Lumen View Technology v. Findthebest.com

7/28/2019 Lumen View Technology v. Findthebest.com

1/29

IN THE UNITED STATES DISTRICT COURTFOR THE SOUTHERN DISTRICT OF NEW YORK

LUMEN VIEW TECHNOLOGY LLC,

Plaintiff,

v.

FINDTHEBEST.COM, INC.

Defendant.

Civil Case No.

JURY TR IAL DEMANDED

COMPLA INT

Plaintiff Lumen View Technology LLC ("Plaintiff'), for its Complaint against DefendantFindTheBest.com, Inc. ("Defendant"), hereby alleges as follows:

PART IES

1. Plaintiff is a Delaware limited liability company.2. Upon information and belief, Defendant is a Delaware corporation having a

principal place of business at 101 Innovation Place, #A, Santa Barbara, California 93108. Uponinformation and belief, Defendant may be served with process through Thomas N. Harding, SeedMackall LLP, at 1332 Anacapa St., Suite 200Santa Barbara, California 93101.

NATURE OF THE ACTION

3. This is a civil action for the infringement of United States Patent No. 8,069,073(the '"073 Patent") under the Patent Laws of the United States, 35 U.S.C. 1 et seq.

Case 1:13-cv-03599-UA Document 1 Filed 05/29/13 Page 1 of 29


2/29

JURI SD ICT ION AND VENUE

4. This Court has jurisdiction over the subject matter of this action pursuant to 28U.S.C. 1331 and 1338(a) because this action arises under the Patent Laws of the UnitedStates, 35 U.S.C. 271 et seq.

5. This Court has personal jurisdiction over Defendant because Defendant isregistered to conduct business in New York and has purposely availed itselfof the privileges andbenefits o f th e l aws o f t he S t at e o f New York.

6. Upon information and belief, more specifically, Defendant, directly and/orthrough authorized intermediaries, ships, distributes, offers for sale, sells, and/or advertises(including the provision of an interactive web page) its products and services in the United Statesand the State of New York. Upon information and belief, Defendant has committed patentinfringement in the State of New York. Defendant solicits customers in the State of New York.Defendant has many paying customers who are residents of the State ofNew York and who eachuse Defendant's products and services in the State ofNew York.

7. Venue is proper in this judicial district as to Defendant pursuant to 28 U.S.C. 1391 and 1400(b).

THE PATENT - IN SUIT

8. Paragraphs 1-7 are incorporated by reference as if fully set forth herein.9. On November 29, 2011, the '073 Patent entitled "System And Method For

Facilitating Bilateral And Multilateral Decision-Making" was duly and lawfully issued by theUnited States Patent and Trademark Office ("PTO"). The '073 Patent is attached hereto asExhi bi t A .



3/29

10. Plaintiff is the exclusive licensee of the '073 Patent and possesses all rights ofrecovery under the '073 Patent, including the right to sue and recover all damages forinfringement thereof, including past infringement.

COUNT I - PATENT INFR INGEMENT

11. Paragraphs 1-11 are incorporated by reference as if fully restated herein.12. Upon information and belief and in violation of 35 U.S.C. 271(a), Defendant

has infringed and continues to infringe one or more claims of the '073 Patent by making, using,providing, offering to sell, and selling (directly or through intermediaries), in this district andelsewhere in the United States, a computer implemented method for facilitating evaluation, inconnection with the procurement or delivery of products or services, in a context of at least oneof a financial transaction and operation of an enterprise, such context involving a first class ofparties in a first role and a second class of counterparties in a second role. More specifically, andby way of non-limiting example, Defendant offers recommendation services via thewww.findthebest.com website ("Defendant Website") for individuals seeking the purchase ofproducts and individuals offering to sell the products.

13. For purposes of the '073 Patent, the Defendant Website executes a computer-implemented method for facilitating evaluation, in connection with the procurement or deliveryof products or services, in a context of the operation of an enterprise, such context involving afirst class of parties (e.g., consumers) in a first role and a second class of counterparties in asecond role (e.g. , individuals sel ling goods). The Defendant Website retrieves first preferencedata from a digital storage medium, the first preference data received from the individual(s)registering on and/or using the Defendant Website, and assigns attribute levels based on the



4/29

choices made by the individuals (first class of parties). The Defendant Website retrieves thesecond preference data from a digital storage medium, the second preference data received fromthe individuals registering on and/or using the Defendant Website, and assigns attribute levelsbased on the choices made by the individuals (counterparties). The Defendant Website performsmultilateral analysesof the selected party's preference data, e.g., an individual seeking products,and the preference data for each of the counterparties, e.g., individual's selling products, andcomputes a closeness-of-fit value based thereon, namely by way of their AssistMe service.Further, the Defendant Website uses the closeness-of-fit value to derive and provides a list to theselected party matching at least one of the counterparties, e.g., a list matching the individualseeking products with individual's offering products for sale. Namely, the Defendant Websiteprovides "AssistMe Results," allowing consumers to see a list of "products best fit [for their]requirements."

14. To the extent such notice may be required, Defendant received actual notice of itsinfringement of the '073 Patent at least as early as the filing of the original complaint in thisaction, pursuant to 35 U.S.C. 287(a).

15. Defendant's aforesaid activities have been, intentional , without authori ty and/orlicense from Plaintiff.

16. Plaintiff is entitled to recover from the Defendant the damages sustained byPlaintiff as a result of the Defendant 's wrongful acts in an amount subject to proof at trial, which,by law, cannot be less than a reasonable royalty, together with interest and costs as fixed by thisCourt under 35 U.S.C. 284.



5/29

17. Defendant's infringement of Plaintiffs exclusive rights under the '073 Patent willcontinue to damage Plaintiff, causing irreparable harm for which there is no adequate remedy atlaw, unless enjoined by this Court.

PRAYER FOR R E LIE F

WHEREFORE, Plaintiff Lumen View Technology LLC respectfully requests that thisCourt enter judgment against Defendant FindTheBest.com, Inc. as follows:

A. An adjudication that Defendant has infringed the '073 Patent;B. An award of damages to be paid by Defendant adequate to compensate Plaintiff

for its past infringement and any continuing or future infringement up until thedate such judgment is entered, including interest, costs, and disbursements asjustified under 35 U.S.C. 284 and, if necessary to adequately compensatePlaintiff for Defendant's infringement, an accounting of all infringing salesincluding, but not limited to, those sales not presented at trial;

C. A declaration that this case is exceptional under 35 U.S.C. 285;D. An award to Plaintiff of its attorney fees, costs, and expenses incurred in

prosecuting this action; andE. An award to Plaintiff of such further relief at law or in equity as the Court deems

just and proper.



6/29

DEMAND FOR JURY TR IAL

Plaintiff hereby demands trial byjury on all claims and issues so triable.Dated: May 29, 2013 AETON LAW PARTNERS LLP

tP4jU*JdCL^ Jf^^t^J^Damian Wasserbauer (DW3507)[email protected] Centerpoint Drive. Suite 105Middletown, CT 06457Telephone: (860)724-2160Counsel for PlaintiffLumen View Technology LLC



7/29

Exhib i t A

United States Patent No. 8,069,073



8/29

(i2) United States PatentShapiro et al.

(54) SYSTEM ANDMETHOD FOR FACILITATINGB ILATERAL AND MULTILATERALDECIS ION-MAKING

(75) inventors: Eileen C. Shapiro, Cambridge, MA(US); Steven J. Mintz, Saddle River, NJ(US)

(73) Assignee: Dalton Sentry, LLC, Wilmington, DE(US)

( * ) Notice: Subject to any disclaimer, the term of thispatent is extended or adjusted under 35U.S.C. 154(b) by 0 days.This patent is subject to a terminal disclaimer.

(21) Appl.No.: 12/754,291(22) Filed: Apr. 5,2010(65) Prior Publication Data

US 2010/0191664A1 Jul . 29, 2010Related U.S. Application Data

(63) Continuation of application No. 11/711,249, filed onFeb. 27, 2007, now Pat. No. 7,725,347, which is acontinuation of application No. 11/171,082, filed onJun. 29, 2005, now Pat. No. 7,184,968, which is acontinuation of application No. 09/538,556, filed onMar. 29, 2000, now Pat. No. 6,915.269.

