IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`Case: IPR2013-00034
`Attorney Docket No.: 30693-0090IP1
`
`In re Patent of: Cheng et al.
`U.S. Patent No.: 7,970,674
`Issue Date:
`June 28, 2011
`Appl. Serial No.: 11/347,024
`Title:
`AUTOMATICALLY DETERMINING A CURRENT VALUE
`FOR A REAL ESTATE PROPERTY, SUCH AS A HOME, THAT IS
`TAILORED TO INPUT FROM A HUMAN USER, SUCH AS ITS OWNER
`DECLARATION OF DR. RICHARD BORST
`
`1.
`
`My name is Richard Borst of 41 Wharton Drive, Glen Mills, PA
`
`19342. I have been asked to offer technical opinions with respect to prior art
`
`references cited in this Inter Partes Review (“IPR”). I base these opinions on my
`
`work regarding Computer Assisted Mass Appraisal Systems (CAMAs) and
`
`Automated Valuation Models (AVMs). My current curriculum vita is attached.
`
`2.
`
`I hold a Doctor of Technology from the University of Ulster, Northern
`
`Ireland. My Doctoral work was in the field of geostatistics applied to Computer
`
`Assisted Mass Appraisal Systems (CAMA) and Automated Valuation Models
`
`(AVM).
`
`3.
`
`I have been involved in the mass appraisal of real property since 1973
`
`when I was the project manager on a contract with the State of New York. My
`
`project team at Calspan Corporation developed a residential CAMA system for the
`
`New York State Board of Equalization and Assessment.
`
` 1
`
`MICROSTRATEGY 1023
`Microstrategy, Inc. v. Zillow, Inc.
`IPR2013-00034
`
`

`
`4.
`
`In December of 1974, I took a position with North America’s then-
`
`largest mass appraisal company, Cole Layer Trumble Company. Initially, I
`
`managed the system development team. Under my direction of that team, the
`
`company began offering a CAMA system using multiple regression analysis and
`
`comparable sales analysis by mid-1975. The technology developed by me and my
`
`team at Cole Layer Trumble Company in the mid to late 1970’s is still being
`
`utilized in the United States, Canada, England, Australia and other places
`
`throughout the world. By 1981, I was president of the company and was a leader
`
`in the effort to transform U.S. Assessment offices from paper-based assessment
`
`systems to systems that leveraged computer readable and updateable methods and
`
`procedures. I was in that position until 1989.
`
`5.
`
`Since leaving my position as the president of Cole Layer Trumble
`
`Company, I have held various positions within the property valuation industry.
`
`For example, in 2003, I was the project executive of a team tasked with developing
`
`an AVM for the British government. That AVM was designed to value 22,000,000
`
`residential properties for the Council Tax. The AVM we designed is currently in
`
`use by the Valuation Office Agency (VOA), which is an executive agency of Her
`
`Majesty’s Revenue and Customs. As project executive, I directed the project team
`
`and personally calibrated the AVM to value well over one million properties.
`
` 2
`
`

`
`6.
`
`As of 2013, my responsibilities have included calibrating AVM
`
`models for large jurisdictions in Australia, Georgia (US), Ohio and Pennsylvania.
`
`7.
`
`I have published a number of peer-reviewed articles on the subject of
`
`CAMA and AVM methodology. I also have given presentations on AVM
`
`methodology at real estate and property tax conferences in Austria, Brazil, China,
`
`Italy, Germany, Mexico, Moldova, South Africa, the United States, and Canada.
`
`In 1990, as a chairman of an ad hoc committee organized by the Appraisal
`
`Standards Board of the Appraisal Foundation, I helped rewrite Standard 6 of the
`
`Uniform Standards of Professional Appraisal Practice (USPAP), which establishes
`
`requirements for the development and reporting of mass appraisals. The USPAP
`
`contains the generally accepted standards for professional appraisal practice in
`
`North America. I also served as a reviewer of the International Association of
`
`Assessing Officers (IAAO) “Standard on Automated Valuation Models.” See Ex.
`
`1019, p. 2.
`
`8.
`
`I have no financial interest in either party or in the outcome of this
`
`proceeding. I am being compensated on an hourly basis at a rate of $450 per hour.
`
`9.
`
`In preparing this declaration, I studied U.S. Patent No. 7,970,674 (the
`
`ތ SDWHQW

