... I was retraining a statistician last year for more adventurous modelling, and had him generate data according to a model, then recover the model (ie, some...
726
T.S. Lim
tslim@...
Oct 19, 2000 9:24 am
... There're some resources listed in the "Data Sets" section of http://www.kdcentral.com For example, Dataset Generator (DatGen). -- T.S. Lim ...
727
Thomas Hellström
thomash@...
Oct 20, 2000 10:05 am
Does anyone know of investigations done concerning optimal values for k in the k-nn algorithm ? It should be affected by things like: 1. size of the training...
728
Samy Bengio
bengio@...
Oct 20, 2000 11:02 am
... you could use simple cross-validation techniques to set K. for instance, if you have enough data, train many k-nns with different K values over a training...
729
Gianluca.Bontempi@...
Oct 20, 2000 11:18 am
Our work on lazy learning proposed a fast cross-validation technique to find the optimal number of neighbors. You can take a look at these web pages ...
730
Ian Cloete
ian.cloete@...
Oct 20, 2000 12:36 pm
*************************************************************************** CALL FOR PAPERS ... ...
731
Vitorino Ramos
vitorino.ramos@...
Oct 20, 2000 3:17 pm
Hi Thomas: I have dealed with this problem with another approach. Using Genetic Algorithms to explore all the combinations for a specific problem. Here is the...
732
Mathias Kirsten
mathias.kirsten@...
Oct 21, 2000 9:34 am
Dear Thomas, although not very recent, it may be worth to have a look at @Article{Milligan:85, author = "G. W. Milligan and M. C. Cooper", year =...
733
Davide Roverso
roverso@...
Oct 23, 2000 8:44 am
This is a call for participation to an International cooperative effort for the definition and realization of one or more benchmarks for transient...
734
Franklin Wayne Poley
culturex@...
Oct 24, 2000 1:34 am
Subject: [FUTURE-CITIES] Verbot for Premier Dear Mr. Mauldin: <http://www.vperson.com> I expect there will be a BC Election in a few months and that before the...
735
Ron Blue
rcb5@...
Oct 25, 2000 10:07 am
http://washingtonpost.com/wp-dyn/articles/A17618-2000Jul20.html In light of the finding that IQ is strongly linked to a small area in the frontal lobes I would...
736
Robert BURBIDGE
r.burbidge@...
Oct 25, 2000 12:36 pm
[apologies if you receive this more than once] CALL FOR PARTICIPATION ++++++++++++++++++++++ YOR12 Data Mining Stream The University of Nottingham 27-29 March...
737
Martin Sewell
martin@...
Oct 28, 2000 1:40 am
Hi New Support Vector Machine mailing list: http://www.jiscmail.ac.uk/lists/support-vector-machines.html Cheers Martin...
738
T.S. Lim
tslim@...
Oct 31, 2000 2:24 am
... Provide details! What's SVM? What's the goal of the list? Who should join? Is the list moderated/unmoderated? Follow the excellent service provided by...
739
guido@...
Nov 1, 2000 1:12 am
Hello, Neural Networks are black-boxes. It's difficult to explain their responses. In many classification problems "If-Then ..." rules are useful to explain ...
740
a.allen@...
Nov 1, 2000 1:12 am
University of Aberdeen Department of Engineering Research Studentship available: Neural systems engineering for embedded machine vision Applications are...
741
Martin Sewell
martin@...
Nov 1, 2000 1:45 am
... The Support Vector Machine (SVM) is a powerful new learning algorithm based on recent advances in statistical learning theory. SVMs have become one of the...
742
Michael Goebel
mgoebel@...
Nov 1, 2000 10:07 am
Dear all, my name is Michael, I'm a PhD student within the ML group at the University of Auckland, New Zealand. Does anybody know where I can find a copy of...
743
correspondence@...
Nov 1, 2000 7:07 pm
Please note the November issue of SecondMoment has been posted at http://www.secondmoment.org . This month we are featuring a discussion with Dr. Hal White on...
744
Franklin Wayne Poley
culturex@...
Nov 2, 2000 11:20 pm
If I were to connect a camera to my desktop pentium, is there software available which would enable it to "itemize" all objects in a typical, cluttered room...
745
Franklin Wayne Poley
culturex@...
Nov 3, 2000 10:05 pm
... Date: Fri, 3 Nov 2000 14:04:24 -0800 (PST) From: Franklin Wayne Poley <culturex@...> Reply-To: FUTURE-CITIES@egroups.com To:...
746
T.S. Lim
tslim@...
Nov 5, 2000 2:54 pm
... Get paid to write review! http://recursive-partitioning.epinions.com...
747
Franklin Wayne Poley
culturex@...
Nov 5, 2000 9:00 pm
If I analyze the visual field before me, I am aware of objects within edges and the important fundamental variables which define those objects are hue,...
748
beimfohr@...
Nov 6, 2000 12:51 am
Regarding a technical simulation, I have the task to construct several recurrent networks, one for each parameter setting of the experiment. These different...
749
Borys Omelayenko
borys@...
Nov 6, 2000 2:31 pm
Dear colleaques, The network of excellence called OntoWeb: Ontology-based information exchange for knowledge management and electronic commerce...
750
Alexander G. Parlos
a-parlos@...
Nov 6, 2000 9:00 pm
... Hi Frank, Our experience w/ recurrent networks for forecasting (or multi-step-ahead prediction in general) has been quite positive. But at the same time...
751
Robert Grundulis
robgru@...
Nov 7, 2000 1:19 am
Hello All! Now this may seem stupid asing this, but I would appreciate any help you can offer me on this... Could anyone tell me some real-world examples where...
752
Gianluca.Bontempi@...
Nov 7, 2000 8:48 am
It is not neural network but anyway it could help you. One of our paper on tuning local learning methods for multi-step-ahead prediction in time-series...
753
Dr. Pralhad Rao
rao@...
Nov 7, 2000 2:18 pm
Please inform me how to generate gaussian clusters for the numerical data. Thanks. P.R.Rao. ... Mailing Address :- Dr.P.R.Rao,Reader, Department of Computer...
754
Franklin Wayne Poley
culturex@...
Nov 7, 2000 3:28 pm
... Anon: I'm not concerned about also finding those "spurious bits". If you get patterns and details within edges so what? The task is to itemize all the...