(60) Provisional application No. 60/173,259, filed on Dec.23, 1999.

US008069073B2

(io) Patent No.:(45) Date of Patent:

US 8,069,073 B2*Nov.29,2011

(51) Int. CI.G06Q10/00 (2006.01)(52) U.S.CI. 705/7.14; 705/327; 705/321

(58) Field of Classification Search 705/7.14,705/327, 32 1See application file for complete search history.

(56) Re f e rence s C i t edU.S. PATENT DOCUMENTS

4.258.351 A * 3/1981 Shigetaetal 340/9426.711.522 B2 * 3/2004 Shiraietal 702/179

OTHER PUBLICATIONS

Gilbert, "The Myth of Fingerprints", Stumbling on Happiness,pp.245-257(2006).Lehrer. "Choking on Thought",How WeDecide, pp. 133-166 (2009).* cited by examinerPrimary Examiner ~ Johnna Loftis(74) Attorney. Agent, or FirmSunstein Kann Murphy &T imbe r s LLP(57) AB S T R ACTTechniques for facilitating evaluation, in connection with theprocurement or delivery ofproducts or services, in a contextof at least one of (i) a financial transaction and (ii) operationof an enterpr is e, are d isc losed. The techn iques involveretrieving party and counterparty preference data from digitalstorage media; performing multilateral analyses of the combined preference data by computing a closencss-of-fit value;and delivering a list matching the at least one party and the atleast one counterparty using the computed closeness-of-fitvalues.

9 Claims, 11 Drawing Sheets

Obtainfromnn&partyma flrflghagfflipflnireitoq fan fft oftpmidQfl.ddd ffiPO

I umBmdckrarapimscADA

Derive Budpenamieprofile

Dbwq a (HfjArJ[nbtsn

AtttBTCfi piwtjraee pnnflm, todsrivt tifl,lot eachpBrty, ofralarJrelyolo fitofprtftnficoi

U

Comrmim'ciHBOu nfobtalp"nttJaj)Qnutfmpar&am evtt



9/29

U.S. Pa t en t Nov. 29, 2011 Shee t 1 o f 11

Obtain from each party in a firstclass responses to a first set of

questions, and store11

Obtain from each counterparty ina second class responses to asecond set of questions, and store

11

Derive a first preference profilefor each ofparties

11

Derive a second preferenceprofile for each of counterparties

14

Analyze preference profiles, toderive a list, for each party, ofcounterparties providing a

relatively close fit of preferences15

Communicate list of closelyfitting counterparties to eachparty

16

US 8,069,073 B2

F IG . l



10/29

U.S. Paten t Nov. 29,2011 Sheet 2 o f 11 US 8,069,073 B2

FIG. 2



11/29

U.S. Paten t Nov. 29 , 2011 Sheet 3 o f 11 US 8,069,073 B2

First Class ofParties301

1st Question andResponse Module-ls t Set ofQuestions

302

1st Storage Medium303

1st Profi le Processor307

Closeness-of-FitAnalyzer

309

Output

FIG. 3

Second Class ofCounterparties

304

2nd Question andResponse Module-2nd Set ofQuestions

305

2nd Storage Medium306

2nd Profile Processor308



12/29

U.S. Patent Nov. 29,2011 Sheet 4ofll

Primary Data Entry401

Supplemental Data Entry402

Preference Form Completion403

Co-Evaluator Preference Form Completion404

Preference Form Review/Correction405

Review; Revision; Matching Authorization406

Data Release Authorization407

Relatively Close Fits List Disseminated408

Party/Counterparty Action;Matching Facilitation

409

US 8,069,073 B2

FIG. 4



13/29

U.S. Patent Nov. 29,2011 Sheet 5of11 US 8,069,073 B2

Decision Satisfaction Inquiry511

Update Option513

Match-Again Option514

FIG. 5



14/29

Applicant

FIG. 6

overall team performance Ilower base pay with high options

unclear and unbounded rolesmaximize financial returns exclusively

fix problems quicklyrecognition based on power

amorphous structure for getting resourcesincomplete control over significant resources

long work weeksocializing expected in off hours

dea l with external customersfrequent overnight travel

3

z

4

O

ernre

a

0 0ONo-J



15/29

Company

unclear an d unbounded roleslower base pay with high optionsoverall team performancefix problems quickly

maximize financial returns exclusivelyrecognition based on power

amorphous structure for getting resourceslong work week

incomplete control over significant resourcessocializing expected in off hours

dea l with ex te rna l cu s tomer sfrequent overnight travel

FIG. 7

18 20

P

so

S3

O

STCD

-J

c0 0O

w



16/29

Applicantoverall team performancelower base pay with high options

unc l ea r and unbounded ro lesmaximize financial returns exclusively

fix problems quickly_recognition based on poweramorphous structurefor getting resources

incomplete control over significant resourceslongworkweek

socializing expected in off hours_dea l with external customersfrequent overnight travel

FIG. 8

0Company

un c le a r a n d unbounded ro l e slower base pay with high options

overall team performancefix problems quickly

maximize financial returns exclusivelyrecognition based on power

amorphous structure for getting resourceslongwork weekincomplete control over significant resources __socializing expected inoff hours

dea l with external customersfrequent overnight travel

0

8 10 12 14 16 18 20

10 12 14 16 18 20

asfb53

2

O

4.

C00On

W



17/29

U.S. Patent Nov. 29, 2011 Sheet 9of11 US 8,069,073 B2

iAeo jHb

Type the number next to your cho ice , assuming every th ing e l se t o b e eipial ;

1 DECISIONS: Bui ld ing eonsenus i s p r i o r i t y , even if t ine -consuming2 DECISIONS: f ix problems i s 1 st , even i f consensus damage occurs

[yijo nunboK ESC. to -hnv.h uji CTRL END to |i< i t

FIG. 9



18/29

U.S. Patent Nov. 29, 2011 Sheet 10 of11 US 8,069,073 B2

Opt ion ft IS BETTER f o r yo u t it an Opt io n B. Assuming t h a t every th inge l s e is equa l how much be t t e r do yo u f e e l Opt ion ft is ?Type a numbe r f rom the i to 4 sca le be lou .

A: DECISIONS: Fix problems i s 1 s t . even if consensus danage occu r s

B: DECISIONS: Bui ld ing eonsenus i s p r io r i t y , even if t i ne -consuning

I MUST HftUE Option fi ; Opt ion B would be extremely unsa t i s f ac t o ry .1 GKEnlLV PREFER Option R over Option B.1 MODEF.fiTELV PREFER Opt ion A over Option B.I JUST SLIGHTLY PREPER Opt ion ft over Opt ion B.

Ty pe n umKir ,BU S

"I _ ESC t o. . b ac jt u p i CIHL END to q u i t" ' i-'" r,' '':"' T - : ,;.-...;/ ,- , v .- ,; ,. -. .; '- ; , .

FIG. 10



19/29

U.S. Pa ten t Nov. 29,2011 Sheet 11 of 11 US 8,069,073 B2

Acq.exe W HCOMPARE JOB "A " TO JOB "B " AND IHEN TYPE ft NUMBER FROM THE "1" TO "9"SCALE SHOWN AT THE BOTTOM OF THE SCREEN TO INDICATE YOl'R PREFERENCE.


20/29

US 8,069,073 B21 2

SYSTEMAND METHOD FOR FACILITATING In alternativeembodiments, the list may beranked accord-Bn.ATERAL AND MULTILATERAL ingto closeness of fit.