 $GGLWLRQDOO\ , KDYH UHYLHZHG WKH IROORZLQJ WKH 3DWHQW 2ZQHU
`V
`
`Response to the Revised Petition for Inter Partes Review; the Declaration of John
`
`Kilpatrick Under 37 C.F.R. § 1.132 (the “Kilpatrick Declaration); U.S. Patent No.
`
` 3
`
`

`
`5,857,174 (the “Dugan Patent”); and U.S. Publication No. 2005/1015465 (the
`
`“Kim Application”).
`
`10. My findings, as explained below, are based on my education,
`
`experience, and background in the fields discussed above.
`
`11.
`
`This declaration is organized as a response to the statements made by
`
`Dr. Kilpatrick. In particular, the Kilpatrick Declaration advances two positions
`
`with which I disagree: (1) that the term “automatic valuation” as it is used in
`
`LQGHSHQGHQW FODLPV  DQG  RI WKH ތ SDWHQW UHIHUV WR DQ $XWRPDWHG 9DOXDWLRQ
`
`Model (AVM); and (2) that all AVMs necessarily include features that are not
`
`shown in the Dugan Patent and the Kim Application. Moreover, in both his
`
`declaration and his cross-examination testimony, Dr. Kilpatrick has made a
`
`number of factual statements with which I disagree, and I address several of these
`
`below.
`
`A. Interpretation of “Automatic Valuation”
`
`12.
`
`In paragraph 26 of his declaration (attached as Exhibit 1013), Dr.
`
`.LOSDWULFN VWDWHV WKDW FODLPV  DQG  RI WKH ތ SDWHQW ³FODLP DQ DXWRPDWLF
`
`valuation model.” Moreover, in paragraph 45 of his declaration, Dr. Kilpatrick
`
`states that “A person of ordinary skill in the art on February 3, 2006, would
`
`XQGHUVWDQG WKDW WKH ތ SDWHQW FODLPV ZKHQ UHIHUULQJ WR D µYDOXDWLRQ¶ RU PRUH
`
` 4
`
`

`
`precisely an ‘automatic valuation’), refer to a market valuation based on a
`
`regression or similar process across a large data set.” I respectfully disagree.
`
`13. Claims 2 and 15 simply recite an “automatic valuation” of a
`
`“distinguished property” or “distinguished home.” These claims do not recite the
`
`use of a “model,” and they make no mention of a “data set,” much less the size of
`
`the data set being used. The mere recitation of “automatic valuation” does not
`
`mean more than the sum of its parts, as “automatic valuation” is not a term used
`
`with special meaning in the field, nor is it said to have any particular definition or
`
`VSHFLDO PHDQLQJ LQ WKH ތ SDWHQW 5DWKHU WKH SODLQ PHDQLQJ RI WKHVH ZRUGV LV
`
`well understood to those of skill in the art, and it differs from specialized terms like
`
`AVM. As such, “automatic valuation” would not be read by those of skill to
`
`require any particular type of model, nor to require any size of data set.
`
`“Automatic valuation” simply refers to the nature of the valuation.
`
`14.
`
`In greater detail, unlike the term “automatic valuation,” Automated
`
`Valuation Model (AVM) is a term specifically used within the field of property
`
`valuation to describe a type of computer system. For example, the IAAO has
`
`developed the “Standard on Automated Valuation Models (AVMs).” See Ex.
`
`1019. Similarly, the USPAP addresses the use of AVMs with regard to appraisals
`
`in its Advisory Opinion 18. See Ex. 1014.
`
` 5
`
`

