Daniel E Margolis - user overview - MetaOptimize Q+A

Daniel E Margolis

106

karma

Registered user
member since	Dec 26 '12 at 17:55
last seen	Nov 06 '14 at 22:20

overview recent activity karma history favorites

5 Questions

0

votes

0

answers

359

views

How to formulate probability for a new worker on a new sample in a crowdsourcing platform?

May 08 '13 at 06:26 Daniel E Margolis106●5●5●10

bayesian crowdsourcing instance-based

0

votes

1

answers

867

views

Measuring similarity/distance between biographical essays or scientific papers?

Mar 29 '13 at 21:11 Daniel E Margolis106●5●5●10

clustering text-mining distance data-mining similarity

0

votes

0

answers

449

views

How should I go about creating a synthesized (or semi-synthesized) dataset that fits my criteria?

Jan 11 '13 at 21:38 Daniel E Margolis106●5●5●10

0

votes

0

answers

601

views

What are possible experimental setups involving label cost minimization (reducing the number of labels queried)?

Dec 28 '12 at 14:38 Daniel E Margolis106●5●5●10

query cost-sensitive labeling active-learning

2

votes

1

answers

760

views

Active learning to reduce the cost and number of stored instances for "instance-based" (memory-based) methods?

Dec 28 '12 at 16:25 Daniel E Margolis106●5●5●10

memory-based active-learning history-based-models cost-sensitive instance-based

14 Answers

Neural network learning fast, getting false positives

Large Number of Categories with few shared features (Best approach?)

Leave one out SMO sigmoid kernel

New trends on algorithms for classification task?

Cross validation on kNN

Relation between k and error in kNN

Which approach and model to use to find most effective keywords?

Activation function that depends on whole layer?

Difference between RMSE and Spearman Correlation

"ValueError: setting an array element with a sequence" during scikit-learn classification

Neural network learning fast, getting false positives

Examples of papers that list and compare a large number of similar techniques?

How to detect poorly written text

What methods scale to really big data?

8 Votes

3

5

46 Tags

machine-learning × 7
classification × 5
text-mining × 3
neural-networks × 3
clustering × 2
numpy × 2
instance-based × 2
rmse × 2
distance × 2
regression × 2

scikit-learn × 2
cost-sensitive × 2
data-mining × 2
backpropagation × 2
correlation × 2
performance × 2
computer-vision × 2
similarity × 2
knn × 2
active-learning × 2

shuffling × 1
practical-advice × 1
model-selection × 1
grammar-checking × 1
papers × 1
algorithms × 1
activation × 1
rapidminer × 1
memory-based × 1
spelling-correction × 1

optimization × 1
text-classification × 1
query × 1
survey × 1
crowdsourcing × 1
categorization × 1
comparison × 1
big-data × 1
crossvalidation × 1
bayesian × 1

keywords × 1
dataset × 1
references × 1
rbm × 1
labeling × 1
history-based-models × 1

8 Kudos

● Conspicuous Question × 5
● Prominent Question × 5
● Well-Known Question × 5
● Pronouncement × 1
● Advocate × 1
● Appreciated Question × 1

● Revisionist × 1
● Guide × 1

learn more | privacy | contact

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.