DECISION-MAKING In alternative embodiments, the method may furtherinvolve supplying to at (eastone party or counterparty co-CROSS-REFERENCE TORELATED -s cvaluator a scries of forced choice questions so as to elicitAPPLICATIONS co-evaluator responses, wherein the list matches the at leastonepartyand theat leastonecounterpartyaccordingtoanaly-This application is a continuation of our application Ser. sisof preference profiles determined usingsuchco-evaluator

No. 11/711,249, filed Feb. 27, 2007, which is a con ti nuat ion responses.ofourapplication Ser. No. 11/171,082, tiled Jun. 29, 2005 I0 ,n alternative embodiments, the party responses may(U.S. Pat. No. 7.184,968), which is a continuation of our reveal, with respect loeach level of eachof a first series ofapplication Ser. No. 09/538,556. filed Mar. 29, 2000 (U.S. attributes, a utility value which indicates the value that thePat. No. 6,915,269), which claims the benefit of our provi- P3^Places on the level ofthe attribute. Such party responsessional application Ser. No. 60/173,259, filed Dec. 23. 1999; mav reveal me utlhty vatues wimout the utiIitv values beingthese related applications are hereby incorporated herein bv pnmded ^P11*Additionally the counterparty responsesreference " ma? revea1, Wltn resPecl I0 eacn level * each ofa secondseries of attributes that complements the first series ofTFCHNICAI FIFI D attributes, a utility value which indicates the value that thecounterparty places on the level of the attribute. Such counterparty responses may reveal the utility values without theutility values being provided explicitly.he present invention relates to bilateral and multilateralevaluation methods and systems. ,n altemalive embodiments, the series of forced choiceHArkrrrn rwn questions and/or the list may beobtained froma remote serverover a communica t ion ne twork. The series of forced choice

25 questions may be supplied by making available a set of webConsumers constantly decidewhich products and services pages providmg a se, 0fquestions and permitting entry ofbest satisfy their needs and desires. Producers correspond- responses thereto. The party responses and the counterpartyingly decide how best to configure their products and ser- responses may beprovided to a remote serverover acommu-vices, from amongst a wide array of choices. They must not nication network,only choose a sui table pr ice, but al so must de cid e whi ch 30combinationof other attributes of their products and services BRIEF DESCRIPTION OF THE DRAWINGSwill best satisfy consumers.In order to facilitate these decisions, there have therefore The foregoing features of the invention will be morearisen a varie ty of marke ting resea rch techn iques. Among read ily understood by reference to the fol lowing detai ledthese are forced trade-off or forced choice methodologies, 35 description, takenwith reference to the accompanyingdraw-including conjoint analysis. Through statistical methods, ings. inwhich;thesetechniquesallowpredict ionofwhichattr ibutesofprod- FIG. 1 shows a block diagram of an embodiment of aucts and services are relatively more and less valuable to a method inaccordance with the present invention for facilitat-given group of constituents. ing bilateral and multilateral decision-making;Based on these conventional techniques, producers of ~> FIG. 2 shows a block diagram of a further embodimentofgoodsand services are able to model buyers' or users' pref- a method inaccordance with the present invention in whicherences, thereby facilitating design or selection of product s con jo int ana lysi s is employed;andprocesses that best satisfy those preferences, l-'or persons FIG. 3 shows a block diagram of an embodiment of aon two sides of a transaction (a producer and a group of system in accordance with the present invention;consumers, for example), conventional techniques permit 45 FIGS. 4 and 5 illustrate the logical flowof amethod accord-persons on one side of the transaction to model the prefer- ing to an embodiment of the invention, that may be imple-ences of a group of constituents on the other side of the mented using a web server on the Internet;transaction. Conventionaltechniques may therefore becalled FIGS. 6and 7are histogram representationsofa preferenceunilateral, or one-sided, evaluation techniques. profile of a party who isa job applicant and ofa counterparty

50 employer in accordance with an embodiment of the inven-SUMMARY OF THE INVENTION fion;

FIG. 8 presents a side-by-side comparison of the prefer-In accordance with one aspect of the invention there is enceprofilesof FIGS.6 and 7; andprovided a methodfor facilitating evaluation, inconnection FIGS.9,10, and 11arescreenshotsdemonstrating hierar-withtheprocurement ordelivery of productsor services,ina 55 chicallystructuredquestions organized into three stages incontext of at least one of (i) a f inancial t ransac tion and (ii) accordance with an embodiment of the invention,

operation of an enterprise, such context involving a memberof a first class of part ies in a first role and a member of a DETAILED DESCRIPTION OF SPECIFICsecond class of counterparties in a second role. The method EMBODIMENTSinvolves supplying to at least one of the parlies a series of 60forced cho ice quest ions so as to elicit par ty responses; sup- By contras t with convent iona l methods , embodiments ofplying to at least one of the counterparties a series of forced the present invention enable a bilateral evaluation of prefer-cho iceques tions so as to e li ci t counterpa rty responses; and ences: a dec is ion is recommended based on its p roviding adelivering a listmatching the at least one part y and the at least relatively close fit between the preferences of each potentialone counterparty according to analysis of preference profiles 65 pairing of party and counterparty to a potential transaction,determined using conjointanalysis of the party responses and when compared with other possible pairs of parties to thethe counterparty responses. potential transaction. Indeed, embodiments of the present



21/29

US 8,069 ,073 B2inventionmay likewisebe employedwheninformationaboutpreferences is providednotjust by twopartiesto the transaction (a party and a counterparty), but also by at least oneco-evaluator, whoprovidesa useful perspective on the preferences ofa party or a counterparty. Inthis case, the evalua- 5tion is multilateral rather than bilateral.Invarious embodimentsof the present invention, there canbe employedquestions that require a forcedchoice to revealpreferences of the respondent. The benefit of the forcedchoice approachis that it helps to uncoverunderlying prefer- 10ences thatare hiddenand sometimes notconsciously evidenteven to the respondent.In this connection embodiments of the invention mayemploy conjoint analysis. See for example, Cattin, P.and R.R. Wittink, "Commercial Use of Conjoint Analysis: A Sur- 15vey", 45 JournalofMarketing 44-53 (No. 3,Summer, 1982),and "Commercial UseofConjointAnalysis: An Update", 53JournalofMarketing 91-96 (My, 1982): Green, P.E. and Y.Wind, "New Way to Measure Consumers' Judgments,"HarvardBusiness Review, July 1975 ("Green and Wind"); see 20also thereferences identified in the extensive bibliographyofPatrickBohl:ConjointLiterature DatabaseCLD,Universityof Mainz. Germany, 1997. The foregoing articles and references are hereby incorporated herein by reference.As used in this description and the accompanying claims, 25the following terms shall have the meanings indicated, unless

the context otherwise requires:The term "party" includes a natural person or an entity,wherein an entity may be any association, organization, orgovernmental agency. A "counterparty"is similarly any other 10natural person or an entity.

A "f inancial t ransact ion" is a t r ansac tion in which se rv icesor products are being procured or delivered under circumstances involving an expectation that they will be paid for.Thus "financial transactions" include enrollment at a college 35or universityor a privateschool (wherein educational servicesare rendered for tuition), employment by an entity (whereinan employee's servi ces arc rendered for payment by theemployer), engagement of a physician or health maintenanceorganization (wherein health care services are provided for 40compensation), choosing a retirement community, investingin a mutual iund, taking a vacation, or inexecuting a mergeror joint venture or acquisi tion. The terms "services" and"products" include the singularas well as the plural.An "enterprise" is a business organization (regardless of 45form), a government agency or organ, or a non-profit-organization (including a religious, scientific, or charitable organization)."Attributes" of a product or service include characteristics,features, and benefits of the product or service. Ilence (asan 50example) if the service is college education, attributes mayinclude the size of the school, the prestigeof the school, andthe degree of structure of the school's educational program.A "level" o f an a t tr ibute is a value associated with th eattribute that pertains to a characteristic, feature or benefit of 55a product or service. The value may, but need not, be quantitative; the value may be categorical. Hence if the service iscollege education, the levelof the attribute "school size" maybe quantitative, as for example. "9378 students", or may becategorical, as for example, "between 5,000 and 10,000." 60Attribute levels may be categorizedevenwhen more abstractattributes are involved. For example, if the attribute is prestige, a level may be "widely viewed as highly prestigious"; ifthe attribute is degreeof structurein the education program, alevel may be "low degree of structure". 65FIG. 1 is a block diagramof an embodiment of amethod inaccordancewith thepresent invention for facilitating bilateral