`
`15. $V UHYHDOHG E\ WKH ,$$2 DQG 863$3 E\ WKH ILOLQJ GDWH RI WKH ތ
`
`patent, the term “Automated Valuation Model” was in wide and regular use; yet it
`
`ZDV QRWLFHDEO\ DEVHQW IURP WKH ތ SDWHQW OHDGLQJ SHUVRQV RI VNLOO WR LQIHU D
`
`distinction between it and terms like “automatic valuation” that were otherwise
`
`XVHG LQ WKH ތ SDWHQW 7KDW LV QHLWKHU WKH GHVFULSWLRQ QRU WKH FODLPV RI WKH ތ
`
`SDWHQW HYHQ PHQWLRQ $XWRPDWHG 9DOXDWLRQ 0RGHOV 7KH ތ SDWHQW DOVR IDLOHG WR
`
`specifically define the term “automatic valuation” or use it in a context that would
`
`limit that term to the features of an AVM.
`
`16.
`
`Therefore, I believe that the term “automatic valuation” as used in
`
`claims 2 and 15 does not require an Automated Valuation Model as suggest by Dr.
`
`Kilpatrick.
`
`17.
`
`Instead, the meaning of the term “automatic valuation” is plainly
`
`DSSDUHQW WKURXJK LWV XVH ZLWKLQ WKH FODLPV DQG ZLWKLQ WKH FRQWH[W RI WKH ތ SDWHQW
`
`itself. Something that is truly “automatic” happens without human intervention or
`
`thought. Of course, the claims are clear in their indication of what must be
`
`automatic: the “valuation.” A dictionary definition of “valuation” is “the act or
`
`process of valuing.” See http://www.merriam-webster.com/dictionary/valuation.
`
`Applying this definition in the context of claims 2 and 15, which otherwise recite,
`
`for example, “refining an automatic valuation of a distinguished home,” the recited
`
`“valuation” is the process by which a value of the distinguished property or home
`
` 6
`
`

`
`is calculated. And, modifying valuation with automatic, as recited, the plain
`
`meaning of the term “automatic valuation” is revealed as a calculation of a value of
`
`a distinguished property or home performed without human intervention.
`
`18. Returning to the subject of the term automatic, namely a valuation, it
`
`is apparent that the phrase requires nothing of processes performed before or after
`
`valuation. Rather, as indicated above, it merely requires the calculation of the
`
`property value to be performed without human intervention. As such, the amount
`
`and type of user input received before calculating a valuation does not impact
`
`whether the valuation is itself “automatic,” as recited by the claim language.
`
`Rather, a process that results in an “automatic valuation” may rely upon any
`
`amount of human input (e.g., through definition of a model or through
`
`specification of parameters for use by the model), as long as the ultimate
`
`calculation of the valuation itself is performed automatically (i.e., without human
`
`intervention or thought).
`
`19.
`
`7KLV LQWHUSUHWDWLRQ RI DXWRPDWLF YDOXDWLRQ LV FRQVLVWHQW ZLWK WKH ތ
`
`patent description, which describes “steps typically performed by the facility to
`
`automatically determine current values for homes in a geographic area.” Ex. 1001,
`
` 7KH ތ SDWHQW GHVFULEHV WKDW ³UHTXHVWV >WR GHWHUPLQH FXUUHQW YDOXHV@
`
`may be individually issued by users.” Ex. 1001, 5:50-58. As a prerequisite to
`
`determining the current value of a home, “a home’s attributes may be inputted by a
`
` 7
`
`

`
`person familiar with them, such as the owner, a listing agent, or a person that
`
`derives the information from the owner or listing agent.” Ex. 1001, 3:64-67.
`
`20. After reading automatic valuation more narrowly than warranted, to
`
`include un-recited features attributed to an AVM, Dr. Kilpatrick suggests that
`
`Dugan fails to disclose the recited “automatic valuation,” arguing that Dugan’s
`
`system requires the operator to manually select the comparable properties and
`
`perform the appraisal. See Ex. 1022, 174:12-22. On the contrary, Dugan makes
`
`clear that his system automates valuation of a selected property once the subject
`
`and comparable properties have been selected. Indeed, Dugan describes that “the
`
`system will proceed to determine an appraised value, step 62.” Ex. 1003, 8:32-34
`
`(emphasis added). Moreover, Dugan describes that his system is capable of
`
`suggesting comparable properties to the operator. Ex. 1003, 7:63-65. In other
`
`words, the Dugan system may accept selection input from an operator but the
`
`Dugan system then automates calculation of a valuation for the selected property to
`
`produce an “automatic valuation.”
`
`21. Dr. Kilpatrick similarly attempts to distinguish Kim from the claimed
`
`“automatic valuation” by arguing that Kim’s system is necessarily a manual
`
`process in which the user manually selects the comparable properties and performs
`
`the appraisal. See Ex. 1022, 193:5 to 195:7. On the contrary, Kim describes a
`
`valuation system implemented by a computer that performs various automated
`
` 8
`
`

