--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Greetings,

At DOS prompt, I am using the following command to print PostScript files:

	 C:\> copy "filename" /b lpt3:

"filename": The name of the file including its path,
/b: implying binary format,
lpt3: implying that the laser printer is connected to port 3,

Needless to say, your laser printer should have PostScript capability.

In case, you are having problem creating the .ps file, do not hesitate to
let me know.

Hope this helps


Best of luck.
Abedini
                       \|||/
                       (o o)
+=================oOO==(_)==OOo=================+
|_____|_____|_____|_____|_____|_____|_____|_____|
|__|_____|_____|_____|_____|_____|_____|__|__|__|
|_____|                                   |_____|
|__|__| Mohammad J. Abedini               |__|__|
|_____| School of Engineering             |_____|
|_____| University of Guelph              |_____|
|_____| Guelph  Ont, N1G 2W1              |_____|
|__|__| Tel.:(519) 824-4120 ext. 4321 (W) |__|__|
|__|__| Tel.:(519) 821-1199           (H) |__|__|
|_____| Fax :(519) 836-0227               |_____|
|_____| e-mail: mabedini@...      |_____|
|__|__|___________________________________|__|__|
|_____|_____|_____|_____|_____|_____|_____|_____|
+===============================================+
The world is your exercise-book, the pages on which you do your sums.
You are also free to write nonsense, or lies, or to tear the pages.

                                                                Richard Bach








On Mon, 12 May 1997, Sumit Dhar. wrote:

> Hello All!!
>  I desperately need some help on gslib out put *.ps files. i have tried
> taking the printout from HP Lasejet with *ps card. But the machine just hangs.
>  On the other hand the ghoshscript on linux refuses to compile any
> of them and produce any output. i tried dumping them into a file using the
> following command.
>  gs -sDEVICE=djet500 -sOutputFile=<out.prn> -f *.ps
>
> Is there something special about these output from GSlib? are these the
> same as the one's from ADOBE.
>  How do u actually take the printout of these files?
> can someone please help. i figured out that these group would have the
> maximum number of users of GSLIB and would be in n position to help me.
>
> -please try and help.
> thanking in advance.
>
> -Sumit Dhar
>
>
>
################################################################################
> "Computers are useless. They can only give you answers."
> 		 - Pablo Picasso.
>
################################################################################
> Sumit Dhar
>
> Department of Mining Engg.
> IIT Kharagpur
> Kharagpur-721 302
> WB, INDIA
>
> Ph: +91 3222 (55221 to 55224,77202 , 77390, 77393)  Extn 5482
> Ph: +91 3222 77698 (Res.)
>
###############################################################################
>
>
>
> --
> *To post a message to the list, send it to ai-geostats@....
> *As a general service to list users, please remember to post a summary
> of any useful responses to your questions.
> *To unsubscribe, send email to majordomo@... with no subject and
> "unsubscribe ai-geostats" in the message body.
> DO NOT SEND Subscribe/Unsubscribe requests to the list!
>

--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Hello All!!
	 I desperately need some help on gslib out put *.ps files. i have tried
taking the printout from HP Lasejet with *ps card. But the machine just hangs.
	 On the other hand the ghoshscript on linux refuses to compile any
of them and produce any output. i tried dumping them into a file using the
following command.
	 gs -sDEVICE=djet500 -sOutputFile=<out.prn> -f *.ps

Is there something special about these output from GSlib? are these the
same as the one's from ADOBE.
	 How do u actually take the printout of these files?
can someone please help. i figured out that these group would have the
maximum number of users of GSLIB and would be in n position to help me.

-please try and help.
thanking in advance.

-Sumit Dhar


################################################################################
"Computers are useless. They can only give you answers."
			 - Pablo Picasso.
################################################################################
Sumit Dhar

Department of Mining Engg.
IIT Kharagpur
Kharagpur-721 302
WB, INDIA

Ph: +91 3222 (55221 to 55224,77202 , 77390, 77393)  Extn 5482
Ph: +91 3222 77698 (Res.)
###############################################################################



--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Greetings,

I believe this can be helpfull for some of you.

Best regards,


Gregoire


<gregoire.dubois@...>
<Moderator of the ai-geostats mailing list>
<http://java.ei.jrc.it/rem/gregoire/>

we are distributing a free educational software for
map projections.  at this point, it has been tested
on sgi's only.  it is available through the ff url:

http://www.cse.ucsc.edu/research/slvg/map.html

if possible, can you please give it a try and provide
us with feedback.  thanks!

jeff and alex

Dear ALL
         In the world of spatial statistics i haven't seemed to come across
any 'paper' or book that discusses Spatial autoregressive models (SAR)
models with random cofficients, that is models of the form

         Y=Beta.W.Y + Xt+error

where X.t is a random factor.

Does anyone know of any key references/books that discusses this issue.

Thanks you for your assistance.

Regards Vince



--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

SSD'97

               5th International Symposium on Spatial Databases


                          CALL      FOR      PARTICIPATION


             July 15-18, 1997                       Berlin, Germany

                      http://www.inf.fu-berlin.de/~ssd97


VENUE
Computer Science Institute, Freie Universität Berlin


1. PRELIMINARY PROGRAM
2. TUTORIALS DESCRIPTION
3. CONFERENCE REGISTRATION FORM
4. HOTEL  REGISTRATION FORM
5. EXHIBITS


1. PRELIMINARY PROGRAM
    ===================


Tuesday , July 15: Tutorials
-----------------
8:30        Preregistration

9:30-12:30  Tutorial 1 -  Spatial Databases
                           Hanan Samet (University of Maryland, USA)

14:00-16:00 Tutorial 2 -  Spatial Relations: Models and Inferences
                           Max J. Egenhofer (NCGIA and University of Maine, USA)

16:30-18:30 Tutorial 3 -  Constraint Databases for Geographic Applications
                           Stephane Grumbach (INRIA, France)

18:45       Icebreaker party (Computer Science Institute, FU Berlin)
             and preregistration



Wednesday, July 16: Conference
------------------
8:00        Registration

9:00        Welcome and opening remarks

9:30-10:30  Keynote speech
             Spatial Internet Marketplaces: a Grand Challenge?
             David J. Abel (CSIRO, Canberra, Australia)

11:00-12:30 Session 1 - Spatial Similarities
             o 3D Similarity Search by Shape Approximation
               H.-P. Kriegel, T. Schmidt, T. Seidl (University of Munich,
Germany)

             o Finding Boundary Shape Matching Relationships in Spatial Data
               E. M. Knorr, R.T. Ng, D. L. Shilvock (Univ. of British Columbia,
               Canada)

             o Spatial Data Mining: A Database Approach
               M. Ester, H.-P. Kriegel, J. Sander (University of Munich, Germany)


14:00-15:00 Session 2 - Geo-Algorithms

             o A Simple, General and Efficient Implementation of Geometric
               Operators and Predicates
               E. P. F. Chan, J. N. H. Ng (University of Waterloo, Canada)

             o Materialization Trade-Offs in Hierarchical Shortest Path
Algorithms
               S. Shekhar, A. Fetterer, B. Goyal (University of Minnesota, USA)

15:30-16:30 Session 3 - Spatial Constraint Databases
             o Manipulating Spatial Data in Constraint Databases
               A. Belussi (Politecnico di Milano, Italy), E. Bertino, B. Catania
               (Universita` degli Studi di Milano, Italy),

             o Constraint-Based Interoperability of Spatiotemporal Databases
               J. Chomicki (Monmouth University, USA), P. Z. Revesz (University
               of Nebraska, USA)


17:00-19:00 Tutorial 4 - Spatial Data Mining & Spatial Data Warehousing
                          Jiawei Han  (Simon Fraser University, Canada)