and multilateral decision-making. In this embodiment, sixactivities are involved. As shown in item 11, first there isobtainedfromeachparryina first classresponsestoa first setofquestions, andthe responses are stored ina suitable digitalstoragemedium. Also, in item 12, there isobtained from eachcounterparty in a second class responses to a second set ofquestions, and the responses are stored in a suitable digitalstoragemedium. (Wediscuss thenatureof suitablequestionsin connectionwith later figures.)(Note items 11 and 12 neednotbe contemporaneous and need not be sequencedin anyorder.) In item 13, a first preference profile is generated foreach party, based on the party's responses to the first set ofquestions; and, similarly, a second preference profile is generated, in item 14, for each counterparty, based on the counterparty's responses to the second set of questions. The questions and the resulting profiles may be developed using any ofa wide range of approaches. In some embodiments, asdescribed below,there may be employedconjoint analysis orother forced-choice methodologies. It is within the scope ofthe invention to utilize a firstmethodology in connectionwiththe first set of questions and a second methodology in connection with the second set of questions. Once these preference profiles have been generated, the method next analyzes,for each party in the first class, the preference profiles ofcounterparties inthe secondclass, and derives a ranked list ofcounterparties that provide the closest fit of preferences withthat party, as compared with the fitof all counterparties inthesecond class (item 15). Finally, in item 16, the list of closestfitting counterparties for each party is communicated to thatparty. (Similarly for each counterparty, the method derives aranked listof parties that provide the closest fitof preferenceswith that counterparty, as compared with the fit of all partiesin the first class; and the list of closest fitting parties for eachcounterparty is communicated to that counterparty.) By providing such a list in each case, based on a bilateral or multilateral preference analysis, the method facilitates parties andcoun terparties in mak ing decis ions that arc based on thecloseness of the fit between their preferences.FIG. 2 shows one embodimentofa method according totheinvention, in which, first, preference profiles for parties andcounterparties are generated using conjoint analysis techniques. Conventional conjoint analysis techniques, used in aunilateral fashion, are described in th e references describedabove near the beginning of this section of the description.

Once preference profileshave been generatedaccording tothe embodiment ofFIG. 2. they are then used to recommendtoeach party a setofcounterpartieswho provide the closestfitof preferences amongst the counterparties considered. Sincethe embodiment uses a preference profile for bothpartiesandcounterparties to evaluate the closeness of fit of preferences,it isa bilateral preference analysis method as opposed to theunilateral methods of the prior art.

We n ow c on si de r t he embodiment o f FIG. 2 in furtherdetail. First, in the embodiment of FIG. 2. a set of questions201 isposed toeach party andcounterparty.The questions aredesigned to reveal the utility va lue that each respondentplaceson the possible levelsof a set of attributes {aP a2,am) related to theproposed transaction.Bila teral p reference methodo logi es according to theembodiment of FIG. 2 are useful in, but are not limited to,three exemplary contexts.In the first exemplary context, an individual party wishestoenter a transaction with an organization counterparty. In thetransaction, the party seeks to identify an organization withrespect to which the preferences of such party are a good fitrelative to thealternatives. The organization counterparty, on

the other hand, seeks to give entry to parl ies who will be



22/29

US 8,069,073 B2successful within the organization. Examples of the firstexemplary context include a student (asthe individual party)choosingcollegesto attend (the organization counterparty),andan employee(theindividual party)choosinga corporateemployer(the organization counterparty). 5

In the first exemplary context, the questions asked of thepartyare designedto reveal the utility valuestliatthe partyplaces on possible levels of a set of attributes related to theenvironment withinthe counterparty organization. The preference profile created by the party's answers can thus be 10calleda "valueprofile"in thiscontext.Bycontrast,the questions asked of each potential counterparty organization aredesignedto revealthepreferenceprofilethat thecounterpartyconsiders necessary for a party to be successful within itsorganization. The preference profile created by the counter- 15party's answers in this context can thus be cal leda "successprofile". In the first exemplary context, a decision is recommended tothe party based on a relatively close fitbetween theparty's value profile and the counterparty's success profile.Note that questions for a counterparty organization in the 20first exemplary context are not necessarily directed to revealing profiles of successful individuals within its organizationin the past. The questionsmay instead elicit the value profilesof individuals that the counterparty believes will be successful within its organization in the future. 25

In a second exempla ry contex t, both the par ty and thecounterparties to a potential transaction are organizations. Inthe tr ansaction, the part y and coun terpa rt y seek to jointogether to form one organization. An example of such acontext is a corporate merger or acquisition. In the second 30exemplary context, the questions asked of both the party andpotential counterparties are designed to reveal the value profile that each respondent considers necessary for successwithin its organization. Thus, in the second exemplary context, a decisionis recommended basedon a relatively close fit 35between the success profiles of the party and counterparty. Ina merger example, such a recommendeddecision maximizes"culture fit" betweenmerging companies.

Finally, in a thirdexemplary context, boththe party and thecounterparties toa potential transaction are individuals. Inthe 40transaction, the party and counterparty seek to enter a financial relationship. For example, an individual party may seek acounterparty partner for a joint venture. In this third context,the questions asked of the party and each potential counterparty arc designedto reveal the utility value that each respon- 45dent places on possible levels of a set of attributes related tothe proposed relationship. A decision is then recommendedbased on a relatively close fit between the resulting valueprofiles of the party and a counterparty.

For convenience in this section of the description, we refer 50p redominant ly to a " po ten tia l transaction". However,embodiments of the present inven tion may also be used indealing with operation of an enterprise. In such a case, theparty and the counterparty may (but need not) be differentconstituents of the same enterpr is e and the issues of fit 55between the constituentsmay involve, for example, addressing organizational inefficiencies in the workplace and a widevariety of other activities. In one example, the party andcounterparty may be management and labor, and the issue offitmay involve a company policy todeal with staggeredwork 60hours. Alternatively, the party and counterpa rty may be amanagers of two different divisions of a company havingcompeting claims on a common resource to them, such asmarketing. Or, as yet anotherexample, the enterprise may becity government, the party maybe the police force, the coun- 65terparty may be the mayor, and the issuesof fit may be relatedtoemployee benefits, includingterms ofa health insurance, to

coverthepoliceforce. Inanycase,the technical approach forembodiments of this type is similar to that described belowwith respect to a potential transaction between party andcounterparty.A common featureof questions used ineach of the exemplary contexts is that the attributes are chosen so that those oftheparties"mirror,"or otherwisecomplement, those of thecounterparties. For example, for a party who is a collegeapplicantdecidingwhichcollegetoattend,the attributes mayinclude: population of the locality in which the school islocated,degreeof structure of the learning environment, andaverageclass size.The counterparties forthis transactionmaybe college admissions personnel, and their "mirror" attributesin helping studentsto decidewhetherthis schoolwouldprovide a close fit of preferences may be: "students who do wellhere prefer being in a locality with what population"; "students who do well here prefer attending a school with whatdegree of structure of the learning environment"; and "students who do well here prefer having classes ofwhat averageclass size."An example of questions according to one embodiment ofthe invention is provided inTabic 1 for collegeapplicants andforcollegesseekingapplicants.Table2 provides a similar setof questions for mutual fund purchasers and mutual fundsseeking investors. Table 3 provides a set of questions for jobseekers and employers seeking job candidates.In process 201 of the embodiment of FIG. 2, each respondent is posed N questions {Q,, Q,, . . . Q^f. As describedabove, questions for counterparties typically mirrorthe questions for parties. These questions may be fashioned ina widevarietyof forms. Inone form of questioning, each respondentis provided a series of individual multi-attribute descriptionsand asked to rate each of these descriptions. In another formof questioning, each respondent is provided with a series ofpairs ofmulti-attribute alternative descriptions, and is asked,for each pair, to select a desired one of the alternatives. Inyetanotherform ofquestioning,each respondent may be askedtoselect from among two or more alternative multi-attributedescriptions.Proper design of the questions permits statistical evaluation of the responses , from which may be derived uti lityvalues for each respondent. For example, the college applicant may be asked to rate a se lection of potential collegesfrom 1to 10. with 10 beingmost favorable; each college maybe characterized by a level for each of a scries of attributes.For example, in the case of attributes such as population oflocality, degree of structure of the learning environment, andclass size, one of the colleges to beranked may becharacterized by levels as fol lows: in a local ity with populat ion 100,000, unstructuredlearningenvironment, and small class size.In general, each attribute {a,} will have possible attributelevels which characterize i tin the example , there may bepossible college locations with populations between 15,000and 100.000; two options for learning environment (structured or unstructured); three class sizes (small, medium, andlarge), and so on. Note that the a tt ribu te levels need not benumbers, but may also be yes/no choices, or choices of itemsfrom a list of categories. Furthermore, note that attributes, andlevels of the attribute, may also be directed to "soft" characteristics related to a transaction; that is, characteristics whichare more emo ti onal in nat ur e and less quantif iable. Forexample, in an employment setting, a relevant attribute couldbe the degree of expected after-work socializing with fellowemployees , and the levels of the att ribute could be "rare .""moderate," and "frequent."Thequestions{Q,} neednot,however, askeachrespondentlo evaluate a list of all possible combinations of attribute