`
`operations including a gathering operation 1406 that automatically selects
`
`comparable properties and an estimating operation 1414 that automatically
`
`“estimates a subject property value.” Ex. 1004, ¶¶ 0032, 0055-0056, 0092. In
`
`other words, the Kim system automates selection of the comparable properties and
`
`automatically values a subject property based on these comparable properties to
`
`produce an “automatic valuation.”
`
`22.
`
`For at least these reasons, I disagree with Dr. Kilpatrick’s
`
`interpretation of the term “automatic valuation” as it is used in the context of
`
`LQGHSHQGHQW FODLPV  DQG  DQG WKH ތ SDWHQW PRUH EURDGO\
`
`B. What is an AVM?
`
`23. After arguing that claims 2 and 15 require an AVM, Dr. Kilpatrick attempts
`
`to define characteristics that are necessarily present in all AVMs. See, e.g., Ex.
`
`1013, ¶¶ 26, 37-38; see also Ex. 1022, 22:4 to 23:5, 63:24 to 68:16, 75:7-12, 79:24
`
`to 82:10. Based on these characteristics, Dr. Kilpatrick attempts to draw a
`
`distinction between AVMs and systems like those described in Dugan and Kim,
`
`which he refers to “appraiser-oriented appraisal systems.” See Ex 1013, ¶¶ 38-39,
`
`48. In particular, Dr. Kilpatrick attempts to establish a “spectrum” on which
`
`AVMs and manual appraisals are on opposite ends. See Ex 1013, ¶¶ 38-39. While
`
`FODLPV  DQG  RI WKH ތ SDWHQW GR QRW UHFLWH RU RWKHUZLVH UHTXLUH DQ $90 HYHQ
`
` 9
`
`

`
`if they had, the Dugan patent and Kim application describe AVMs, consistent with
`
`the meaning attributed to that term by the industry.
`
`24. Dr. Kilpatrick argued that “[t]he distinguishing feature of the AVM is
`
`the use of computer algorithms to generate values in a statistical fashion based on a
`
`large data set of sales, without any need for an appraiser to identify comparable
`
`properties.” Ex. 1013, ¶ 37. Dr. Kilpatrick further argued that “by 2006, we’ve
`
`gotten to a point where we recognized, as we recognize today, that AVMs are
`
`basically regression models.” Ex. 1022, 124:6-9. He specifically distinguished
`
`AVMs from neural networks and artificial intelligence systems. See Ex. 1022,
`
`178:2-10. However, these narrow definitions of an AVM are contrary to the
`
`written authorities on AVMs and my own understanding of that term based on my
`
`studies and my lifetime of practice in the field.
`
`25.
`
`The USPAP describes an AVM as “a computer software program that
`
`analyzes data using an automated process.” Ex. 1014, 178:15. As far back as
`
`1997, the USPAP clearly recognized that AVMs do not require regression, as Dr.
`
`Kilpatrick suggests. Specifically, the USPAP describes that “AVMs may use
`
`regression, adaptive estimation, neural network, expert reasoning, and artificial
`
`intelligence programs.” Ex. 1014, 178:15-17.
`
`26.
`
`This understanding of the types of systems that qualify as AVMs did
`
`not change by 2003, when the IAAO approved and published the “Standard on
`
`
`10
`
`

`
`Automated Valuation Models (AVMs).” The IAAO’s Standard recognizes that, in
`
`addition to regression, adaptive estimation and neural networks were appropriate
`
`techniques for calibrating AVMs. See Ex. 1019, pp. 11-16. Rather, according to
`
`the IAAO’s Standard, “[t]he distinguishing feature of an AVM is that it is a market
`
`appraisal produced through mathematical modeling.” Ex. 1019, p. 5. However, as
`
`described below, the IAAO’s Standard recognizes that the mathematical models
`
`used by AVMs may take various forms.
`
`27.
`
`The IAAO’s Standard specifies that “AVM models are based upon
`
`one or more of the three approaches to value (cost, sales comparison, and
`
`income).” Ex. 1019, p. 9. With regard to the sales comparison method, the
`
`IAAO’s Standard describes that “one model is developed to identify comparable
`
`sales and a second model is developed to make adjustments for differences
`
`between the subject property and the identified comparables.” Id.
`
`28.
`
`Interestingly, Dr. Kilpatrick appears to recognize that Dugan utilizes a
`
`sales comparison method of valuation that is similar to the one described in the
`
`IAAO’s Standard. See Ex. 1013, ¶ 20. In particular, Dr. Kilpatrick describes
`
`Dugan as relying upon a “traditional sales comparison approach” and describes this
`
`approach as “requir[ing] adjustments to the sales price of the comparables.” Ex.
`
`1013, ¶ 20. This description is consistent with the “sales comparison method”
`
`recognized by the IAAO Standard as one method that may be employed by AVMs.
`
`
`11
`
`

