the inter
`section of the roWs and columns of the matrix to indicate that
`such articles are associated With the intersecting key terms.
`Although the index is shoWn in pair groupings, the index can
`be extended to include triplicate or other associations as
`Well, i.e. separate boxes for Alpha-Beta-Gamma
`combinations, etc.
`The index shoWn beloW has empty boxes because some
`intersecting groupings are substantially equivalent to other
`intersecting groupings. As such, one of these intersecting
`groupings, i.e. Alpha-Beta or Beta-Alpha, can be left blank.
`Alternatively, the equivalent groupings could be used to
`record and distinguish betWeen the order of key terms in a
`search query. In other Words, a search query of “Alpha AND
`Beta” could include the grouping Alpha-Beta, Whereas the
`
`Petitioner Apple Inc. - Exhibit 1006, p. 5
`
`
`
`6,006,222
`
`8
`If the key term total score is also used for each key term
`score of each article and is altered every time a search query
`matches that key term grouping of that matched article, the
`index Would then look like this:
`
`7
`search query “Beta AND Alpha” Would include the grouping
`Beta-Alpha. In such case, the empty boxes in the matrix
`Would be used.
`In the example above, article A1 is the only article Which
`is associated With both the key terms Alpha and Beta.
`Accordingly, article A1 can be listed in at least the Alpha
`Alpha box, in the Alpha-Beta box, and in the Beta-Beta box,
`for example. Doing this for all key term groupings of the
`articles in the example above Would give an initial index that
`looked like this:
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`10
`
`Alpha
`
`A1 - 1/2 A1 - 1/1
`A2 - 1/1
`
`A1 - 1/2 A2 - 1/1 A1 - 1/1
`A3 - 2/2 A3 - 1/1 A3 - 1/1
`
`Index
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A1 - 1/1
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A1 - 1/1
`A3 - 1/1
`
`A2 - 1/1 A1 - 1/1
`A3 - 1/1 A3 - 1/1
`
`15
`
`A1 - 1/1
`
`A1 - 1/1
`
`A1 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1 A1 - 1/1
`
`A3 - 1/1
`
`A3 - 1/1
`A2 - 1/1 A3 - 1/1
`
`20
`
`A3 - 1/1
`
`A1 — 1/1
`A3 - 1/1
`
`Alpha
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`During a search, an entered search query Would typically
`include one or more key terms. The search engine could
`separate these key terms into one or more groupings. For
`example, the search query “Alpha AND Beta” could have
`only the one grouping Alpha-Beta, or could be separated into
`three groupings: Alpha-Alpha, Beta-Beta, and Alpha-Beta.
`For larger queries, the search query “Alpha AND Beta
`AND Gamma” could have three groupings: Alpha-Beta,
`Beta-Gamma, and Alpha-Gamma, but could additionally
`include the single groupings Alpha-Alpha, Beta-Beta, and
`Gamma-Gamma. As an additional example, a more complex
`query such as “Alpha AND (Beta OR Gamma)” could have
`the groupings Alpha-Beta and Alpha-Gamma, and could
`additionally include the single groupings Alpha-Alpha,
`Beta-Beta, and Gamma-Gamma.
`As described above, the invention Will then accept a
`search query from a user and a search engine Will identify
`articles Which are indexed With the key terms that match the
`search query. A search query of “Alpha AND Gamma”
`Would identify matched articles A1 and A3 because they are
`both indexed With the key term groupings Alpha-Alpha,
`Gamma-Gamma, and Alpha-Gamma. The key term group
`ings Which match the search query are called matched key
`term groupings. The search engine Will then display a squib
`of each of the matched articles. The user can then scroll
`through the squibs of the articles and select a desired one of
`the matched articles.
`Once the user has selected a matched article, the key term
`scores for the selected matched article under the matched
`key term groupings can be altered to indicate that the user
`believes that those matched key term groupings are properly
`associated With the selected matched article. To alter the key
`term scores, for example, the positive score can be added to
`the key term scores, although any other possible type of
`indication can be used.
`If the user selected only article A3, the key term scores for
`selected matched article A3 under the matched key term
`groupings Alpha-Gamma Would be altered. Additionally, the
`key term scores for selected matched article A3 under the
`matched key term groupings Alpha-Alpha and Gamma
`Gamma could also be altered since the key terms Alpha and
`Gamma are each represented individually in the results of
`the search query.
`
`25
`
`30
`
`35
`
`40
`
`45
`
`55
`
`60
`
`65
`
`Beta
`
`Gamma
`
`Delta
`
`Epsilon
`
`A3 - 2/2
`
`A1 - 1/1
`
`A1 - 1/1
`
`A1 - 1/1
`
`A1 - 1/2 A3 - 1/1 A1 - 1/1
`
`A3 - 2/2
`
`A3 - 1/1
`A2 - 1/1 A3 - 1/1
`
`A3 - 1/1
`
`A1 — 1/1
`A3 - 1/1
`
`As explained above, the invention could then organiZe the
`articles by ?rst comparing the key term scores as related to
`the key term total scores for each of the matched articles
`under each matched key term grouping. To this end, the key
`term scores of each matched article under each of the
`matched key term groupings could then be compared or
`associated With the respective key term total score in any
`knoWn manner to create the key term probability score for
`that matched article under each matched key term grouping.
`For example, the key term scores of each matched article
`under the matched key term groupings could be divided by
`the respective key term total score to create the key term
`probability score for that matched article under those respec
`tive key term groupings.
`The key term probability scores of each matched article
`under each of the matched key term groupings could then be
`associated in any knoWn manner to create the comparison
`score for each matched article. For example, the key term
`probability scores could be multiplied together to create the
`comparison score for that matched article for the search
`query. The matched articles can then be displayed to the user
`in order of superiority, such as by displaying the matched
`article With the highest comparison score ?rst.