23/29

US 8,069,073 B2levels. Rather, the set of questions actually posed to therespondents are selected to achievea balance across independent contributions of each attribute (or, alternatively, suchthat every point in the space of possible attribute level combinations may be represented as a linear combination of thechosencombinations). In otherwords, thequestionsmay bedesigned sothatresponses tothem can beanalyzed interms ofattributes that, in mathematical terms, are orthogonal to oneanother or nearly so.In order to increase efficiency of the process of obtaininginformation from respondent or, to enhance the collection ofinformation that is most pertinent, questions may be structured hierarchically. In this way, responses for one or morequestions may be used to gate the selection of subsequentquestions. Alternatively, or in addition, questions may be insuites, with each suitedealing with a given areaof inquiry. Forexample, in the college selection example, one suite of questions may address factors governing the experience of life atthe school such as school size, social activities, geographiclocation, climate, facilities, nature of housing accommodations, and another suite may address conditions associatedwith pursinga givenmajor (say history or engineering) at theschool (conditions such as c la ss size, expec ted hours perweek studying, use of teaching assistants or use of full professors). Also, for example, inthe job example, one suite mayaddress company-related factors (such asexpectation/participation in company-sponsored events, expectation aroundconsensus building, locations, and emphasis on cross-training between fiinctions), and another suite may address function specific matters (such as frequency of overnight travel,work week hours, and type of job training programs).Inone particularembodiment, the questions are organizedinto three stages. In the first stage, the respondent ranks thelevelsofeachattribute, indescendingorder of preference. Forexample, "1" could signify the most preferred level, and "3 "the least preferred level, for three possible levels of anattribute. In the second stage, the respondent is asked to ralehis or her degree of preference for the mostpreferredlevelofeach attribute, over its least preferred level; for example, thedegrees of preference could be "1, slightly preferred"; "2,moderately preferred"; "3, greatly preferred"; "4, I musthavethe least preferred level would be upsetting." Finally,inthe third stage, a series of two-option choices isgiven to therespondent, forcing the respondent to express the degree towhich he or she would prefer one of two multi-attributecombinations. For example, the respondent could be presentedwith optionA and option B, each having different levelsoftwo attributes, and asked to rank them on a scale of 1 to 9 (1meaning "strongly prefer option A", 5 meaning "the two areequal," and 9meaning"strongly preferoption B"). Examplesof questions from each of these three stages are shown inFIGS. 9 through II .Once each respondent has provided a set of ratings {RL,R2, . . . Rjy} in answer to the questions (process 202), the

embodimentofFIG. 2 next calculates a preference profile foreach respondent, which includes the utility value that eachrespondent places on possible levels of the attributes {a,}related to the proposed transaction. The preference profile isgenerated in process 203 by establishing, for each respondent, a utility function U,(a,) for each attribute; this functionprovides a utility value corresponding to each level of theattribute a,. The utility functions are generated by first calculating a total utility for each example combination that wasranked by the respondent. The total utilities are calculated byevaluating proposed utility functions for each attribute at theattribute levels composing each combination, and, for eachcombination, summing the resultingutility values. The func-

45

50

SS

60

8tionsarethenchosenfromamongsttheproposedfunctions bythe criterion that a ranking of the total utilities should correspond to the respondent's actual rankings as closely as possible.The result, for each respondent, is a utility functionU,(a,) chosen foreachattribute {a[,a2. . . . am}. Eachutilityfunction translateseach levelof itsattributeintoa utilityforthat respondent. So, for example, a utility function will beestablished for the college applicant's evaluation of the college location attribute (with a utility value corresponding toeach locat ion A. B, and Csay 0.3, 0.2, and 0.4), the classsize attribute (with a utility value corresponding to small,medium, and large class sizessay 0.5, 0.2, 0.1), and soon.As in conventional conjoint analysis methods, the utilityfunctionsare normalizedto permit comparisons between theutility values of given levels ofdiffcrcnt attributes. However,in conventionalmethods, respondents are typically treated asa class and their responses are analyzedcollectively. Herethecontext is typically different, and the responses of each party(and counterparty) are typically analyzed separately, so thatfor each party and each counterparty there is obtained a separate set of utility functions. Furthermore, conventional methods produce util ity funct ions in a one-s ided, or uni la tera l,fashion. For example, a producer conventionally obtains a setof utility functions describing the preferences of consumers.By contrast, the method of the embodiment of FIG. 2 produces a set ofutility functions for the party and each potentialcounterparty, and continues with a bilateral analysis as discussed below. However, as described below, in some circumstances the preferences of a group may be evaluated collectively. Also, the responses of any individual to questionsmaybeaugmentedandextrapolatedon thebasis ofdata previouslyob ta i n ed fo r s imi la r individuals .The set of utility functions associated with a respondent(bethe respondent a party or counterparty) are sufficient to characterize the preferences of the respondent. For example, therecan be determined the relative importance that the respondentplaces on each attribute by calculating, for thai attribute, therange of the utility function over the interval of possibleattribute levels. A higher range for an attribute's utility function indicates a greater relative importance for that attribute.As an example, consider the hypothetical in the paragraphbefore last; the respondent's range of utility values for thecollege location attribute was 0.2 (from a low of 0.2 toa highof0.4); and the range for the class size attribute was 0.4 (froma low of0.1 toa high of 0.5). Class size is therefore relativelymore important for that respondent than college location.More generally, there may be derived from the utility functions U,(a,) for a respondent, a range vector {R,}having aseries of components R corresponding in each case to therangeof the utility function U,(a,)over levelsof the attributear . From the utility functions of a respondent there can besimilarly determined the level ofeach attribute giving rise tothe greatest utility experienced by the respondent. In otherwords, from the utility functions can be derived the attributelevels most preferred by the respondent. One may thereforedetermine a valuevector {V,} foreach respondent, as shownin process 204. The components of the value vector {V(}representthe levelsof each attribute {a,} that maximizetherespondent's utility function with respect to that altribule. Inparticular, if, forcounterparty number two. three levelsA, B.and C of attribute one (at) correspond to utility functionvalues of 0.2, 0.3, and 0.4 respectively, then level C will bechosen asV,, since itgives the maximum utility value fortiusat t r ibute.Given the seminal nature of the utility functions, the preference profile for each respondent, in this embodiment, is ihe



24/29

US 8,069,073 B29 10utility function vector {U,(a,)} for each attribute {a,,a2,. . . attribute a,: U/(a,) is the counterparty's utility function foram}. In other embodiments, the preference profile may be attribute a,; andthevertical barnotation indicates evaluationcomposed ofoneormore of the value vector {V,} and the of thefunction at attribute level Lrrangevector{R,}. Inprocess208of the distancevaluemethodof FIG. 2, theOnce the utility functions are generated, the process of 5 counterparty that,whenpaired withthe party, produces thedetermimng the counterparties having the closest fitwith a lowestdistancevaluePisidentifiedashavingtheclosestfitofparty begins. As shown in FIG. 2, there are two alternate preferences with the party. Similarly counterparties yieldingembodiments ofthemethod of FIG. 2. Inthe first, called the hi&herdistance values when paired with theparty areviewedaggregate value method, a list of counterparties having the as havin8 aPorer fit "''preferences with the party. By select-closest fit is determined by following processes 204, 205, 10 *Bagroup of the lowest distance valued counterparties, there206. and 208. In a second, alternative embodiment of the Can ^ P""^. a list of counterparties having a relativelymethod of FIG. 2, called the distance value method, the list ^ fat ' P/flerences ,wlth *pof the Party. While the. , . . c ., ,_ _ , ' ,. illustration above uses a linear d istancem easure t hat is m ini-may-begenerated by following.processes 207 and 208 (in- mize^ Qa ^^ , be ( dstead of processes 204.205.206, and 208). c , 7- c . , _ 3 , . w \i -* c i_ , . _. 15 including, for example, a least-squares approach. In such aIn process 205 ofdie aggregate value method, avector ,s ^ embodimen, of FIG. 2aUowsp^ and counter-generated correspondmg to a pairing of each counterparty jes to make decjsions ^ ja, 1nmsac6ms basedwith he pty. These vectors are formed^by evaluating the Qn abilaten|1^^ of preferences.party s utility functions (from process 203) ateach counter- Wbj]e ^ embodiment ofFIG 2^ been^^ wilhparty svalue vector levels (from process 204)-that is, at the ^ refernce ]o a^ ofcount ^ bein ided to a ^counterparty s utility-maximizing values There is thus hshould ^ understood mat ^ven ^ classes ofparties