`
`29.
`
`I agree with Dr. Kilpatrick’s determination that Dugan describes “a
`
`computer-assisted valuation model.” Ex. 1022, 161:25 to 162:8. However, I
`
`disagree with Dr. Kilpatrick that the Dugan patent and the Kim application fail to
`
`describe types of AVMs. The Dugan and Kim systems each produce a market
`
`valuation through mathematical modeling (i.e., a sales comparison model). Thus,
`
`viewed in light of the IAAO’s Standard and the USPAP’s description of AVMs,
`
`the Dugan and Kim systems qualify as AVMs that utilize the sales comparison
`
`model.
`
`30. Dr. Kilpatrick attempts to draw a distinction between AVMs and
`
`“appraiser-oriented appraisal systems” in order to demonstrate a difference
`EHWZHHQ WKH ތ SDWHQW DQG WKH 'XJDQ DQG .LP V\VWHPV See, e.g., Ex 1013, ¶¶
`
`38-39, 48. However, such a dichotomy does not accurately represent the true
`
`nature of these systems. The USPAP defines an appraiser as “one who is expected
`
`to perform valuation services competently and in a manner that is independent,
`
`impartial, and objective” and an appraisal as “the act or process of developing an
`
`opinion of value.” Uniform Standards of Professional Appraisal Practice, 2012-
`
`2013 Edition, p. U-1 (accessed at http://www.uspap.org/#/22/). In other words, an
`
`appraisal is an opinion of value developed and approved by an appraiser.
`
`31.
`
`,UUHVSHFWLYH RI WKH ODQJXDJH XVHG LQ WKH ތ SDWHQW WKH 'XJDQ SDWHQW
`
`and the Kim application, none of these three systems are producing an “appraisal.”
`
`
`12
`
`

`
`The Appraisal Standards Board (ASB) of the Appraisal Foundation is clear on this
`
`point. See Ex. 1018, p. 1. In particular, the ASB states that “[a]n appraisal is
`
`defined as ‘an opinion of value,’ which is distinctly different from an estimate or
`
`calculation of value.” Id. As a result, “[a]n AVM uses automated processes and
`
`cannot produce an opinion of value because only individuals can exercise
`judgment and form opinions.” Id. 7KXV E\ GHILQLWLRQ WKH ތ SDWHQW WKH 'XJDQ
`
`patent, and the Kim application can only produce estimates of market value. The
`
`use of the word “appraisal” in the Dugan patent and the Kim application is simply
`
`being used in a more general sense to describe a property valuation, not the more
`
`specific definition used by the USPAP and relied upon by Dr. Kilpatrick to draw
`
`his distinction.
`
`32.
`
`In fact, the IAAO’s Standard uses the word “appraisal” to describe the
`
`output of an AVM. See Ex. 1019, p. 5 (“The distinguishing feature of an AVM is
`
`that it is a market appraisal produced through mathematical modeling” (emphasis
`
`added).) Moreover, Dr. Kilpatrick himself uses the word “appraisal” to refer to the
`process described in the ތ SDWHQW See Ex. 1013, ¶ 42 (“Another difference
`
`between the Dugan Patent and the '674 Patent is the type of value which results
`
`from the two different appraisal processes.”)
`
`33.
`
`Therefore, despite being labeled as “appraisal” systems, the Dugan
`
`SDWHQW DQG WKH .LP DSSOLFDWLRQ OLNH WKH ތ SDWHQW DFWXDOO\ GHVFULEH DQ
`
`
`13
`
`