Thursday, July 17: Conference
-----------------
9:00-10:30  Session 4 - Spatial Query Processing
             o Improving Spatial Intersect Joins Using Symbolic Intersect
               Detection
               Y.-W. Huang, M. Jones (University of Michigan, USA),
               E. A. Rundensteiner (Worcester Polytechnic Institute, USA)

             o Data Partitioning for Parallel Spatial Join Processing
               X. Zhou, D. J. Abel (CSIRO, Canberra, Australia), D. Truffet
               (University of Queensland, Australia)

             o Orthogonal Polygons as Bounding Structures in Filter-Refine
               Query Processing Strategies
               C. Esperanca (Universidade Federal do Rio de Janeiro, Brazil),
               H. Samet (University of Maryland, USA)


11:00-12:30 Session 5 - Systems
             o From GeoStore to GeoToolKit : The Second Step
               O. Balovnev, M. Breunig, A. B. Cremers (Univ. of Bonn, Germany)

             o Geo-Opera: Workflow Concepts for Spatial Processes
               G. Alonso, C. Hagen (ETH-Zurich, Switzerland)

             o Physical Database Design for Raster Images in CONCERT
               L. Relly, H.-J. Schek, O. Henricsson, S. Nebiker (ETH Zurich,
               Switzerland),


14:00-15:30 Session 6 - Spatial Data Models
             o Topological Error Correcting in GIS
               T. Ubeda (INSA Lyon, France), M. J. Egenhofer (NCGIA and
               University of Maine,  USA)

             o Vague Regions
               M. Erwig, M. Schneider (FernUniversitaet Hagen, Germany)

             o On the Assessment of Generalisation Consistency
               V. Delis (Computer Technology Institute, Patras, Greece and
               University of Patras), T. Hadzilacos (Computer Technology
               Institute,  Patras, Greece)

20:00 Banquet
       Restaurant Die Moewe, Berlin-Mitte



Friday, July 18: Conference
---------------
9:00-10:00  Keynote speech
             Perspectives in GIS Database Architecture
             Dick Newell (Smallworldwide plc)

10:00-11:00 Session 7 - Spatial Access Methods
             o New Linear Node Splitting Algorithm for R-trees
               C. H. Ang, T. C. Tan (NUS, Singapore)

             o The S-Tree: An Efficient Index for Multidimensional Objects
               C. C. Aggarwal, J. Wolf, P. Yu (IBM Watson Research Center, USA),
               M. Epelman (MIT, USA)

11:30-13:00 Panel discussion
             Title: "The Forces that Will Drive the Spatial Data Industry"
             Chair: John R. Herring (Spatial Systems, Oracle Corporation, USA)

13:00       Conference ends



2. TUTORIAL DESCRIPTIONS
    =====================

_____________ Tutorial 1
Spatial Databases
Hanan Samet, Computer Science Department, University of Maryland

Description:
A discussion of issues involved in the implementation of spatial databases
including the differences between the underlying data and their effect on the
index chosen, the types of queries to be supported, the integration of spatial
and non-spatial data, and user interface issues.  A live demonstration will be
given of a spatial database management system that highlights these issues.

Intructor:
Professor Samet is the author of the two books "The Design and Analysis of
Spatial Data Structures" and "Applications of Spatial Data Structures: Computer
Graphics, Image Processing and GIS", Addison-Wesley, Reading, MA, 1990.


_____________ Tutorial 2
Spatial Relations: Models and Inferences
Max J. Egenhofer, NCGIA and University of Maine, USA

Description:
This tutorial will review the role and importance of spatial relations
in geographic database systems, spatial query languages, and data
exchange. It focuses on models for spatial relations (topological
relations, cardinal directions, approximate distances) and their use in
spatial reasoning. We will round up the tutorial with a discussion of
the integration of spatial relations into spatial query languages and
new applications enabled by the formalisms for spatial relations.

Intructor:
Dr. Max J. Egenhofer is the director of the National Center for
Geographic Information and Analysis (NCGIA) at the University of Maine,
an associate professor in Spatial Information Science and Engineering,
and a cooperative associate professor in Computer Science. His research
interests include geographic databases, spatial reasoning, and user
interface design for geographic information systems.

_____________ Tutorial 3
Constraint Databases for Spatial Applications
Stephane Grumbach, INRIA, France

Description:
The manipulation of large quantities of data requires the design of
high level data models that allow reliable and efficient retrieval,
querying and restructuring of the data.  In this tutorial, we will
show that the constraint database paradigm offers an appropriate
framework for spatial information. It allows the definition of a
sound data model, compatible with the vector mode representation, and
the design of query languages, based on a small number of primitives.

Instructor:
Dr. Stephane Grumbach is a researcher in the Verso project at INRIA
Rocquencourt, where he works on databases, both from theoretical and
practical perspectives.  His recent work is on the design of data
models and query languages for new database applications.


_____________ Tutorial 4
Spatial Data Mining and Spatial Data Warehousing
Jiawei Han, Simon Fraser University, Canada

Description:
This tutorial provides a comprehensive survey on spatial data mining and
spatial data warehousing technologies, including spatial data cube
construction, spatial OLAP, concept description and comparison, spatial
association, classification, clustering, pattern-directed mining, and data
visualization.  Some spatial data mining and/or data warehousing systems
and methods for integration of spatial data warehousing and spatial mining
will also be discussed.

Instructor:
Jiawei Han, Professor of Computing Science, Simon Fraser University, Canada.
He has been working on data mining, data warehousing, deductive and
object-oriented databases, and spatial databases, with numerous publications.
He is currently leading a project developing DBMiner and GeoMiner systems.


-----------------------------------------------------------------




3. CONFERENCE REGISTRATION FORM
    ============================

Note that a PostScript version of the following form (1 page)
is available at URL http://www.inf.fu-berlin.de/~ssd97




                           REGISTRATION FORM
               FIFTH INTERNATIONAL SYMPOSIUM ON SPATIAL DATABASES
                                SSD'97
                    JULY 15-18, 1997, Berlin, GERMANY

          ----------------------------------------------------------------
          SEND or FAX to: Agnes Voisard, SSD'97, Institut fuer Informatik,
          FU Berlin, Takustr. 9, D-14195 Berlin. Fax: +49-30-838 75 109

          --          There is NO electronic registration               --
          ----------------------------------------------------------------


Name _____________________________________________________________________
         Last                                                 First

Organization _____________________________________________________________

Address __________________________________________________________________

         Street           City    No       State        Zip         Country

Telephone ______________________________      Fax ________________________

E-mail _________________________________

Do you have any special needs? ___________________________________________


CONFERENCE REGISTRATION FEES
----------------------------
Registration includes all sessions, refreshment breaks, opening reception,
conference banquet on July 17, and one copy of the conference proceedings.
Additional banquet tickets may be purchased upon request at DM 100 each.