formed, for each counterparty paired with the party, avector ôa^^y^^^d^^^îesr êmbodiment{U((a,)l}, where the vertical bar nolat.on indicates eyalua- 0fFia2canequallybeusedtorecommendalistofpartiestotion oftheparry sutility function for attribute a,ata=V and gcoume ^ te a Hshed b si j fo|low.Vîsthecounterparty sutil.ty-maximizing value lorattnbute 2S mg me described processes, but replacing the term -party"a'" with"counterparty," andvice versa. Generally, it should beIn process 206 of the aggregate value method there is undemoodthatembodimentsoftheinventionaresymmetri-computcd an aggregate value for each vector {U((a,) I} by calwi t h t t0mo sjdes ofa^^^^ mha, Uiey^ysummmg the componentsU(a,) Iofthe vector; i.e. by evalu- beUSfid tQ Tecon]metld decisions tooneside astotheaungthesum 30 olher

Furthermore, where embodiments are described in which,first,a preference profile is generated for persons on one sideV(/(a] ofa transaction, andthena preferenceprofileisgeneratedforfri* v, persons on the othersideof the transaction, it should bes5 understood by those of ordinary skill in the art that the order

of questioning the persons, and of generating the preferenceIn process 208ofthe aggregate value method, the counter- profiles, is nol essential. Thus, where it is described to askparty that, when paired with the party, produces the greatest questions of persons on one side ofa transaction first, andaggregate value is identified as having the closest fit ofpref- ibea ofpersons on tne other, itshould be understood that itiserences tothe party. Similarly counterparties yielding lower m equaIlypossible to reverse theorderofquestioning (byaskingaggregate values when paired with the party are viewed as questionsofthe opposite sideofthe transaction first), orevenhaving apoorer fit ofpreferences tothe party. By selecting a to ask questjons 0fboth sides simultaneously,group of the highest ranking counterparties, there can be Inafurther related embodiment, parties and counterpartiesprovided a listofcounterparties having a relatively closefitof are enabIed lomakedecjsj0ns based on a multilateral evalu-preferences with those of the party. 45 atjon 0f preferences. In such an embodiment, the methodIn the distance value version ofthe embodimentofFIG. 2, proceeds as described for FIG. 2, except that questions area list ofcounterparties providing a relatively close fit ofpref- asked notonlyofparties and counterparties, but also ofoneorerences isgenerated by using adistance measure between the moreco-evaluators.Aco-evaluatormaybeany natural personutility functions generated in process 203 for the party and oranentity, as with the parties and counterparties. The party,each counterparty. First, a utility function vector {U(a,)} is 50 and my oftne possible counterparties, may wish to includegenerated for the party andeach counterparty as described in me mpuXofaco_evaluatorasan aid to decision-making. Thus,process 203 above. Then, in process 207, for each possible for exampie, a college applicant party may wish to have acounterparty that canbepairedwith theparty,adistance value guidance counselor orhis parents evaluate the circumstancesisgenerated by comparing the utility functions ofthe pair. For underwhich he performs best,orseems mostcontent, inorderexample, a linear distance valueDmay be computed using a S5 to aid ^ in de^ng whicn coUege toattend. Similarly, adistance measure as follows: college counterparty maywish tohave input from alumni/ae,

faculty, and current students to guide in selection of studentstoadmit. Ina merger, a corporation party may wish tohave thefA , , . , ,, [Equation 11 members of its various departments, and even some of its4^4t" ' lj j" 60 customers and/or suppliers,act as co-evalualors, to assist indetermining thedegree of "culture fit"with a corporate counterparty with which it is merging,where the distance value D is calculated for each possible Ineachcase, the co-evaluatorchosen has a useful perspec-counterparty pairedwith the party; and whereAbs { } indi- live on the party or counterparty's preference profile. Thecatesthe absolutevalueof thesubtraction result inthe brack- 65 question array, ranking, utility function, and value vectorets; m is the number of attributes {a,}; J, is the number of proceduresare followedasin boxes 201 through 204.exceptlevels of attribute a,; U,(a,) is the party's utility function for lhalin thismultilateral embodiment theyare performed forat



25/29

US 8,069,073 B211

least oneco-evaluator, based onhisorherown perception oftheassociatedparty's or counterparty'spreferences, inaddition to being performed by the parties and counterpartiesthemselves.Co-evaluators may fall into two exemplary categories, 5although they are not limited to these categories. In the firstexemplary category, theco-evaluatorprovidesinputconcerningthecircumstances underwhichhisorherassociated partyor counterparty ismost contentor satisfied. In thiscategory,theco-evaluator canbesaidtoprovide a preference profile for tohis or herassociatedpartyor counterparty.In the secondexemplary category, the co-evaluator provides input concerning the circumstances under which his orher associatedparty or counterparty performs best. In thiscategory, the co-evaluator can be said to provide a success isprofileforhis or her associatedparty or counterparty.Attributes forco-evaluators typicallymirrorthose for parties and counterparties. For example, in the college-admissions example discussed in connectionwith FIG. 2. attributesfor a guidance counselor co-evaluator couldbe: "Prospect 20does best in environments that..." or "Prospectis happierwith products or services that.. ." Similarly, questions for aco-evaluator for a counterparty may be structured to elicitanswers to the questions: "People who do well here typicallylike jobs that ..." or "Users who are satisfied with this 25purchase typically prefer items that . . ."A co-evaluator for a party or counterparty need not be asingleperson;h couldalso bea groupofpeople. Forexample,a corporate counterparty may wish to use the members of agiven department as its co-evaluators ina transaction. Insuch 30a case, i.e. where a co-evaluator consists of a group of individuals, questionsare asked of each member of the group ofco-evaluators, and rankings are obtained from each. Then asingle setof utility functions (one function foreach attribute)isgenerated for the group of co-evaluators. This may be done 35by averaging utility functions for each member of the group;by weighting somemembers' utilityfunctions morehighly, ina weighted average of functions (with the optimal weightingdetermined based on the context of the transaction); or byallowing the counterparty (or party) associa ted with the co- 40evaluator to choose which group member's profi le to use asthe co-evaluator's profile.Where there is a group co-evaluator, orwhere there ismorethan one co-evaluator for a single party or counterparty, itmay also be useful to provide a visual display of each co- 45evaluator's preference profile to parties and counterparties.Sucha visualdisplaycouldtakethe formof a histogram,witha bar indicating the relative value of attributes; or the visualdisplay could graphically display a utility function for eachattribute, for each co-evaluator. Another useful visual display 50could be a scatterplot or dis tr ibut ion (characterized, forexample, by a mean and standard deviation) of thepreferenceprofile results from more than one co-evaluator. By usingsuch visual displays, parties and counterparties may beenabled to weigh the input of multiple co-evaluators in a 55comparative and qualitative fashion.Whena party or counterparty uses a co-evaluator, methodsaccording to embodiments of the invention may require theparty or counterparty's permission, before releasing a co-evaluator 's preference profile to other respondents in the 60decision-making process.Once a preference profile has been obtained for the party,each counterparty, and each co-evaluator, the next process in

a multilateral embodiment of th e invention is, as with thebilateral embodiments describedabove, to recommend a list 65of counterparties providing a relatively close fit of preferences. First il must be determined, for each party and coun-