`
`automated process of rendering a value estimate. It is only after review and
`
`approval by an appraiser that the output of any one of these three systems could be
`
`considered an “appraisal” as defined by the USPAP. Whether the value
`
`estimations provided by these systems are actually used by an appraiser to render
`
`an appraisal is immaterial to their classification as an AVM.
`34. $FFRUGLQJO\ HYHQ LI FODLPV  DQG  RI WKH ތ SDWHQW UHTXLUHG DQ
`
`AVM, the Dugan patent and Kim application describe AVMs, as that term would
`
`have been understood by those of skill in the art.
`
`35.
`
`I hereby declare that all statements made herein of my own
`
`knowledge are true and that all statements made on information and belief are
`
`believed to be true; and further that these statements were made with the
`
`knowledge that willful false statements and the like so made are punishable by fine
`
`Code.
`
`or imprisonment, or both, under Section 1001 of Title 18 of the United States
`Richard
`Digitally signed by:
`Richard A Borst, PhD
`DN: CN = Richard A
`A Borst,
`Borst, PhD C = US O =
`Richard Borst Consulting
` PhD
`Date: 2013.08.26 09:29:
`35 -04'00'
`Signature: _____________________________
`
`Date: ________________
`
`Richard Borst, Ph.D.
`
`
`14
`
`

`
`ATTACHMENT AATTACHMENT A
`
`
`
`ATTACHMENT A
`(DR. BORST CV)
`
`(DR. BORST CV)(DR. BORST CV)
`
`
`
`
`
`
`
`
`15
`
`

`
`Name: Richard A. Borst, Ph.D.
`
`
`
`Summary
`
`Dr. Borst has an unusual technical/academic/business background. He has been engaged in
`managing the design, development, and implementation of computer-based real property information
`systems since 1973. He rose to the presidency of North America’s largest mass appraisal firm while
`at the same time maintaining his contributions to the technical aspects of mass appraisal systems. As
`to managerial experience, during his term of presidency his company was engaged in two
`simultaneously executed revaluation programs for jurisdictions in excess of one million properties
`each. Employment at the company at that time exceeded five hundred. Regarding his academic
`background, he obtained a Doctor of Technology from the University of Ulster, Northern Ireland July
`2007. His thesis was entitled: Discovering and Applying Location Influence Patterns in the Mass
`Valuation of Domestic Real Property.
`
`His technical background is evidenced in a number of published articles and conference
`presentations. He introduced artificial neural networks to the assessment community in 1990. He
`was appointed to a three-year term as a “Visiting Research Fellow” at the University of Ulster,
`Belfast, Northern Ireland. During this tenure he collaborated with members of the faculty at the
`University performing research in the fields of valuation modeling and the application of location
`effects in the model structuring and calibration process.
`
`Academic Qualifications
`
`Dr. Borst’s undergraduate studies were at the Cleveland State University, Cleveland, Ohio, USA. He
`majored in Engineering Science and graduated with honors.
`
`His master’s Degree was earned at the State University of Buffalo, Buffalo, New York, USA. He
`majored in Operations Research with specialization in mathematics, statistics, and management
`science.
`
`While studying at the University of Buffalo, he was a graduate teaching assistant and taught
`undergraduate courses in Operations Research and while under full academic scholarship he had sole
`responsibility for teaching a survey course in Operations Research to first year graduate students.
`
`His Ph.D. was earned at the University of Ulster, Northern Ireland. It was granted by the Faculty of
`Engineering, School of the Built environment. The title of his thesis was “Discovering and Applying
`Location Influence Patterns in the Mass Valuation of Domestic Real Property”.
`
`Representative Publications
`
`Borst, Richard A., Artificial neural networks: The next modeling/calibration technology for the
`assessment community?, 1991, Property Tax Journal March v. 10 no. 1, 69-94
`
`
`
`16
`
`

