LECTURER/RESEARCHER SCHOOL OF SYSTEMS AND DATA STUDIES The following has appeared at http://www.jobs.ac.uk/jobfiles/IA159.html. See also...
dlyons@...
Sep 1, 1999 8:49 am
223
Call For Papers E-Commerce and Data Mining Special issue of the journal Data Mining and Knowledge Discovery Guest editors: Ronny Kohavi and Foster Provost Will...
Foster John Provost
foster@...
Sep 1, 1999 7:15 pm
224
Hi, there Does anyone know a good survey on the existing techniques of data mining? Thanks. -KC...
chen@...
Sep 2, 1999 3:57 am
225
Hi There is a survey at ftp://ftp.fas.sfu.ca/pub/cs/han/kdd/survey97.ps.gz. Is it good? Read and decide yourself. Ruslan Kravets....
Ruslan Kravets
ruslan@...
Sep 2, 1999 10:11 am
226
chuc-@... wrote: original article:http://www.egroups.com/group/machine-learning/?start=2 25 ... gz. ... The above survey is from a database...
chuck@...
Sep 2, 1999 8:50 pm
227
lim-@... wrote: original article:http://www.egroups.com/group/machine-learning/?start=2 26 ... =2 ... s. ... r/ ... pr ... "A Survey of Data...
limt@...
Sep 2, 1999 10:27 pm
228
The following could be of general interest: "Keyword Indexed Bibliography on Evolutionary Computation" ...
Tjen-Sien Lim
limt@...
Sep 2, 1999 10:41 pm
229
Hi, everybody, Is it possible to build a security enhancing tools Using machine learning techniques such as Bayesian network (maybe analyze some log files and...
Gang LI
yqtuglic@...
Sep 6, 1999 6:11 pm
230
I'm wondering if someone has the electronic copy (in PostScript or PDF) of the following paper. Our local library doesn't have the 1995 proceeding and it'll...
Tjen-Sien Lim
limt@...
Sep 7, 1999 12:56 am
231
... IBM has used bio-sequence pattern discovery program called TEIRESIAS to "learn" all common network traffic usage patterns on IP level (I think). Then, all...
Jaak Vilo
vilo@...
Sep 7, 1999 8:12 am
232
Back in 1993 WizSoft's WizRule presented a similar approach. WizRule is includes an association rules algorithm that rveals all the if-then rules in the data,...
Abraham
abraham@...
Sep 7, 1999 8:54 am
233
MANNING & NAPIER INFORMATION SERVICES ROCHESTER, NY Manning & Napier Information Services (MNIS) develops and deploys cutting-edge technology for large-scale...
randy@...
Sep 7, 1999 7:44 pm
234
Hello: Who could help? I am looking for a good fitness function for IF THEN rules, which are produced depending on attributes and their values within a...
Christian Setzkorn
m9cs@...
Sep 8, 1999 3:22 pm
235
Dear Cristian, Factors to consider including in a fitness function are: - accuracy on training examples (but needs a coverage term too). Better still,...
philip@...
Sep 8, 1999 9:13 pm
236
Dear all, ... I should clarify that "validation set" here refers to separate unseen data, not the test data which of course should not be used in training....
philip@...
Sep 9, 1999 4:56 am
237
One method of evaluating the validity of if-then rules is calculating the error probability of the rule. This is the probability the rule exists accidentally...
Abraham
abraham@...
Sep 9, 1999 10:11 am
238
I've just been to a seminar in which the presenter described the self-organising map (SOM). It seems to be similar to what I as a statistician would call...
Murray Jorgensen
maj@...
Sep 10, 1999 4:19 am
239
Opitz, D. and Maclin, R. (1999) "Popular Ensemble Methods: An Empirical Study", Volume 11, pages 169-198. Available in HTML, PDF, PostScript and compressed...
Tjen-Sien Lim
limt@...
Sep 9, 1999 7:31 pm
240
... This is an interesting article. But the methods used for training neural nets seemed to involve no attempt to get good generalization by choosing an...
Warren Sarle
saswss@...
Sep 9, 1999 11:59 pm
241
... See "What are conjugate gradients, Levenberg-Marquardt, etc.?" in the Neural Network FAQ, part 2 of 7: Learning, at ftp://ftp.sas.com/pub/neural/FAQ2.html ...
Warren Sarle
saswss@...
Sep 10, 1999 12:03 am
242
Dear All, I am doing research on shape analysis. I would be grateful for indication all possible good sources of the classification and description of shapes...
Magda Les
Magda.Les@...
Sep 10, 1999 12:35 am
243
Go to http://www.recursive-partitioning.com/bib.html for the link to SOM bibliography. Mailing-List: contact machine-learning-owner@egroups.com X-Mailing-List:...
Tjen-Sien Lim
limt@...
Sep 10, 1999 5:07 am
244
Decision Analysis and Artificial Intelligence for the Management of Biophysical Systems Post-Doctoral research position opened at INRA, Biometry and Artificial...
Frederic GARCIA
fgarcia@...
Sep 10, 1999 7:47 am
245
OK, thats good. I've now got 215 pages containing references for 3343 works about SOM. Do any of these works answer my questions? Murray Jorgensen [Friday...
Murray Jorgensen
maj@...
Sep 10, 1999 11:00 am
246
... You might be better off comparing the SOM to a clustering algorithm. The SOM does not try to preserve the order of distances between observations (ordered...
Harry Zuzan
hzuzan@...
Sep 10, 1999 2:34 pm
247
lim-@... wrote: original article:http://www.egroups.com/group/machine-learning/?start=2 40 ... Empirical ... tml> ... necessarily ... I agree...
limt@...
Sep 10, 1999 10:34 pm
248
... That would be OK if they used early stopping or Bayesian learning. Even if they had trained the networks to convergence, then the results would have said...
Warren Sarle
saswss@...
Sep 11, 1999 1:27 am
249
Dear all, I am looking for good methods to evaluate the results of a trained SOM. Especially after cluster building to conclude some statements about: 'Which...
Christian Setzkorn
m9cs@...
Sep 12, 1999 9:17 am
250
Re: How to identify which variable are important in the SOM model: Once objects are placed into a cluster, the cluster number can be used as a categorical...
William Shannon
shannon@...
Sep 12, 1999 5:24 pm
251
There was a paper on non-linear PCA a few years back (de Emers or something like that [San Diego?] - I have a copy in a different office). It's role was...