12terparty who used a co-evaluator. how to use the co-evaluator*s preference profilein the analysis. Inone embodiment,this is performed by the following algorithm:1)Determine theclosenessof fitof the party or counterparty's preference profile with that of its associated co-evaluator. Thismay be done using the aggregate value methodorthe distance value method (each described above for a bilateral embodiment).2) If theprofileof the co-evaluatorisclose enoughto thatof the party or counterparty, as judgedagainst a pre-established standard, then thepartyor counterparty's ownprofilewill be used for comparison with potential partners to thetransaction.3)If,however, theco-evaluator'spreferenceprofiledifferssufficiently from that of its associatedpartyor counterparty(as judged against the pre-established standard), ihen theassociated party or counterparty is given a choice as towhichpreference profileto use forcomparisonwith potentialpartnersto thetransaction. Thepartyor counterpartymaychoosetouse its own profile only, orthat of the co-evaluator only, or(optionally, for an additional fee) to use each profileseparately andobtain results using each.Once itis determinedwhichpreference profilewill be usedfor the party and each counterparty, a multilateral embodiment of the invention proceeds as described above for bilateral embodiments. ITie result of this multilateral embodiment, then, is to provide a list of counterparties to theproposed transaction who provide a relatively close fit ofpreferences with those of the party, in a way that takes intoaccount the perspective of at least one co-evaluator.Because decisions are recommended based on the preferences ofmore than one party to a transaction, embodimentsofthe invention are particularly advantageous for long-term,relational transactions. Examples have been provided aboveof utilizationof embodiments in situations where parties andcounterparties may lack any previous business relationship.However,such a circumstance is in no way a necessary foundation for application of embodiments of the present invention. Forexample, embodimentsof the present inventionmaybe employed for evaluation ofexisting relationships betweenemployer and employee. Quest ions in such a circumstancemay, for example, be directed to particular work conditions,such as scheduling of employee's work hours during the day,work rules and changes to physical facilities. In this manner,management and employees may usefully evaluate potentialissues of importance in the work environment . As anotherexample, embodiments of the present invention may beapplied within corporationsto determinewhere there is"geargrinding"within the organization (inefficientor counterproductive relationships), or areas of difficulty in "culture fit"between merged companies.Similarly, embodiments of the present invention may beused in tandem with more traditional evaluation techniques.For example, potential employees may be identified usingtraditional techniques, and thereafter promising candidatesalong with human resources managers may be subjected toco-evaluation in a c co r dan ce wi th an embodiment of th epresent invention.It is equally possible to refine evaluation techniques invarious embodiments herein. One method of refinement is toconsiderinstanceswherein a close fithas been predicted byanembodiment, but wherein experience later shows there to bea problem. (Or alternatively, a close fithas not been predicted,but nevertheless resulted.) When the reasonforthe outcome isuncovered, it may be due to an attribute that had not beenpreviouslyidentified,ordue toan ineffectiveor badlywordedquestion. In such cases, the questions posed lo respondents



26/29

US 8,069,073 B213 14may be modified to take into account a new attributeor to provided, for example, in Gratia. Howthe Internet WorkscorrectmefTectivequestions.Thequestionsmaythenbeused (Ziff-Davis Press, 1996), which is hereby incorporated byfornew submissions to future respondents orcanoptionally reference; seeespecially pages 44-49.be resubmitted to former respondents. Alternatively, or in Infurtherembodimentsofsystemsandmethodsaccordingaddition, the problem may be attributable to improper analy- 5 to the invention,communication with a server and informa-

sis of the answers to thequestions, andthesematters can be tionprocessing maybe implemented usingwireless devices,adjusted by modifying, for example, the utility functions FIGS.4and5illustratethelogicalflowofamethodaccord-associatedwith thepartyor counterparty asappropriate, and ing to an embodiment of the invention that may be imple-re-performing the analysis. mented using a web server on the Internet. This embodimentFIG. 3 shows a block diagram of an embodiment of a to also illustrates use of the processes described above in con-system inaccordancewith the present invention.A first ques- nection with the system of FIG. 3.tion and response module 302 obtains responses from each Inbox401. a system, whichmay bea website serveron thememberofa first classofparties301to a firstsetof questions. Internet, receivesprimarydata from partiesand counterpar-The questions are designed toelicit revelationof preferences ties, via guidedtemplates for data entry.Each party or coun-thatcan be used to estimate the closeness of each party's fit 15 tcrparty enters the site, registers basic information (forwith potential counterparties to the transaction. The first example name, address, and other contact information), andquestion and response module then stores the party's selects a decision area from a set of pre-set parameters. Theresponses in a first d igital storage medium 303. pre-set decision areas may be, for example, college selectionSimilarly, a second quest ion and response module 305 or employment searching. For college selection, the partyobtains responses from each member of a second c las s of 20 could be a col lege applicant , and the counterparty may be acounterparties 304 to a second set of questions. These ques- college looking for or decided which students to admit; fortions are, similarly, designed to elicit revelation of prefer- employment searching, the party may be a job candidateence s that can be used to est ima te the closeness of each looking for a job, and the counterparty may be an employercounterparty's fitwithpotential partiesto the transaction.The lookingforemployees ordeciding amongst candidates.Aftersecond question and response module then stores the coun- 25 receiving the decision area choices, the system prompts theterparty's responses ina second digital storagemedium306. party or counterparty, via guided templates, for informationA first profile processor 307 uses the responses stored in onco-evaluatorsthat he or she wishes to include in the deci-firststoragemedium303toderiveafirstpreferenceprofilefor sion-making process.Thesystemalsogivesthepartyorcoun-each party, and a second profile processor 308 uses the terpartythe optionof usingdata fromonly the co-evaluatorsresponses stored in secondstoragemedium308 to derivea 30 in making the decision (with no input from the party orsecondpreferenceprofile foreachcounterparty. counterparty himself)- Finally, the party or counterpartyA closeness-of-fit analyzer 309 analyzes the preference authorizes payment, and the systemreceives and verifies theprofilegenerated foreachparty by firstprofileprocessor307 payment method(for example,creditcardpayment).in relation to the preference profiles generated by second Next,inbox402.thesystempromptseachpartyandcoun-profileprocessor308. Foreach party,theresult is an output 35 terparty, via guided templates, for supplemental data thatrankedlist310of counterparties providinga relatively close might be useful later in the process of evaluation. Forfit of preferences with that party, compared withthe other example, a job candidate party may be prompted for, andpotential counterparties. The closeness-of-fit analyzer com- register, a formatted resume.Acollege applicant partymaybemunicates sucha listto eachparty. prompted for, and register, a summary of his academic his-Inembodimentsofsystems accordingto the invention, the 40 tory. Ineach case, the promptedsupplemental data is poten-first andsecondquestionandresponsemodules302and305, tiallyusefulto a counterparty (e.g.an employerora college)the firstand secondprofileprocessors307and 308, and the laterintheprocessofevaluation(describedbelow).Similarly,closeness-of-fit analyzer309 may be implemented as com- thesystempromptscounterparties forsupplemental data thatputerprocesses running onmultiplecomputers incommuni- are potentially useful to parties laterin thedecision-makingcationwithcachother(forexampleovcranetwork,including 45 process. For example, if the counterparty is a companythe Internet), oras processes running ona singlecomputer. searching for job candidates, the company's supplementalSimilarly, the first andseconddigital storage media 303and datamay be"leads"on housing opportunities, which would306maybe separate storage devices, or portions of a single be attractive tojob candidates whoneedto find housing neardigitalstoragemedium. the company. Once the parties and counterparties haveIna preferredembodiment, thesystemof FIG.3 isiraple- 50 enteredsupplemental data, the systemassignsa uniqueiden-mentedasa hostcomputeraccessibleovera network, suchas tifierto each user, includingparties, counterparties, andanytheInternet.Inparticular, parties301andcounterparties 304 co-evaluators thattheyhavenamed.The systemalsocreatesmayaccessthe systemusing remotecomputerswhichare in a file for each user, and associateseach filewith the corre-communicationwith a hostcomputer viaWebpagesofa web spending unique identifier.siteontheWorld WideWeb. Thehostcomputeristhena web 55 Thesystemnext,in box403,disseminatesaquestionnaireserver, which runs computer processes that implement the formto each party and counterparty, along with the uniquefirstand secondquestion and responsemodules 302 and 305, identifierassigned toeach. (This process isomittedfor a partythe first and secondprofileprocessor307 and 308, and the orcounterparty whohaselectedtohaveevaluationperformedcloseness-of-fitanalyzer 309. The server stores responses to bya co-evaluatoronly,as described above inconnectionwithquestions in an associated content storage device (for 60 box 401.) The system then administers the questionnaireexampleatleastoneharddiskdrive),whichservesas firstand forms to eachpartyor counterparty. Forexample,the systemsecondstoragemedia303and 306.The servermaycommu- mayguidethepartyorcounterpartythrougha seriesofques-nicate with parties and counterparties using e-mail, or by tions formatted as templates on Web pages on a web site,makinginformationavailableon a web site, or byother com- through which the system receives each parry's (or counter-munication methods. 65 party's) answers to the questions. The questions on the ques-Further information concerning the Internet and E-mail tionnaireformaredesignedtoelicittheutilityvaluewhichthe(bothofwhichtermsare usedthroughout thisspecification)is respondentplacesonpossiblelevelsof eachattribute,without