`
`Borst, Richard A., Thompson John F., Mathematical and computer system considerations related to
`the development of a generalized cost approach, 1987, Property Tax Journal v. 6 no. 3, 173-191
`
`Borst, Richard, McCluskey William, Comparative Evaluation of the Comparable Sales method with
`Geostatistical Valuation Models, 2007, Pacific Rim Property Research Journal, 13:1, 106-129
`
`Borst, Richard, McCluskey William, J., Using geographically weighted regression to detect housing
`submarkets: modeling large-scale spatial variations in value, 2008, Journal of Property Tax
`Assessment & Administration, 5:1, 21-54
`
`Borst, Richard A., Ph.D., 2008, Evaluation of the Fourier transformation for modeling time trends in
`a hedonic model, Journal of Property Tax Assessment & Administration 2008 v. 5 no. 4 (5)
`
`Borst, Richard A., Ph.D., 2008, Time Weighted Regression: Modelling the Effect of Sale Date in a
`Hedonic Model, Journal of Property Tax Assessment & Administration 2009 v. 6 no. 4 (3)
`
`Representative Conference Presentations
`
`Borst, Richard A., Statistics, models, calibration, valuation and comparables: Part I, 2002, 68th
`International Conference on Assessment Administration IAAO, 325-334
`
`Borst, Richard A, McCluskey, William J., (2006) Market Segmentation Models Developed Using
`Geostatistical Techniques, International Property Tax Institute, 9th Annual Conference, Kuala
`Lumpur
`Borst, Richard A. and McCluskey William J., 2007, Capturing the Premium for Location in
`Significant Neighborhoods, International Property Tax Institute Mass Appraisal Valuation
`Symposium, May 7th-8th, 2007, Toronto, Canada
`
`Borst, Richard A. and McCluskey William J., 2007, 2007 Evaluation of the Fourier Transformation
`for Modelling Time in a Hedonic Model: A Mass Appraisal Perspective, Economy, Valuation and
`Management of Real Estate and Natural Resources. Belarusian State Technological University, Minsk, April
`18-20, 2007
`
`Key Qualifications Commercial Sector:
`
`Computer Assisted Mass Appraisal Systems Knowledge
`Dr. Borst has had project or executive responsibility for the design, development and deployment of CAMA
`systems now used by units of government in twenty-eight States and two Canadian Provinces and the Country of
`England. Tens of millions of parcel valuations have been produced by these systems. Under his leadership the
`systems evolved from batch oriented cost approach to on-line, market approach, fully integrated property tax
`solutions. He directed the effort to develop an Internet Browser based property tax (including CAMA) system.
`Analytical Background
`Has background and experience in the use of multivariate analytical techniques. He has authored papers on
`the use of multiple regression, artificial neural networks and Geostatistical methods. He provided instruction
`in statistics to real estate appraisers. Graduate school courses included and mathematical statistics. He
`provided a three hour seminar in “Statistics, Models, Calibration, Valuation and Comparables: Part I” at the
`IAAO’s 2002 annual fall conference. During calendar year 2011 he will have been responsible for developing
`
`
`17
`
`