`For example, a neW user entering the search query “Alpha
`AND Gamma” Would result in the search engine again
`identifying matched articles A1 and A3. Under the key term
`grouping Alpha-Alpha, the key term probability score for
`matched article A1 Would equal the key term score of 1
`divided by the key term total score of 2, for a key term
`probability score of 0.5. Also, the key term probability score
`for selected matched article A2 under the key term grouping
`Alpha-Alpha Would equal the key term score of 2 divided by
`the key term total score of 2, for a key term probability score
`of 1.
`Similarly, under the key term grouping Gamma-Gamma,
`the key term probability score for matched article A1 Would
`be 0.5, and the key term probability score for selected
`matched article A3 Would equal 1. Under the key term
`grouping Alpha-Gamma, the key term probability score for
`the matched article A1 Would equal 0.5, and the key term
`probability score for selected matched article A3 Would
`equal 1.
`By multiplying the key term probability scores for each
`key term of each matched article together to get the com
`parison score, for example, the comparison score for article
`A1 Would equal 0.5 times 0.5 times 0.5, for a total of 0.125.
`The comparison score for article A3, on the other hand,
`Would equal 1 times 1 times 1, for a total of 1. The invention
`could then display the article A3 to the user in a superior
`
`Petitioner Apple Inc. - Exhibit 1006, p. 6
`
`
`
`6,006,222
`
`position to article A1 because the comparison score for
`matched article A3 is higher.
`
`CATEGORIES
`The invention can also be used to organize articles by
`category. To this end, the key terms of the index may simply
`comprise category key terms represented by the generic
`
`5
`
`10
`terms can have a similar Weight as other key terms or may
`be increasingly or decreasingly Weighted to represent rela
`tive importance of the categories to the search query. An
`example of such an index Wherein all articles are initially
`equally ranked in all categories and under all key terms is
`shoWn beloW.
`
`Index
`
`Alpha
`
`Beta
`
`Gamma Delta
`
`Epsilon
`
`CAT1 CAT 2
`
`CAT3
`
`A1-1/1 A1-1/1 A1-1/1 A2-1/1 A1-1/1
`A2-1/1
`A3-1/1 A3-1/1 A3-1/1
`A3-1/1
`
`A1-1/1
`A1-1/1 A1-1/1
`A1-1/1 A3-1/1 A1-1/1
`A3-1/1
`A3-1/1
`A2-1/1 A3-1/1
`A3-1/1
`
`A1-1/1
`A3-1/1
`
`Alpha
`
`Beta
`Gamma
`
`Delta
`
`Epsilon
`
`CAT1
`
`CAT2
`
`CAT3
`
`A1-1/1 A1-1/1 A1-1/1
`A2-1/1 A2-1/1 A2-1/1
`A3-1/1 A3-1/1 A3-1/1
`A1-1/1 A1-1/1 A1-1/1
`A1-1/1 A1-1/1 A1-1/1
`A3-1/1 A3-1/1 A3-1/1
`A2-1/1 A2-1/1 A2-1/1
`A3-1/1 A3-1/1 A3-1/1
`A1-1/1 A1-1/1 A1-1/1
`A3-1/1 A3-1/1 A3-1/1
`A1-1/1 A1-1/1 A1-1/1
`A2-1/1 A2-1/1 A2-1/1
`A3-1/1 A3-1/1 A3-1/1
`A1-1/1 A1-1/1
`A2-1/1 A2-1/1
`A3-1/1 A3-1/1
`A1-1/1
`A2-1/1
`A3-1/1
`
`labels CAT1, CAT2, CAT3, CAT4, etc. The articles can each
`be associated With one or more of these category key terms,
`and the key term score is associated With each article for
`each of the category key terms. Additionally, the index may
`also include the key term total score for each category key
`term score of each article, as described above for the key
`terms.
`
`35
`
`For example, an initial index setting may look like this:
`
`Index
`
`CAT 1
`
`CAT 2
`
`CAT3
`
`CAT4
`
`CAT 5
`
`A1 - 1/1
`
`A2 - 1/1
`
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1
`
`A2 - 1/1
`
`A3 - 1/1
`
`A1 - 1/1
`
`A3 - 1/1
`
`45
`
`This embodiment of the invention, operating separately
`from or in addition to the embodiments described above,
`Would permit the user to enter or select a category key term
`for inclusion in the search query. In this embodiment, the
`invention Would operate in a similar manner for the category
`key terms as described above for the key terms alone. The
`invention may alloW a user to enter one or more category
`key terms in formulating a search. For example, the user
`may enter the category key terms “Apartments” and “Los
`Angeles” or the category key terms “Romantic” and “Com
`edy” to ?nd articles (i.e. advertisements or movies) Which
`fall under tWo or more category key terms.
`
`55
`
`60
`
`Moreover, the category key terms can be incorporated
`into the index of key terms as just another key term and
`included in the association of the comparison score and, if
`used, the key term probability scores. The category key
`
`This embodiment of the invention Works in a substantially
`similar manner as the key term groupings described above,
`except that the key term groupings may also include a
`category key term. For example, the search query “CAT1
`AND Beta” could include just the one grouping CAT1-Beta,
`or could be separated into three groupings: CAT1-CAT1,
`Beta-Beta, and CAT1-Beta.
`In yet another embodiment of the invention, the category
`key terms can be incorporated into one side of the index of
`key terms and associated With the key terms in the index to
`form the key term groupings. In this embodiment, the
`category key terms each function as just another key term to
`form the key term groupings and are included in the asso
`ciation of the comparison score and, if use