27/29

US 8,069,073 B215 16necessarily asking for preferences directly. For example, the preference analysis. Optionally, such information may bequestions may be structured to elicit from the parties answers presented ina histogramorother graphical display, in order toto the questions: "I do best in environments that..." or "I'm visually display the results of the analysis. An example ofhappier with products or services that . . ." Similarly, the such a histogram for a job applicant is shown in FIG. 6; aquestions for counterparties may be structured to elicit 5 corresponding histogram for a counterparty employer isanswers tothequestions: "Peoplewhodo well here typically shown in FIG. 7;anda histogram showing a side-by-sidelike jobs that ..." or "Users who are satisfied with this comparison of the two is shown in FIG. 8. Note that histo-purchase typically prefer items that. ..." grams may beused todisplay weights orvalues or both; theyAs previously discussed. Tables 1-3 show examples of are sed for values only where the values in question arequestions that may be used, in various embodiments of the to


28/29

US 8,069,073 B217

party byusingaweb linkona web site,or hyusingaweb linksent to the party as part of an e-mail; or the system mayprovide phone numbers or other contact information, asauthorized in box 407 (above).Continuingwith box510 inFIG.5, thesystemnextquerieseach party and counterparty as towhat decisions it has madein the decisionarea forwhichthe analysiswasperformed; forexample,a partycould beaskedwhat job he or sheaccepted,orwhat product he or sheselected; anda counterparty couldbe asked which job candidates it selected for employment.These decision inquiries are repeated at time intervals that areeither chosen by theregistrant in the primary data entry process of box 401, or are specified to the registrant during theprimary data entry process, or are otherwise scheduledby thesystem.

Inbox 511, the system communicates aquery to each partyand counterparty as to its satisfaction with the decisionthat itmade, after ithas been informed that the party or counterpartyhas made the decision. This decision satisfaction inquiry isperformed at a time interval after process 510 that is eitherchosen by the registrant in the primary data entry process ofbox 401, or pre-dctermincd in the system.Next, inbox 512, the system, in one embodiment, performsa post-decision analysis. It analyzes key attributes that contributed to each party and counterparty's degree of satisfaction, by comparing each one's reported degree ofsatisfaction(from box 511) with the analyzed preference form resul tsobtained in boxes 406 and 408. The system communicates toeach party and counterparty its individual post-decisionanalysis, and provides each with a structured opportunity torespond to the analysis, e.g. by providing a set of web-pagetemplates enabling the party or counterparty to comment onthe key attributes identified in the post-decision analysis.Additionally, the system storesthe results of thepost-decisionanalysis, and the comments on it. Owing to the collection, inthe course of pract ic ing embodiments discussed in thisdescription, of substantial quantities ofdata that tend to beofa personal nature, it is within the scope of various embodiments to preserve the confidentiality of such data and todisclose such data only under circumstances to which theaffected individuals and organizations have given their cons en t .Large discrepancies between a preference form analysisand a post-decision report may indicate that the respondentdid not understand its own preferences well. Thus, such post-decision reports may help parties and counterparties to learnabout themselves, and therefore to make better decisions inme future.Results ofpost-decision analyses may beused to revise thesystem's method of preference form analysis, or to revise thequestions which are asked in each decision area. For example,if it is discovered that some college applicant parties havedecided toattend colleges with which theywere unhappy,andsome were unhappy based on attributes that the preferenceform did not elicit, then the preference form for the collegechoice decision area could be altered to incorporate the overlooked at t r ibutes .As part of the post-decision analysis, the system may alsoprovide a co-evaluatorwith a repor ton the par ty or counterparty's reported degreeofsatisfaction. For example, a collegeguidance counselor co-evaluator can be provided with areport on a college applicant party's (or a group of parties')degree of satisfaction, so that the counselor can modify his orher counseling in the future.Note, however, that in some contexts it is preferable toguarantee that a party's post-decision report will be kept inconfidence with respect to (at a minimum) the counterparty

18

20

25

30

withwhichthe partyentered a transaction(andvice versafora counterparty's confidences).Forexample, it is preferabletoguaranteeconfidentiality to an employeepartywho reportsdissatisfaction withanemployercounterparty ina post-decision report.Inbox 513, thesystem invites each party and counterpartyto update its preference profile. If the party or counterpartyagrees, the process begins anew, beginning with box 401,above.For suchupdates, theprocess retains thedata fromtheoriginal analysisprocess, andupdates itaccordingto thenewinputwhich thepartyor counterparty provides.The pricingtousers may be configured so that additional charges may bemade for updates, as opposed to original analyses. A party orcounterparty may also initiate the update process itself, without an invitation; this may. for example, be implemented byproviding an update option for regis trants on a web site. Aspart of an update, the system also allows a party or counterparty to add or delete co-evaluators. Ifthe update option isnotselected, the process proceeds to box 514.

In box 514, the system invites each party and counterpartyto perform a new matching process for closeness of fit,basedon its current preference profile. If the party or counterpartyagrees to do so, the process beginsanewat box 407. with thepricing changed accordingly.Although this description has set forth the invention withreference to several preferred embodiments, one of ordinaryskill in the art will understand that one may make variousmodifications without departing from the spirit and the scopeo f th e invention, as se t forth in t he c laims .

We c laim :1.A computer-implemented method for facilitating evaluation, inconnectionwith the procurementor delivery ofproducts or services, in a context of at least one of ( i) a financialtransaction and (h) operation of an enterprise, such contextinvolving a first class of parties in a first role and a secondclass ofcounterparties in a second role, the method comprising:in a first computer process, retrieving first preference datafrom a first digital storage medium, the first preferencedata including attribute levels derived from choicesmade by at least one of the parties in the first class:ina second computerprocess, retrievingsecond preferencedata from a second digital storage medium, the secondpreference data including attribute levels derived from

choices made by at least one of the counterparties in thesecond class;in a third computerprocess, for a selected party, performing multilateral analyses of the selected party's preference data and the preference data of each of the counterparties, and computing a closeness-of-fit value basedthereon; an d

in a fourth computer process, using the computed closeness-of-fit values to derive and provide a list matchingthe selected party and at least one of the counterparties.2.A method according to claim 1,wherein the list is rankedaccording to closeness of fit.3. A method a ccording to cl aim 1. fur ther c ompr is ingreceiving co-evaluator choices made by a party co-evaluator

or a counterparty co-evaluator, wherein the list matches the atleast one party and the at least one counterparty according toa multilateral analysis of preference data determined usingsuch co-evalua to rcho ices .4. A method according to claim 1, wherein the partychoices reveal, with respect to each level of each of a firstseries of attributes, a utility value which indicates the valuethai the party places on the level of the altribule.

45

50

55



29/29

US 8,069,073 B219

5. A method according to claim 4, wherein the partychoices reveal the utility values without the utility valuesbeing provided explicitly.6.Amethodaccordingtoclaim4, whereinthecounterpartychoices reveal, with respect to each level of each of a secondseries of attributes that complements the first scries ofattributes, a utility value which indicates the value that thecounterparty places on the level of the attribute.7.Amethodaccording toclaim 6,wherein thecounterpartychoices reveal the utility values without the utility valuesbeing provided explicitly.

208. A method according to claim 1,wherein at least one ofthefirstpreferencedata, secondpreferencedata,and thelistis

obtained from a remote server over a communication network.9. A method according to claim 1, wherein the partychoicesand thecounterpartychoices are providedto a remoteserver over a communication network.


Documents

Lumen View Technology v. Findthebest.com