`
`hedonic models for the valuation of over one million properties. In addition he provides training to clients in
`the use of the company’s geostatistically based market valuation system.
`GIS Background
`While at Day & Zimmermann, Inc., he instituted a GIS practice. The goal of the practice was to convert existing
`paper or Mylar maps into GIS format. Technology included artificial intelligence for pattern recognition. Current
`activities include the development if systems that integrate GIS, Imaging and Computer Assisted Mass Appraisal.
`Consulting
`Dr. Borst has worked with several dozen of governmental units during the past thirty years. The “consulting” role
`was usually that of a participant on a project development team or as the manager of a large organization. He was
`the sole consultant on a study prepared for the Valuation and Lands Agency of Northern Ireland. This study
`resulted in a recommendation for moving forward with a reappraisal of domestic properties in this Province. More
`recently he has developed AVM models for large jurisdictions in the United States. This involves exploratory data
`analysis, model specification and testing and implementation. Of particular relevance was he was the Project
`Executive for the company’s effort in developing an AVM for England. He performed management and technical
`functions including calibrating a large number of multiple regression models. The scope of the project involved
`22,000,000 domestic real properties.
`Service to the Property Tax Community
`Dr. Borst has been and continues to be a reviewer of papers submitted for publication by the International
`Association of Assessing Officers. He is generally called on to review articles on sales ratio studies and novel
`valuation methodologies. In 1998 he was appointed to a three-year term as a “Visiting Research Fellow” at the
`University of Ulster, Belfast, Northern Ireland. During this tenure he collaborated with members of the faculty at the
`University performing research in the field of valuation modeling and the application of location effects in the model
`structuring and calibration process.
`
`Experience Record:
`
`2012
`Expert Witness services in an AVM patent dispute.
`2005-Present
`Senior Research Scientist, Appraisal & Tax Division, Tyler Technologies Inc.
`1998 - 2005
`Chief Strategy Officer, Cole Layer Trumble Company, a wholly owned subsidiary of Tyler Technologies, Inc.
`1975 - 1998
`Day & Zimmermann, Inc.
`Philadelphia, PA
`Has held several positions during over 24 years of continuous employment with Day & Zimmermann, Inc. Titles, dates,
`division names and responsibilities follow:
`Chief Technical Officer/Regional Manager
`West Chester, PA
`Apr. 1995 - 1998
`Cole Layer Trumble Company
`As regional manager, had operations (P&L) responsibility for $20M backlog of systems installation and field appraisal work
`in the Mid-Atlantic Region. Included the integration of GIS, EDMS, and Company developed property imaging and mass
`appraisal systems. Company mass appraisal system is developed in the ORACLE® RDBMS. As CTO provides strategy for
`systems offerings by the Company and participates in acquiring new business with selected strategic accounts. Introduced and
`marketed Web applications for the property tax.
`Senior Vice President, Business Development
`Jan. 1994 - Apr. 1995
`Day & Zimmermann Information Solutions Radnor, PA
`Responsible for the sales of systems and services for this new division. The division was comprised of several practices
`including, EDMS installation and integration, GIS installation and map conversion, Lotus Notes application development, and
`specialty software development for engineering applications.
`President
`
`
`18
`
`

`
`Radnor, PA
`April. 1992 - Dec. 1993 Applied Image & Information
`He started a new business unit to pursue the EDMS and GIS market as part of Day & Zimmermann, Inc. family of Companies.
`Target accounts included large chemical companies and power generating utilities. Responsibilities included all activities
`associated with a start-up operation, i.e. writing a business plan, staffing, acquiring technology, selling, establishing the
`delivery organization. He negotiated agreements to acquire exclusive U.S. rights to market EDMS software, and negotiated
`exclusive U.S. rights to the VectorVision process for map conversion. In January of 1994, Applied Image & Information was
`merged with four other technology units within the Corporation to form Day & Zimmermann Information Solutions.
`Group Senior Vice President, Business Development
`Radnor, PA
`Jan. 1990 - Apr. 1992
`Day & Zimmermann, Inc.
`Staff position reporting to the President of Day & Zimmermann, Inc., Management Services Group, an approximately $250M
`diversified services business unit. Assisted in pursuing new business opportunities for several business units within the group.
`Chief Executive Officer
`Dayton, OH
`Cole Layer Trumble Company
`1981 - 1990
`Full P&L responsibility for this business unit. During this period the Company enjoyed several years of gross revenue and net
`contribution that have not been surpassed to date.
`Company Officer (VP, Sr. VP, Executive VP, President)
`Dayton, OH
`Dec. 1974 - 1981
`Cole Layer Trumble Company
`Started in 1975 as Vice President, Systems development. Responsibilities progressively increased to include sales and
`marketing, then full responsibility for the unit.
`(This concludes the Day & Zimmermann, Inc. sequence of positions.)
`
`Section Head
`Buffalo, NY
`Calspan Corporation
`Jan. 1972 - Dec. 1974
`Managed a group of scientists, engineers and systems analysts in two rather distinct technical areas. The first related to
`supporting the development of the Anti-Ballistic Missile (ABM) System. The second was the development of real property
`information systems for the State of New York. Performed both business development and project management functions
`with this account.
`
`
`
`
`
`19
`
`

`
`ATTACHMENT BATTACHMENT B
`
`
`
`ATTACHMENT B
`(EXHIBIT 1013)
`
`(EXHIBIT 1013)(EXHIBIT 1013)
`
`
`
`
`20
`
`

`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`In re Patent of:
`U.S. Patent No.:
`Issue Date:
`App\. Serial No.:
`Filing Date:
`Title:
`
`Attorney Docket No. 30693-00901PI
`
`Cheng et a\.
`7,970,674
`June 28,2011
`11134