ADVANCE REGISTRATION (By June 13)            LATE (after June 13, *FIRM*)
                                              and ON-SITE REGISTRATION
June 13 (FIRM)

<  > ACM/SIGMOD member - DM 400            <  >  ACM/SIGMOD member - DM 450

ACM/SIGMOD membership number _____________________________________________


<  > Non-ACM member - DM 450               <  > Non_ACM member -DM 500

<  > Full-time student - DM 150            <  > Full-time student - DM 200
      (students must send a copy of their student card)


                                                            Subtotal:   ... DM

                                       (...) extra banquet ticket(s):   ... DM


TUTORIALS
---------
<  > Tutorial 1, Tuesday, July 15, 9.30-12.30: Spatial Databases

<  > Tutorial 2, Tuesday, July 15, 14-16:  Spatial Relations:
                                            Models and Inferences

<  > Tutorial 3, Tuesday, July 15, 16.30-18.30: Constraint Databases for
                                                 Geographic Applications

<  > Tutorial 4, Wednesday, July 16, 17-19: Spatial Data Mining & Spatial
                                             Data Warehousing


Costs:
1 tutorial: DM 50, 2 tutorials: DM 80, 3 tutorials: DM 105, 4 tutorials: DM 120


                                                            Subtotal:   ... DM


                     ----------------------------------------

METHODS OF PAYMENT ACCEPTED (please check one):

<  > Payment by credit card:

                      <  >  VISA         <  > MasterCard/Eurocard

Credit card number ________________________

Cardholder name ___________________________        Exp. Date ______________



<  > Payment by bank transfer in German currency (DM) made out to:
      Agnes Voisard, SSD'97, Berliner Bank account No. 8106776300, BLZ 10020000.


<  > Payment by check in German currency (DM) made out to:
      Agnes Voisard, SSD'97.


IMPORTANT: ALL EXTRA FEES EMERGING FROM CHECK PAYMENT AND BANK TRANSFER
            MUST BE COVERED BY THE PARTICIPANT.



Date and Signature _____________________              Total Enclosed:   ... DM


Refund Policy:
Before June 13, 1997:              100% refund
Between June 13 and July 1, 1997:  DM 50 processing fee
After July 1, 1997:                No refund







4. HOTEL REGISTRATION FORM
    =======================

Hotel accomodations for SSD 97 are managed by DER Deutsches Reisebuero,
Berlin. DER is Germany's largest travel agency, and is primarily owned
by Deutsche Bahn AG (German Railway) and Deutsche Lufthansa AG. DER
has reserved a block of hotel rooms at considerably reduced rates. All
hotels are very near to subway lines which will take you easily to the
university (metro station "Dahlem-Dorf".
They are centrally located around the famous street
Ku'damm, the main shopping area of Berlin. DER offers also the
opportunity to organize your personal trip to Berlin. Please check
here if interested or send a fax with date and place of departure
and date of return.

Description of the hotels:
(pictures of the hotels available at http://www.inf.fu-berlin.de/~ssd97)

- Sorat art'hotel
(Joachimstaler Str. 29)
It is a very modern 3-star hotel.
The hotel has 133 luminous and comfortable design-rooms, which were
created by the world famous designer Wolf Vostell. It is centrally
located near Kurfuerstendamm. It has the shortest connection to the
subway system which will take you to the university. With the Bus
109/X9 you have a direct connection to airport Tegel. You=B4ll find of
course aircon-ditioning, telephone, cable-TV, radio and minibar. We
specially recom-mend this  hotel. Conference rates are: 125 DM per
night (single), 150 DM per night (double).

- Berlin Mark Hotel
(Meinekestr. 18-19)
It is a modern 3-star hotel, located in the heart of the city,  just a
few minutes walk from the world-famous Kurf=FCrstendamm. The hotel has
231 comfortable rooms. The hotel=B4s central location, within easy reach
of all modes of public transport, makes it an ideal starting point for
discovering Berlin. Conference rates are: 150 DM per night (single),
150 DM per night (double).

- Hotel Imperial
(Lietzenburger Str. 79-81)
It is a modern 3-star hotel, located next to Kurf=FCrstendamm.
The excellent location offers direct access to bus and underground
connections to airports and railroad-stations. During the summer, the
garden-restaurant with its open air swimming pool welcomes you to a
green oasis in the center of Berlin. The hotel has 81 comfortable
rooms equipped with minibar, TV, telephone and hairdryer. Conference
rates are: 115 DM per night (single), 162 DM per night (double).

All prices include breakfast-buffet, service and V.A.T.


-Jugendgaestehaus am Zoo
(Hardenbergstr. 9a)
It is a hotel for students and young people. All rooms have running
water, shower-bath and WC are on the floor. The hotel is equipped with
multiple-bedded rooms. Conference rates are: 41 DM per night (1
person). Continental Breakfast is included. The Deutsche Reiseb=FCro
calculates per person 25 DM handling fee for reservations in this
hotel.

It takes about 30 minutes from the hotels to the university.
Take line U1 direction Krumme-Lanke to reach the station
Dahlem-Dorf.
Once you are at the Subway station Dahlem-Dorf, follow the instructions
on the map that can be downloaded from URL http://www.inf.fu-berlin.de.
The computer Science Institute is at a 5 minutes
walking distance from the metro station.



Please fax hotel reservation form to:
BWO Marketing Service GmbH
Mohrenstr. 63-64
D-10117 Berlin, Germany
Tel.: +49-30-226684-70
Fax: +49-30-226684-64
e-mail: drooff@...

For further requests please contact:
DER Deutsches ReisebueCro, Berlin
Augsburger Stra=DFe 27
D-10789 Berlin, Germany
Tel.:  +49-30-21998-997
Fax: +49-30-211 81 50


Note that a PostScript version of the following form
is available at URL http://www.inf.fu-berlin.de/~ssd97


Title  ____________________________ ________

Family
Name _________________________________________________________________
_________ Sex  o female       o male

Given
Name _________________________________________________________________
_________

Affiliation __________________________________________________________
________________

Address  _____________________________________________________________
_____________

City  ________________________________________________________________
__________

State  ____________________________ ______________

Zip  ______________

Country  ___________________________________________________________

Phone  ____________________________ ________

Fax  ____________________________ ________

E-mail  ______________________________________________________________
____________

Hotel 1. Choice ________________________________

o single  o double

Hotel 2. Choice ____________________________ ________

No. of nights ____________________________ ________

Arrival Date* ________________________________ *(Please indicate, if
you arrive later than 6.00 p.m.)

Departure Date* ____________________________ ___  *(Check out time is
12.00 noon) Further
Comments _____________________________________________________________
_______


Deadlines and Cancellation:
The deadline for the reservation of accomodation is 13 June 1997.
Reservations after this date cannot be guaranteed, but DER will
probably find accomodations for you. In case of cancellation later
than 5 working days before arrival, cancellation fees may have to be
paid. Payment All room fees are paid to DER. For all those paying by
credit card, all fees will be charged at the end of their stay. Please
note that DER accepts Visa and Mastercard. Those paying by cheque must
include a deposit of 200 DM with their registration and pay the
remaining balance by 13 June 1997.  Those who would like to pay with
an EuroCheque have to send the cheque (noting the usual restrictions
on amount) with their registration. There is no fee for using an
EuroCheque. If you would like to pay with a cheque in DM, please send
the cheque with your registration. Note that DER has to charge a fee
of 25 DM for processing foreign cheques. Charges for  other extra
services have be paid to the hotel.

