0.25
0.5
0.75
1.25
1.5
1.75
2
Cross-lingual Global Media Monitoring
Published on Mar 06, 20153741 Views
Global media monitoring in real-time assumes handling large amount of textual data across different languages. We propose using text mining methods together with semantic processing to identify events
Related categories
Chapter list
Real-Time Cross-Lingual Global Media Monitor00:00
Outline00:12
Introduction00:46
The overall goal00:48
Real-Time Cross-lingual News collection01:50
Systems/Demos used within the presentation02:18
Global Media Monitoring pipeline - 102:59
Global Media Monitoring pipeline - 203:07
Global Media Monitoring pipeline - 303:18
Global Media Monitoring pipeline - 403:25
Global Media Monitoring pipeline - 503:29
Collecting Media Data03:35
Get references to news publishers03:37
From a newspaper home-page to an article04:38
Collecting global media data04:51
Downloading the news stream - 106:03
Downloading the news stream - 206:16
Document Enrichment06:41
How can we annotate a document?06:46
Enrycher (http://enrycher.ijs.si/)06:55
Enrycher Architecture09:04
Anaphora resolution - 109:26
Anaphora resolution - 209:56
Anaphora resolution evaluation - 110:20
Anaphora resolution evaluation - 210:39
Cross-linguality11:10
How to operate in many languages?11:36
Languages covered by XLing (top 100 Wikipedia languages) - 112:05
Languages covered by XLing (top 100 Wikipedia languages) - 212:50
Cross-lingual article matching12:51
XLing (XLing.ijs.si)13:08
Example: Cross-lingual News Recommendation15:00
Example: Social Media Recommendation16:32
Example: Article clustering18:00
Cross-lingual cluster linking19:09
Event Representation20:57
What is an event? (abstract description)21:08
How to represent an event?21:53
Feature vector event representation22:59
Example of “feature vector” event representation: Event Registry “Chicago” related events23:46
Structured event representation24:09
“Event Taxonomy”: preview to the current development - 125:04
“Event Taxonomy”: preview to the current development - 225:14
Prototype for event Infobox extraction: semi-automatic annotation service25:52
Event sequences & Hierarchical events26:34
An example event: Microsoft Windows 927:18
Similar events example: similar events to Microsoft Windows 9 event27:37
Event sequence identification27:50
Example Microsoft hierarchy of events27:57
Zoom-in Example Microsoft hierarchy of events28:05
Event Visualization28:14
Event description through entities and Semantic keywords28:19
Collection of events described through Entity relatedness30:55
Collection of events described through Entity relatedness33:20
Collection of events described through trending concepts35:38
Collection of events described through three level categorization38:10
Events identified across languages40:23
Collection of events described through Reporting dynamics41:34
Collection of events described through a story-line of related events46:41
Event Registry API47:24
Python code to access “Event Registry”47:27
Searching for events using Python47:37
Result of the query48:14
Getting info for a particular event48:20
Searching for articles48:31
Future: Challenges and opportunities48:41
Summary48:42
Scientific Challenges49:31
Business/Innovation opportunities51:00