Policy Aware Content Reuse

(CC Attribution License Violations Detection + Semantic Clipboard)

Oshani Seneviratne

Initiated the project during the "Networks for Web Science Student Exchange" at Soton
Became my SM thesis work
Consists of 3 sub projects:
- Assessment of Copyright Violations on the Web
- Attribution License Violations Validator for Flickr Images
- Semantic Clipboard on Tabulator

Goal: Find how many CC licensed images are embedded in Web sites without proper attribution.
Criteria: Attribution to original source or the owner specified within reasonable proximity to the embedded image.
Samples: Generated using the Technorati blog crawler. Fair Sample ensured by randomizing seed URIs, and by using the Technorati "Authority Rank".

Sample	Websites	Total Number of Images	Correctly Identified Images	Misattribution	Precision
1	67	426	183	78%	55%
2	70	241	113	80%	42%
3	70	466	268	94%	39%

Site Crawler search for all the links embedded in the given site.
Flickr Query Evaluator obtains license specific information using the photo id.
License Checker checks the license and proper attribution details.
Notification System gives the instances of license violations in a human-friendly manner.

RDFa Extractor extracts all the RDFa embedded in an HTML page on Page Load.
UI Enhancer overlays the user interface for better license awareness.
Attribution XHTML Constructor composes attribution XHTML code snippet.
User Interface consists of a context menu on images that can be copied to the System Clipboard.

Right Click on the Image with RDFa

Triples corresponding to the subject of the image are extracted from the page.

Attribution XHTML is constructed and placed on the System Clipboard.

CC License Use and Restriction categories mapped in the menu selection.

Tool tip text displays whether the image can be copied or not.

Attribution License Violations Validator:
- Tracking provenance
- Tracking subsequent changes in the license
- Scope of the human readable attribution notice
- Limited Flickr support for license expression
Semantic Clipboard:
- License granularity in the HTML DOM
- Browser dependence
- Usability vs. Operating System independence

oshani@csail.mit.edu