Payment Method  o EuroCheque 		 o Cheque
     o VISA 			 o MASTERCARD

Card No.: _________/_________/________/________

Exp. Date: ________________________

Holder:  _____________________________________________________________
____


Date/Signature: ______________________________________________________
___________



5. EXHIBITS
    ========

There will be exhibits during the conference.
Exhibits are free of charge for universities and sponsors.
Fees are DM 1000 for other industrials.
Please contact the Exhibit Chair for further information:

Guenter Feuer
Institut fuer Informatik
Freie Universitaet Berlin
Takustr. 9
D-14195 Berlin, Germany
feuer@...
Fax: +49-30-838 75 109



----------------------------------------------------------------------

In Cooperation with ACM SIGMOD, NCGIA

Sponsors
--------
Bull AG
Cassini (French CNRS GdR 1041 and Universite' de Provence)
CHOROCHRONOS
DFG (Deutsche Forschungsgemeinschaft)
ESRI, Inc.
ETAK, Inc.
Fraunhofer ISST
Freie Universitaet Berlin
ICSI (International Computer Science Institute, Berkeley)
INRIA
Oracle France
Oracle Spatial Products, USA
SMALLWORLD Systems GmbH


General Chair
-------------
Agnes Voisard, Freie Universitaet Berlin, Germany


Program Committee Chair
-----------------------
Michel Scholl, INRIA and CNAM, France


Local Arrangements
------------------
Heike Friedrichs, Freie Universitaet Berlin
Oliver Grzegorski, Freie Universitaet Berlin
Inge Hellmich, Freie Universitaet Berlin


Tutorial Chair
--------------
Ralf Hartmut Gueting, FernUniversitaet Hagen, Germany


Panel Co-chairs
---------------
Andrew U. Frank, TU Vienna, Austria
John R. Herring, Oracle Corporation, USA


Exhibit Chair
-------------
Guenter Feuer, Freie Universitaet Berlin, Germany


Program Committee
-----------------
David J. Abel, CSIRO, Canberra, Australia
Gustavo Alonso, ETH Zuerich, Switzerland
Elisa Bertino, Universita` degli Studi di Milano, Italy
Rolf A. de By, ITC, The Netherlands
Leila De Floriani,    Universita` di Genova, Italy
Max J. Egenhofer,     NCGIA and University of Maine, USA
Christos Faloutsos,   University of Maryland, USA
Robin Fegeas,         USGS, USA
Andrew U. Frank,      TU Vienna, Austria
Volker Gaede,         Humboldt-Universitaet, Berlin, Germany
Kenn Gardels,         UCB, USA
Christopher Gold,     Universite' Laval, Canada
Ste'phane Grumbach,    INRIA, France
Oliver Guenther,      Humboldt-Universitaet, Berlin, Germany
Ralf Hartmut Gueting, FernUniversitaet Hagen, Germany
Jiawei Han,           Simon Fraser University, Burnaby, Canada
John R. Herring,         Oracle Corporation, USA
Erland Jungert,       FOA, Sweden
Curt Kolovson,       Hewlett Packard, USA
Manolis Koubarakis,   UMIST, Manchester, UK
Robert Laurini,       INSA, Lyon, France
Claudia Medeiros,     UNICAMP, Brazil
Scott Morehouse,     ESRI, Inc., USA
Richard Muntz,        UCLA, USA
Beng Chin Ooi,        University of Singapore
Dimitris Papadias,    Univ. of Science and Technology, Hong Kong
Didier Plateau,       O2 Technology, France
Chinya Ravishankar,  University of Michigan, USA
Elke A. Rundensteiner, Worcester Polytechnic Institute, USA
Hanan Samet,          University of Maryland, USA
Timos Sellis,        NTUA, Athens, Greece
Terence Smith,        UCSB, USA
Jan Van den Bussche,  University of Limburg, Belgium
Peter Van Oosterom,   Kadaster Concerstaf LKK, The Netherlands
Agne`s Voisard,        Freie Universitaet, Berlin, Germany
Frank Wagner,         Freie Universitaet, Berlin, Germany


SSD Steering Committee
----------------------
Dave Abel (CSIRO, Canberra, Australia), Max J. Egenhofer (NCGIA and University
of Maine, USA), Oliver Guenther (Humboldt-Universitaet, Berlin, Germany),
John Herring (Oracle, USA), Beng Chin Ooi (NUS, Singapore), Hans-Joerg Schek
(ETH, Zurich, Switzerland), Terence Smith (UC Santa Barbara)

______________________________________________________________________

--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear All:
	 This query is regarding the ik3d routine in gslib. In the input
parameter file, the cutoff value has to be given as input along with it
the global cdf also has to be given.
	 My question is how is this global cdf calculated?
i have tried (rank -1)/n, for the cluster.dat for the primary variable,
but that doesnot appear to give the required answer.

Can anyone please help...

Thanks in advance.

Sumit Dhar


################################################################################
"Computers are useless. They can only give you answers."
			 - Pablo Picasso.
################################################################################
Sumit Dhar

Department of Mining Engg.
IIT Kharagpur
Kharagpur-721 302
WB, INDIA

Ph: +91 3222 (55221 to 55224,77202 , 77390, 77393)  Extn 5482
Ph: +91 3222 77698 (Res.)
###############################################################################


--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Netsy/Intesys - L. Ongaro wrote:
>
> >>>>
>
>      Return-Path: <carli@...>
>      X-Sender: MASSIMO CARLI <carlim@...>
>      Date: Thu, 15 May 1997 15:53:04 +0200
>      To: nonna@..., 00492aaa@...,
>      puliti@...,
>      ruggeri@..., giudici@...,
>      bertoncello@..., alm@...,
>      deriso@...,
>      garavelli@..., mike@...,
>      MAURI@..., netsy@..., jag@...,
>      a922ts@..., visea@...
>      From: Massimo Carli <carli@...>
>      Subject:
>
>      >From owner-java-it@... Thu, 15 May 1997
>      10:21:26
>      From: "Filippo Bobbi" <f.bobbi@...>
>      To: "Java-it" <java-it@...>
>      Subject: Java-IT: I: (Fwd) Dev-IT: Appello Urgente
>      Date: Thu, 15 May 1997 10:03:37 +0200
>      X-MSMail-Priority: Normal
>      X-MimeOLE: Produced By Microsoft MimeOLE Engine V4.71.0544.0
>
>      Sender: owner-java-it@...
>      Reply-To: java-it@...
>
>      ----
>      Da: Bertellotti Alessandro <abertel@...>
>      A: dojo@...; barmario@...
>      Data: gioved 15 maggio 1997 8.20
>      Oggetto: (Fwd) Dev-IT: Appello Urgente
>
>      > Appello Urgentissimo - VERY URGENT APPEAL
>      >
>      > Urge aiuto per una bambina italiana di sei anni Francesca
>      De Lunas
>      > FRANCESCA DE LUNAS IS A 6 YRS OLD DAUGHTER WHO URGENTLY
>      NEEDS HELP.
>      >
>      > ricoverata presso il reparto rianimazione dell'ospedale
>      Brotzu di
>      > Cagliari
>      > (Sardegna-Italia) Dott. Pettinao.
>      > SHE'S NOW UNDER THERAPY BY DR. PETTINAO AT THE "REPARTO
>      RIANIMAZIONE"
>      > [REANIMATION DEPT.],
>      > OSPEDALE BROTZU, CAGLIARI (SARDINIA, ITALY)
>      >
>      > La bambina presenta una sindrome caratterizzata da:
>      > SHE SHOWS A SYNDROME WHICH PECULIARITIES ARE:
>      >
>      > - ipercoagulazione del sangue;
>      > BLOOD HYPERCOAGULATION
>      >
>      > - trombi che si formano in circa 15 secondi di colore
>      rosso con nucleo
>      > bianco
>      > di lunghezza di circa 2-3 cm.
>      > RED THROMBI WITH WHITE NUCLEUS, ABOUT 2-3 CM LONG, WHICH
>      FORM IN 15
>      > SECS.
>      >
>      > Alla bambina e' stata gia' amputata la gamba sinistra e,
>      finora non e'
>      > stato possibile effettuare diagnosi di alcun tipo.
>      > THE DAUGHTER HAS ALREADY BEEN AMPUTATED THE LEFT LEG. IT
>      HAS NOT BEEN
>      > POSSIBLE
>      > TO MAKE ANY DIAGNOSIS UNTIL NOW.
>      >
>      > Chiunque sia in grado di ipotizzare una diagnosi o di
>      riconoscerla
>      > si metta urgentemente in contatto con le seguenti e-mail:
>      > ANYBODY ABLE TO MAKE A DIAGNOSIS OR RECOGNIZE THIS DISEASE
>      PLEASE GET
>      >IN
>      > TOUCH
>      > URGENTLY BY E-MAIL TO THE FOLLOWING ADDRESSES:
>      >
>      > <mailto:lukrezia@...>lukrezia@...
>      >
>      > <mailto:stefania@...>stefania@...
>      >
>      > schintu@...
>      >
>      > <mailto:psico@...>psico@...
>      > *********************************
>      > THANK YOU VERY MUCH FOR YOUR COOPERATION
>      >
>      > Grazie per l' attenzione.
>      >
>      > By
>      >
>      >
>      --------------------------------------------------------------
>
>      >
>      > Vi prego di forwardare questa e-mail in tutte le mailing
>      list a cui
>      > siete
>      >
>      > iscritti... per favore...
>      >
>      > --
>      > Massimo D'Eramo
>      > Vice Presidente
>      > Grande Associazione Sterratisti - ROMA
>      >
>      > <http://www.ics.it/gas>http://www.ics.it/gas
>      > e-mail:<<mailto:gas@...>gas@...>
>
>      ************************************************************************
>
>      * Bertellotti Alessandro * IRC Nick (on EFNET): Roccia *
>      * mailto:abertel@... * PowWow: abertel@...
>      *
>      ************************************************************************
>
>      * HOME-PAGE "Il primo sito italiano dedicato alle ARTI
>      MARZIALI" *
>      * "MARTIAL NET" --> http://www.programatic.it/websurf/dojo/
>      *
>      ************************************************************************
>
>      * ISCRIVITI ALLA PRIMA MAILING-LIST ITALIANA SULLE ARTI
>      MARZIALI *
>      * MAILTO:majordomo@... - BODY: subscribe dojo *
>      ************************************************************************
>
>      ++++++++++++++++++++++++++++++++++++++++++++++++
>      Ing. Massimo Carli
>      Java and Internet Developer
>      tel. +39 425 362885
>      (0338) 7372320
>      email : carlim@...
>      carli@...
>      Collaboratore Mokabyte
>      http://www.programmers.net/Riviste/moka/current/
>      +++++++++++++++++++++++++++++++++++++++++++++++++
>
> <<<<
>
> Ing. Luciano Ongaro
> Netsy
> software for the network computing age
> a division of Intesys Srl
> via Latason, 5
> 36030 CALDOGNO (VICENZA) - Italy
> tel +39 444 905440 fax +39 444 905441
> Email:<mailto:netsy@...> Web <http://www.netsy.it>

