Dave Lewis - user overview - MetaOptimize Q+A

Dave Lewis

890

karma

Registered user
member since	Jun 30 '10 at 09:10
last seen	Jul 18 '13 at 10:08

overview recent activity karma history favorites

20 Questions

votes

answers

711

views

choosing threshold to hit a recall target with a specified confidence

Jul 25 '13 at 20:33 digdug245●11●16●20 (asked by Dave Lewis890●20●28●46 )

votes

answers

views

name for 1.0 minus negative predictive value?

Jul 18 '13 at 10:11 Dave Lewis890●20●28●46

negative-predictive-value contingency-table recall precision effectiveness evaluation confusion-matrix

votes

answers

973

views

format for primal representation of linear SVMs in PMML?

Sep 27 '11 at 23:20 Dave Lewis890●20●28●46

svm svm-software pmml format

votes

answers

views

examples of solving the wrong problem in NLP

Aug 27 '11 at 12:33 Dave Lewis890●20●28●46

career applications information-extraction task-definition

votes

answers

872

views

setting up prediction tasks on temporal intervals

Aug 26 '11 at 19:55 Daniel Mahler1226●3●13●22 (asked by Dave Lewis890●20●28●46 )

format multimedia interface streaming-data sequence-labeling time-series hci annotation

votes

answers

views

POLL: what do you call the software component that applies the output of a supervised learner?

Jul 12 '13 at 02:49 Matt0●1●1●3 (asked by Dave Lewis890●20●28●46 )

implementation poll classification terminology software

votes

answers

728

views

examples or best practices for use of MiningBuildTask element in PMML

Aug 21 '11 at 18:36 Dave Lewis890●20●28●46

xml logging pedigree pmml format standards parameter-settings configuration software

votes

answers

959

views

examples of wrapping open source learning software in PMML?

Aug 09 '11 at 22:46 Dave Lewis890●20●28●46

open-source pmml datatype data-model dataset formats standards software

votes

answers

views

(why) does PMML (not) have a way of representing data sets?

Aug 05 '11 at 16:10 Dave Lewis890●20●28●46

xml test-set pmml data-mining format training-set standards representation data-set software

votes

answers

views

avoiding cheating in cleaning up a text classification data set

Jul 08 '11 at 20:18 Robert Layton1625●12●26●37 (asked by Dave Lewis890●20●28●46 )

data-quality overfitting errors preprocessing test-collections noise cleaning datasets labeling

votes

answers

views

PMML pros, cons, and competitors

Dec 06 '10 at 17:38 Alex Guazzelli1 (asked by Dave Lewis890●20●28●46 )

xml pmml implementation documentation standards libraries models api software

votes

answers

views

value of randomization in stochastic gradient descent

Sep 14 '12 at 07:36 denis281●2●8●12 (asked by Dave Lewis890●20●28●46 )

data-ordering online-learning stochastic-gradident efficiency randomization optimization preprocessing external-memory sgd

votes

answers

views

exemplary documentation for open source machine learning code

Dec 08 '11 at 10:01 Gael Varoquaux921●4●14●26 (asked by Dave Lewis890●20●28●46 )

online-learning java open-source implementation documentation libraries logistic-regression freely-available software

votes

answers

views

estimating speech recognition accuracy from a sample

Dec 15 '10 at 17:55 dataengines16●1 (asked by Dave Lewis890●20●28●46 )

transcription proportion sampling speech speech-recognition estimation evaluation annotation labeled-data accuracy

votes

answers

views

Text classifiers using accidental features

May 07 '11 at 01:18 Joseph Turian ♦♦5790●51●125●146 (asked by Dave Lewis890●20●28●46 )

overfitting nlp proper-nouns robustness generalization concept-drift text-classification feature-selection

View all questions by Dave Lewis

22 Answers

Libsvm and precision-recall break-even point (BEP)

name for 1.0 minus negative predictive value?

term weighting in text classification

Computing precision/recall curve for unbalanced multilabel document retrieval

SVM vs. Logistic Regression for classification

Most Influential Ideas 1995 - 2005

What is the best way to classify news articles on 50+ topics?

Idiot-proof Machine Learning Methods

Start with a blank approach to an NLP task, or read a lot before starting?

Online, unsupervised technique for inducing document representations?

training or test error for online learning?

Incremental Maximum Entropy Classifier

Adding noise to (text) data

Author identification

Effects of Training data order for Supervised Learning Tasks

Does document length matter in text categorization?

NLP with a linguistic background

What is the best categorizer approach for text tag prediction?

Document Recognition and Classification - Software / Models

Do styles matter in text classification?

The best machine learning consultants?

Algorithm to discover features from raw data?

16 Votes

format × 9
optimization × 7
regularization × 7
external-memory × 7
stochastic-gradident × 7
learning-theory × 7
sgd × 7
speech × 7
hyperplane × 7
sequence-labeling × 7

efficiency × 7
margin × 7
data-ordering × 7
randomization × 7
linear-algebra × 7
computational-geometry × 7
combinatorics × 7
streaming-data × 6
transcription × 6
labeled-data × 6

bioinformatics × 6
time-series × 6
unlabeled-data × 6
proportion × 6
language × 6
recall × 6
semi-supervised × 6
hci × 6
accuracy × 6
multimedia × 6

transductive-learning × 6
interface × 6
precision × 6
speech-recognition × 6
text-classification × 5
nlp × 5
open-source × 5
overfitting × 5
poll × 5
documentation × 5

18 Kudos

● Prominent Question × 20
● Conspicuous Question × 20
● Well-Known Question × 20
● Appreciated Question × 10
● Appreciated Answer × 8
● Prized Question × 3

● Revisitation × 2
● Revisionist × 1
● Esteemed Answer × 1
● Pronouncement × 1
● Blue Pencil × 1
● Guide × 1

● Commentator × 1
● Esteemed Question × 1
● Participant × 1
● Advocate × 1
● Discriminating × 1
● Inquisitive × 1

Registered user

20 Questions

22 Answers

16 Votes

50 Tags

18 Kudos