Views
Graph
Explorer
Focus
Down
Load 1 level
Load 2 levels
Load 3 levels
Load 4 levels
Load all levels
All
Dagre
Focus
Down
Load 1 level
Load 2 levels
Load 3 levels
Load 4 level
Load all levels
All
Tree
SpaceTree
Focus
Expanding
Load 1 level
Load 2 levels
Load 3 levels
Down
All
Down
Radial
Focus
Expanding
Load 1 level
Load 2 levels
Load 3 levels
Down
All
Down
Box
Focus
Expanding
Down
Up
All
Down
Page ✓
Article
Outline
Document
Down
All
Canvas
Time
Timeline
Calendar
Request email digest
Past 24 hours
Past 2 days
Past 3 days
Past week
Add
Add page
Add comment
Add citation
Edit
Edit page
Delete page
Share
Link
Bookmark
Embed
Social media
Login
Member login
Register now for a free account
🔎
Issues of Verification for Unsupervised Discovery Systems
Position
1
#249067
CONTEXT
(Help)
-
OpenSherlock Project »
OpenSherlock Project
OpenSherlock Project☜Fabricating possibly many open source variants along the lines of IBMs Watson. The projects name has changed from SolrDrWatson to SolrSherlock, with thanks to Tom Munnecke. Migrating to OpenSherlock concept where we generalize beyond Solr as the platform core.☜F1CEB7
▲
References »
References
References☜Links to resources related to the SolrDrWatson project☜59C6EF
▲
Conference Papers »
Conference Papers
Conference Papers☜☜FFB597
■
Issues of Verification for Unsupervised Discovery Systems
Issues of Verification for Unsupervised Discovery Systems☜☜59C6EF
►
Use Case 4: Knowledge-based Discovery »
Use Case 4: Knowledge-based Discovery
Use Case 4: Knowledge-based Discovery☜A top-level use case for many activities☜FFFACD
Heading
Summary
Click the button to enter task scheduling information
Open
Details
Enter task details
Message text
Select assignee(s)
Due date (click calendar)
RadDatePicker
RadDatePicker
Open the calendar popup.
Calendar
Title and navigation
Title and navigation
<<
<
October 2024
>
<<
October 2024
S
M
T
W
T
F
S
40
29
30
1
2
3
4
5
41
6
7
8
9
10
11
12
42
13
14
15
16
17
18
19
43
20
21
22
23
24
25
26
44
27
28
29
30
31
1
2
45
3
4
5
6
7
8
9
Reminder
No reminder
1 day before due
2 days before due
3 days before due
1 week before due
Ready to post
Copy to text
Enter
Cancel
Task assignment(s) have been emailed and cannot now be altered
Lock
Cancel
Save
Comment graphing options
Choose comments:
Comment only
Whole thread
All comments
Choose location:
To a new map
To this map
New map options
Select map ontology
Options
Standard (default) ontology
College debate ontology
Hypothesis ontology
Influence diagram ontology
Story ontology
Graph to private map
Cancel
Proceed
+Comments (
0
)
- Comments
Add a comment
Newest first
Oldest first
Show threads
+Citations (
1
)
- Citations
Add new citation
List by:
Citerank
Map
Link
[1]
Issues of Verification for Unsupervised Discovery Systems
Author:
Shou-de Lin, Hans Chalupsky
Publication info:
2004 KDD-2004
Cited by:
Jack Park
7:09 PM 3 February 2013 GMT
URL:
http://www.cs.cmu.edu/~dunja/LinkKDD2004/Shoude-Lin-LinkKDD-2004.pdf
Excerpt / Summary
Developing systems that perform discovery presents unique challenges, e.g., when compared to learning programs, since in general there is no teacher or example library available to train or evaluate a discovery system. In particular, the lack of gold-standard methods or examples makes the verification of discovered results and evaluation of system performance a very difficult problem. In this paper we address several issues of verifying a machine discovery system and discuss what one should and should not expect from the evaluation report of a discovery system. Recognizing that there is no direct way to verify the validity of a true discovery system, this paper proposes several indirect strategies one can adopt to assess the validity of discovered results. We then present our own set of novel link discovery tools as a case study to show how the proposed concepts can be applied to verify a real-world discovery system. In a time where computer science research has become extremely evaluation driven, researchers sometimes shy away from areas where results are difficult to evaluate. By discussing these issues with respect to discovery systems, we hope to provide a useful overview as well as – hopefully - a positive and encouraging signal to attract more researchers to work on machine discovery problems.
+About
- About
Entered by:-
Jack Park
NodeID:
#249067
Node type:
Position
Entry date (GMT):
2/3/2013 7:06:00 PM
Last edit date (GMT):
2/3/2013 7:06:00 PM
Show other editors
Incoming cross-relations:
0
Outgoing cross-relations:
1
Average rating:
0
by
0
users
x
Select file to upload