--
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
                   Christiane Boehner
     Joint Research Centre of the European Commission
       Institute for Systems,Informatics and Safety
                  Tecnology Assessment
             T.P. 361, I - 21020 Ispra (VA)
                ++      ++      ++     ++
                 phone: ++39 332 789517
                 fax:   ++39 332 789394
             e-mail:christiane.boehner@...
     http://taws03.jrc.it/epu/christiane/boehner.html
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear all,

I have a geological data set and I'd like to compare the
interpolation results from different methods (kriging, minimum
curvature and IDW). I have an "ideal" map from geologists and this
is the map of isolines. The question is, how to compare grids which
comes from interpolation methods and this isolines map? I see two
possibilities here:

1) to compare grids. Then I need to interpolate the isolines map to
grid. And this means, I will possibly get some differences due to
this interpolation., not due to "goodness" (or "badness") of
interpolation method of interest. Another question, what is a
criteria to compare two grids? A mean squared difference? And first
of all, I need an algotithmus to interpolate isolines to grid. I
found a description of such method (it calls "Profiles" in TNTmips or
INTERCON in IRDISI), but descriptions are not detailed enough to
write a code. So, does anyone have such a code (or pointers to that)?

2) another possibility is to compare two sets of isolines, but I have
no idea, how.

If you have ideas about all that, drop a couple of lines, I will
summarise the responses.

Thank you in advance,
Kostantin.
Konstantin Malakhanov, wiss. Mitarbeiter/research engineer

IWW, RWTH Aachen
Tel. 0241-807343
Fax. 0241-8888348
E-Mail: kosta@...
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Thanks to Roberto Bruno <roberto.bruno@...> for his
response.

To clear my situation:

1) the data is from  quarternary stratum with about 390 data points at
5km^2 available, so I have no maps from satellite images and no
"true" grid values.

2) The isoline map is drawn per hand directly from data points (there
are wizards among geologists, they can do such things), so there is no
intermediate grid between data points and isolines map.

  If I understand you right, to get a grid from isolines map, you
advise to perform kriging of both data set points and isolines(at
vertex points)? But doesn't that mean loose of information - with isolines I
know the gradients of surface, with data points- no. Or should one
rasterize the isolines?

And what is "test-kriging" ?

And I'm still looking for detailed description of linear "isolines-to-grid"
interpolation....

Thanks,

Konstantin.
Konstantin Malakhanov, wiss. Mitarbeiter/research engineer

IWW, RWTH Aachen
Tel. 0241-807343
Fax. 0241-8888348
E-Mail: kosta@...
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear Geostats group

I have produced semivariograms for some data and for most variables there is

strong evidence of a trend across the sample plot.  I am interested to

remove this to gain information about the underlying spatial dependence in

the data despite this main trend. I have checked for outliers, and also I

have tried to do median polishing in S-Plus, but using twoway, all my

residuals are zero presumably because my data is not of the right format -
Can anyone make suggestions ?  The purpose of the work is to determine the

scale of spatial dependence so as to make inferences about the best way to

sample.

I would greatly appreciate any comments

Best wishes

Alison



________________________________
Dr Alison Riding
ADAS
Woodthorne
Wolverhampton
WV6 8TQ
U.K.

Tel:   +44 (0) 1902 693129
Fax:  +44 (0) 1902 693166
E-Mail:  Alison_Riding@...
________________________________
Use Proportional Font: true
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Re: Konstantin Malakhanov's question:

One way would be to contour the map using GEOEAS, for example, (refer the
Software FAQ on the AI-GEOSTATS Web server). Then compare the map
with the geological map (it would be better to use the same isocontour
interval).
Nine out of ten times you'd probably get a totally different map, unless the
interpolated one had "soft" information built in right from the start. I once
had
to contour a distributary channel system from a couple of control wells, and
then compared that with a kriged map - both looked like it came from two
different
planets.

Regards, Syed

PS/- This man versus machine concept is getting pervasive, re Deep Blue
versus Kasparaov, or a computer versus a geologist.


--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Hello all. I was wondering if somebody could point me to
a site where unix geoeas binaries were available (under
Solaris 2.x). I've been trying to compile the sources
but I've had no success.

Thanks in advance.

P. Sanz
University of Seville
Spain. EU

--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

One must be a little careful, kriging is not a unique
interpolation method hence to say that one is going to
compare "kriging" with "IWD" does not make sense

First of all, is the mean treated as constant or a polynomial
(if the latter what terms are included, i.e., universal
kriging). Even if the mean is considered constant is it
considered known or unknown (simple vs ordinary kriging).
What variogram/covariance model(s) are used, what parameters
(including possible anisotropies). Is a unique neighborhood
used or a moving neighborhood, if the latter what conditions
are placed on it (radii, minimum/maximum number of points,
sectors, # of empty sectors?).

Note that the thin plate spline is a special case, i.e.,
a particular covariance.

Even IWD can vary, see Kane et al (Computers and Geosciences
1982) on an empirical study of the effect of the exponent.

To construct the contour map one must first interpolate onto
a regular grid, the choice of the grid mesh for interpolation
will have some effect.

Are point values interpolated or block values (one might
use either for contouring)?

What is the ultimate use of the contouring? That will have
some bearing on how to judge the comparison.

