Opinion Mining
This month we celebrate the first anniversary of our newsletter and proudly present Eastern Fence Lizard (2014-06 release). Looking back at the inaugural issue one year ago, we shared a testimonial from David Herring, Communications Director of the NOAA Climate Program Office. Since then, the opinion mining technology of webLyzard has continued to play an essential part in NOAA's online evaluation strategy. A new research article in the distinguished journal Knowledge-Based Systems presents the NOAA portal to highlight how opinion mining helps to optimize communication campaigns.
As an example of opinion mining across languages, we have launched an extended "Media Watch on Climate Change" that includes German and French news and social media content during a June 10 press conference (see sidebar).
|
|
|
Demo Portal
Multilingual | Media Watch on Climate Change. User-generated content from social media platforms is noisy, unstructured, and multilingual. Processing this type of content requires robust text mining techniques. Multilingual capabilities are becoming increasingly important when applying these techniques to global networks. This goes beyond merely detecting the language(s) used, a comparably straightforward task. It requires specific language resources (sentiment lexicons), and a system that is capable to process grammatical structures and consider the idiosyncrasies of specific language communities.
[english] [french] [german] |
|
Journal Article
Context-Aware | Novel Opinion Mining Approach. Processing multilingual knowledge resources, automatically considering the specific context of a stated opinion, and the ability to resolve the inherent ambiguities of human languages remain key challenges of opinion mining. This article summarizes the recent progress of our research team in addressing these challenges. It shows how to enrich semantic knowledge bases with external concept knowledge - with a focus on the application of crowdsourcing and opinion mining techniques to Web intelligence platforms and other high-throughput big data applications.
[download article] |
|
|