Finally the comparison will be valid only for the specific
data set and does in general not imply anything about
the comparison for a different data set.

Wanting to compare different possible results is an
interesting/useful question but it will not have a
definitive answer.

Donald E. Meyrs
Donald E. Myers
Department of Mathematics
University of Arizona
Tucson, AZ 85721
http://www.u.arizona.edu/~donaldm
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Konstantin Malakhanov wrote:
>
>
> 2) another possibility is to compare two sets of isolines, but I have
> no idea, how.


I think comparing two sets of isolines doesn't make much sense in general unless
you
have used the same interpolation method or if you have a high number of isoline
levels.
If isolines are handy to use, they also have the magic property to hide
unwanted/desired
information. You never know what is behind the isolines. When drawing your
isolines,
you may have similar  general patterns compared to the "ideal" map but you won't
know
how the underlying data behave. Changing the isoline levels may show very
different
patterns compared to those hidden in the "ideal map".

Converting isolines to a grid rises the same question: how does the data really
look like ?

How would you define the mean values between 2 levels ? Nothing can tell you
that
the average is the arithmetic mean. It may be much lower or higher.

The "ideal" map has been drawn by hand. Therefore the only way to approach the
hand
drawn map would be to convert all assumptions used by the author about the
spatial variability of the analysed variables, the nature and the scale of
the variability into an interpolation model. I hope we are not too far in time
from
a geostats version of Deep Blue...


To be more pragmatic: if you convert your isolines into a grid, you could first
convert
the lines of the isolines into several (depending on the original resolution)
points which
will receive the values defined by the associated levels. Analysing the spatial
correlation
(exploratory variography) of these points may help you to understand the spatial
structure
of the data. The semivariogram can also be compared to the original data. (A
problem is that
you are then working on categorical data and not anymore on supposed continuous
data).
Such approach may help you to recreate the hand drawn map (with kriging)
but again your information may be strongly different from it if the assumptions
are different.

Gregoire


--
Gregoire Dubois (PhD student)   Tel. 39-332-78.99.44
Joint Research Centre           Fax. 39-332-78.54.66
Environment Inst. TP 321        Email: gregoire.dubois@...
I-21020 Ispra (Va), ITALY       URL: http://java.ei.jrc.it/rem/gregoire/
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

auth c7a6e7f6 subscribe ai-geostats dl.craggs@...
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

We are proud to announce the First Isatis Users Meeting adressing the
Oil and gas Industry :


                     The participants:
                     ----------------

Agip :                         Livia Bazzana
Amoco:                         Mike Ross and Steve Winter
Anadarko Algeria Oil :         Richard Drumheller, Todd Fowler and
                                Akli Tegmount
Beb  :                         Willy Hack
Enterprise Oil:                Jon Wardell
Gaz de France:                 Bertrand Coureaud
Geomath :                      Jeffrey Yarus
Institut Francais du Petrole : Brigitte Doligez and Abdelaziz Faraj
Lasmo Plc:                     Ashley Francis
Nam :                          E. Gevers
Shell :                        Yvan Pannatier
Statoil :                      Valerie Langlais
Total :                        Jean Luc Piazza


Centre de Geostatistique:
    Christian Lantuejoul
    Didier Renard
    Yves Touffait
    Yuri Voitsekhovsky
    Hans Wackernagel

Geovariances:
    Jacques Deraisme
    Brigitte Hogan
    Frederic Rambert
    Ludovic Robinot


you can reach them all together using the e-mail adress:

                   users-meeting@...


Venue: Hotel Mercure
        41, rue Royale
        77300 Fontainebleau
        tel 33 (0) 1 64 69 34 34
        fax 33 (0) 1 64 69 34 39

                         Program
                         -------

Monday, June 9, 1997

-9.00  Welcome
        Introduction of the participants (Jacques Deraisme
                                          Geovariances)

10.00  "A standard tool for multivariate EDA and Modeling: Principal
        Component Analysis" (Hans Wackernagel
                             Centre de Geostatistique)

10.45  Break

11.15  "3D Velocity Modeling Building using a Geostatistical Approach"
        (Mike Ross - Amoco)

12.00  Lunch

14.00  "Uncertainties in the volume in place, methodology and
         implementation" (Valerie Langlais
                          Statoil)

14.45  "Volumetrics post-processing in Isatis" (Didier Renard
                          Centre de Geostatistique)

15.30   Break

16.00   Import/Export (Frederic Rambert
                        Geovariances)

16.30   Review and discussion

Tuesday June 10, 1997

-9.00   "Iterative Algorithms for conditional simulations"
                        (Christian Lantuejoul
                         Centre de Geostatistique)

-9.45    "Geovel, a steering geostatistical tool" (Livia Bazzana
                                                    Agip)

10.30     Break

11.00     "Kriging with gradients and external drift"
                          (Yuri Voitsekhovsky
                           Centre de Geostatistique)

12.00     Lunch

14.00     Support and maintenance (Frederic Rambert
                                    Geovariances)

14.30     Open discussion on your requests,
           Isatis version 3.1,
           Future releases
                       (Jacques Deraisme &
                        Frederic Rambert
                        Geovariances)

15.30     Break

16.00     Conclusion (Jacques Deraisme
                       Geovariances)

you can read the available abstracts on our Web pages:

                    http://www.geovariances.fr




---------------------------------------------------------------------
Brigitte Hogan
Sales Manager
GEOVARIANCES - 38, Av. F. Roosevelt - 77210 AVON - FRANCE
Phone: +33-(0)-160.74.90.97   -   Fax: +33-(0)-164.22.87.28
Web: http://www.geovariances.fr
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Don

Thank you for your response to my question about trend removal.  I have used

the loess function in S-Plus,  and analysing the residuals gave quite a

'good' semivariogram.

  Presumably you would now use the semivariogram of the residuals to krige

with?

Also, do you (or anyone else) have a reference that describes this use of

loess residuals?

Thanks a lot for your response, it was very helpful - I will shortly write a

summary of all the responses I had.

Best wishes

Alison
________________________________
Dr Alison Riding
ADAS
Woodthorne
Wolverhampton
WV6 8TQ
U.K.

Tel:   +44 (0) 1902 693129
Fax:  +44 (0) 1902 693166
E-Mail:  Alison_Riding@...
________________________________
Use Proportional Font: true
Previous From: dmck@... (Don McKenzie)
Previous To: Alison Riding@ADASMail
Original to: Alison Riding@ADASMail
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Oooops, a small correction is needed:

you should read:

"The Candidate should have experience preferably in the
following areas (in priority order):


2) C++ programming (basics required)  "


instead of

"The Candidate should have experience preferably in the
following areas (in priority order):


2) GIS C++ programming (basics required)"




--
Gregoire Dubois 	 Tel. 39-332-78.99.44
Joint Research Centre           Fax. 39-332-78.54.66
Environment Inst. TP 321        Email: gregoire.dubois@...
I-21020 Ispra (Va), ITALY       URL: http://java.ei.jrc.it/rem/gregoire/
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

********************************************************

POST-DOC Position

********************************************************

The Radioactivity Environmental Monitoring Unit (REM) of
the Environment Institute (EI) of the Joint Research Centre
(JRC) of the European Commission is looking for Candidates for a

-- Post-Doc Position (maximum 2 years) --

for:

THE DEVELOPMENT OF A GEOGRAPHIC INFORMATION SYSTEM (GIS)
BASED ON GEOSTATISTICS AND SPATIAL ANALYSIS FUNCTION
FOR THE MONITORING AND THE VALIDATION OF RADIOACTIVITY
MEASUREMENTS IN THE ENVIRONMENT


Aim of Work to be performed:

Within a team of 3 persons, the candidate will help:

- to design and develop a GIS,
- to integrate within the GIS advanced spatial statistics
- to develop methodologies for the monitoring of radioactivity
   in the environment.


The Candidate will participate in the following activities:

1) to design and develop a GIS based on a C++ development
toolkit

2) to develop and integrate geostatistics and spatial statistics
function within the GIS

3) to adapt the developed tools to the monitoring of environmental
radioactivity through case studies

4) to develop interfaces between the GIS tool and a database (Oracle)


The Candidate should have experience preferably in the
following areas (in priority order):

1) Spatial statistics, geostatistics (required)

2) GIS C++ programming (basics required)

3) GIS

4) Environmental monitoring

5) Database management

The candidate must have the citizenship of an EU Member State
and have accomplished a PhD.

Probable starting date for the work: January, 1998.

Deadline for subsmission of the proposal: 15th of July 1997

Interested candidates should send their CV by e-mail or FAX
before the 20th of June 1997 to:


Gregoire Dubois 	 Tel. 39-332-78.99.44
Joint Research Centre           Fax. 39-332-78.54.66
Environment Inst. TP 321        Email: gregoire.dubois@...
I-21020 Ispra (Va), ITALY
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

I have been reading with interest the responses to the above inquiry.  I
sent my response directly to Konstantin Malakhanov. I am forwarding it
in this message.
That is right! Comparing isolines from different interpolation methods
may not be a useful thing to do.  Gridding may not be useful either..
Alaa
_________________________________________________________
A. Ali, Ph.D., P.E.
New Mexico State Engineer Office
P.O.Box 25102, Santa Fe, NM 87504
e-mails:  aali@...; or aali@...
Web: http://www.engineering.usu.edu/Departments/cee/Faculty/ulall/
Phone: 505-827-6125           Fax: 505-827-6188
_________________________________________________________

The most suitable way to do it is to use the principle of cross
validation. Given n data points, perform the following for each method:
1) drop data point i, and use the remaining n-1 data points to provide
an estimate at the location of the dropped point.
2) compute the err=y_hat_i - yi,
where y_hat_i is the estimated value at location i
       y_i is the value of dropped data at location i
3) repeat steps 1 and 2 at all n points, and compute sum of squared
error (or the mean square error).
4) repeat steps 1,2,3 for all methods; then compare the resulting sum of
squared errors (or the mean square error).

As you notice, we did not have to create a grid.  However, the method
with least errors is the one that will give the best estimation grid in
terms of this criterion.

Regarding the geologist isoline map, another estimation method, is
subjective and may or may not be better than the a statistical
interpolation method. I am not sure if you can use a statistical method
to judge the adequacy of fit of such a map. However, you can select the
interpolation method that provides the closest estimate to the isoline
map. To do so, select as many points as you can on these isolines, and
consider them data points, then perform cross validation as described
above. select the interpolation method with the least square error
score.

Note: I don't recommend the use of grids to solve your problem.
if you have any further questions, don't hesitate to contact me.
Alaa
_________________________________________________________
A. Ali, Ph.D., P.E.
New Mexico State Engineer Office
P.O.Box 25102, Santa Fe, NM 87504
e-mails:  aali@...; or aali@...
Web: http://www.engineering.usu.edu/Departments/cee/Faculty/ulall/
Phone: 505-827-6125           Fax: 505-827-6188
_________________________________________________________

Call for Papers: 31 May 1997

International ICSC Symposium on Engineering of Intelligent Systems

University La Laguna Campus, La Laguna, Spain
(11 - 13 February 1998)


Applications of Artificial Intelligence have been extended thanks to the
development of new fields, such as
Neural Networks, Fuzzy Logic and Genetic Alogrithms. The possibility of accurate
modeling of complex problems
permitted by these new tools allows an approach to new applications of
artificial intelligence in
control theory, robotics, signal processing and even business, financial
planning management, damage
assessment, security, etc.

The main aim is to bring together theorists from different fields of artificial
intelligence and people
working in applicative areas, in order to state the current state of the
research and applications of
Intelligent Systems and make the basis for rewarding cooperation in the the
area.


More at http://www.compusmart.ab.ca/icsc/eis98.htm




Gregoire
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

>The most suitable way to do it is to use the principle of cross
>validation. Given n data points, perform the following for each method:
>1) drop data point i, and use the remaining n-1 data points to provide
>an estimate at the location of the dropped point.
>2) compute the err=y_hat_i - yi,
>where y_hat_i is the estimated value at location i
>      y_i is the value of dropped data at location i
>3) repeat steps 1 and 2 at all n points, and compute sum of squared
>error (or the mean square error).
>4) repeat steps 1,2,3 for all methods; then compare the resulting sum of
>squared errors (or the mean square error).
>
>As you notice, we did not have to create a grid.  However, the method
>with least errors is the one that will give the best estimation grid in
>terms of this criterion.
>
>Regarding the geologist isoline map, another estimation method, is
>subjective and may or may not be better than the a statistical
>interpolation method. I am not sure if you can use a statistical method
>to judge the adequacy of fit of such a map. However, you can select the
>interpolation method that provides the closest estimate to the isoline
>map. To do so, select as many points as you can on these isolines, and
>consider them data points, then perform cross validation as described
>above. select the interpolation method with the least square error
>score.
>
>Note: I don't recommend the use of grids to solve your problem.
>if you have any further questions, don't hesitate to contact me.
>Alaa
>_________________________________________________________
>A. Ali, Ph.D., P.E.
>New Mexico State Engineer Office
>P.O.Box 25102, Santa Fe, NM 87504
>e-mails:  aali@...; or aali@...
>Web: http://www.engineering.usu.edu/Departments/cee/Faculty/ulall/
>Phone: 505-827-6125           Fax: 505-827-6188
>_________________________________________________________
>

Would it be use an interpolation method to compute a grid, contour the grid
using the same levels as the hand-drawn map, and then compare the contours
directly, perhaps by calculating the area between corresponding contours in
each version? We would have to use the map boundaries to close the polygons
formed by the contour line pairs. This would do away with having to grid
the hand-contoured map.
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

The cross-validation method suggested by A. Ali should be used
with some care. In the case of kriging (which would include
thin-plate splines) the expected value of the mean square error
would be the sum of the kriging variances. In turn the kriging
variances are determined by the variogram/covariance model(s)
(and their parameters), the neighborhood parameters and
the spatial pattern of the data locations. The kriging variances
  are NOT affected by the actual data values. Cross-validation
is more appropriately used to validate the choice of the
the variogram model and to identify unusual data values.

A slight variation on what Ali suggested would be to segregate
the data into two sets, use the one to estimate the others and
use the mean square error. However this is somewhat determined
by the segregation choice and also whether the modeling is done
on the entire data set or only on the control set.

There are obviously other considerations, for example do you
want the interpolating function to be smooth, e.g. a thin
plate spline, do you want to preserve connectedness of highs
, lows?

What is the ultimate use of the contour map and what are the
risks associated with using a less than good contouring method?

It is a delusion to think that there is an absolute way to
compare "methods", it definitely depends on the data set unless
you impose external conditions.

Donald E. Myers
Department of Mathematics
University of Arizona
Tucson, AZ 85721
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear ALL
	 A few weeks ago i directed a question asking for assistance in
regards to models of the form,

         Y=Beta.W.Y + Xb+error    (1)

Where error ~ N(0,sigma^2), W= weights incidence matrix, Y=vector of
observed values,  and Beta=spatial interaction parameter to estimate.

I thank you for your responses, and i was informed to look at the work of
Luc Anselin, Robert L. Haining, Daniel A. Griffith, Cliff-Ord, Brian
Ripley and other key players in this area.

However, I have done a fair bit of reading over the past couple of weeks
and haven't come across on just how to proceed if we have more than one
component of variance to estimate other than the error, that is a model of
the form;

	 Y=Beta.W.Y +Xb+ Zu + error                  (2)


where Zu is  a random effect. And hence

V(Y)=(1-beta.W.Y)^-2(sigma(b)^2ZZ'+sigma(e)^2) .

If we don't have the Beta.W.Y term , estimation of the variance componets
sigma(b) and sigma(error) is fairly easy,but including the spatial
interaction parameter ( Beta.W.y) there isn't any refernces at all, i
believe to it in the literature.

I would be more than grateful if anyone knew of any key papers that
outline estimation theory,and software to use etc.. in order in fitting
models of the form (2) when we have both random/fixed and a beta.W.y to
estimate.

I thank you dearly for your time.

regards Vince

pHD student








--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear all,

This is the summary of responses to my request about comparing two surfaces.
The original question was (briefly):

I have some geological measurements (~390 points at 5km^2). I would
like to compare the results of different interpolation methods (IDW,
kriging, minimum curvature etc.) with "ideal " map drawn (actually
digitized) per hand by geologists.

What are the possible techniques to compare two surfaces?

The idea of mine was either a) to make a grid from isolines map and
to compare two grids or b) to contour the grids from interpolation
methods and then compare isolines.

In both cases, I need comparison criteria and for b) an
"isolines-to-grid" algorithm too.

Many thanks for the answers to Gregoire Dubois
<gregoire.dubois@...>, Donald Myers <myers@...>,
Edson Ribeiro <edson@...>, Roberto Bruno
<roberto.bruno@...<, Steven Zoraster
<szz@...>, Syed <pepsi@...>, Alaa Ali
<Ali_Alaa@...>, Helmut Schaeben
<schaeben@...>, Med Bennett <mbennett@...>,
Glenn Krum <glkrum@...>.

I can sum up the answers as follows:

1) "pessimistic" one ;-) - "comparing isolines from different
interpolation methods may not be a useful thing to do. Gridding may
not be useful either..." (from A.Ali)

2) "not so pessimistic" ones - just visual comparison (Glenn Krum),
because hand-drawn map is subjective and it is difficult to compare
subjective and objective (Robert Bruno, Syed)

3) comparing isolines.

a) calculate area between corresponding contours (M. Bennett) (but
this says nothing about where are the errors)

b) calculate volume differences after generating surfaces from
isolines (by triangularization) (E. Ribeiro)

c) use data points from isolines and perform cross-validation for
interpolation methods, the method with least square method wins
(A.Ali) [remark of D. Myers - this method should be used with care,
as cross-validation in general] (I found the nice discussion of
crossvalidation's uses and abuses in Srivastava and Isaac).

4) comparing grids - one have to compare grids, as isolines are results of
(maybe implicit)
interpolation (R. Bruno).

To get a grid from isolines use DLG2DEM algorithm from USGS (thaks
to Steven Zoraster, Landmarks Graphics) or CTOG from Landmark
Graphics (thanks to Glenn Krum).  Another possibility is given by
AUERBACH & SCHAEBEN, 1990, SURFACE REPRESENTATIONS REPRODUCING GIVEN
CONTOUR LINES: Math.Geol. 22, 723-742, featuring constrained Delaunay
triangulation (thanks to Helmut Schaeben).  I wasn't able to find
DLG2DEM or CTOG on the net and I will look for Math. Geol. 22 in our
university's library.

Or one can use original data points + isolines points (or only
isolines points) to krige it (the problem - we have "another
interpolator to judge" E. Ribeiro).


To compare grids - build maps of differences, and then use all
statistics' power (mean, variance, scatterplot, correlation ranks etc)
and variograms of differences (G. Krum, R. Bruno)

The comparison method, common to both isolines and grids, is
comparison of variograms of "ideal" and interpolated map (Gregoire
Dubois and others).

Some general remarks:

from Donald Myers - It is a delusion to think that there is an
absolute way to compare "methods", it definitely depends on the data
set unless you impose external conditions. One must be a little
careful, kriging is not a unique interpolation method hence to say
that one is going to compare "kriging" with "IWD" does not make
sense. Finally the comparison will be valid only for the specific data
set and does in general not imply anything about the comparison for a
different data set.

Wanting to compare different possible results is an
interesting/useful question but it will not have a definitive
answer.

>From Gregoire Dubois - The "ideal" map has been drawn by hand.
Therefore the only way to approach the hand drawn map would be to
convert all assumptions used by the author about the spatial
variability of the analysed variables, the nature and the scale of
the variability into an interpolation model. I hope we are not too
far in time from a geostats version of Deep Blue...

The problem of subjectivity of hand-drawn maps was also mentioned by
many others.

Well, I had no intention to compare IDW or kriging "in general", but
this work was rather seen by me as getting familiarized with
different interpolation techniqes. And I wanted just to be able to
rank methods for their goodness in this partial case.

Thank you all, it was very interesting and informative discussion.

Konstantin.

Konstantin Malakhanov, wiss. Mitarbeiter/research engineer

IWW, RWTH Aachen
Tel. 0241-807343
Fax. 0241-8888348
E-Mail: kosta@...
--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

Dear Vince,

You wrote:

>        Y=Beta.W.Y +Xb+ Zu + error                  (2)
>
>where Zu is  a random effect. And hence
>
>V(Y)=(1-beta.W.Y)^-2(sigma(b)^2ZZ'+sigma(e)^2) .
>
>If we don't have the Beta.W.Y term , estimation of the variance componets
>sigma(b) and sigma(error) is fairly easy,but including the spatial
>interaction parameter ( Beta.W.y) there isn't any refernces at all, i
>believe to it in the literature.

Of course, it is also easy to estimate SAR models with fixed effect.
For example, Dubin(1992) uses SAR models with Trend Surface, and Can(1990,
1992) estimates SAR models with Spatial Expansion. These models treat
spatial heterogeneity in real estate market as fixed effect.
On the other hand, Jones & Bullen(1994) operate fixed/random effect models
of house prices. They does not simultaneously consider spatial
autocorrelation in the models, but they comment on SAR in the paper.
Personally, it is very complicated to estimate SAR models eith random effect,
and I have not found the literature about such models.

REF:    Can, Ayse (1990), Economic Geography 66, 254-272.
         Can, Ayse (1992), Regional Science & Urban Economics 22, 491-508.
         Jones, Kelvyn and Nina Bullen (1994), Economic Geography 70, 252-272.

Sincerely,
Hajime Takatsuka
----------------------------------------------
                              Hajime TAKATSUKA
Higuchi Lab, Department of Social Engineering
                 Tokyo Institute of Technology
  2-12-1 Ookayama, Meguro-ku, Tokyo -152 JAPAN
               E-mail:takatsu@...
                          TEL:+(81-3)5734-2808
                          FAX:+(81-3)5734-2926
----------------------------------------------

--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!

at
   1 year interval, same dates for all the series). I would like to summarize
the
   data by:

    1 - Classifying the binary time series

    2 - Analyzing the spatial distribution of the resulting classes.

   ... but I'm stuck at the first step!

    How can I make a classification of binary time series?  Can I use a
classical
   cluster procedure with  the appropriate coefficient (Jaccard for example)?

   Thank you in advance.

   Jean-Noel CANDAU
   Canadian Forest Service
   Sault-Ste-Marie - ON
   Canada


--
*To post a message to the list, send it to ai-geostats@....
*As a general service to list users, please remember to post a summary
of any useful responses to your questions.
*To unsubscribe, send email to majordomo@... with no subject and
"unsubscribe ai-geostats" in the message body.
DO NOT SEND Subscribe/Unsubscribe requests to the list!