The Second International Workshop on Diversity in Document Retrieval (DDR-2012)
Dates:
When Feb 12, 2012 – Feb 12, 2012
Where Seattle WA, USA
Submission Deadline Dec 5, 2011
Notification Due Jan 10, 2012
Final Version Due Jan 17, 2012
From the webpage:
In conjunction with WSDM 2012 – the 5th ACM International Conference on Web Search and Data Mining
Overview
=======
When an ambiguous query is received, a sensible approach is for the information retrieval (IR) system to diversify the results retrieved for this query, in the hope that at least one of the interpretations of the query intent will satisfy the user. Diversity is an increasingly important topic, of interest to both academic researchers (such as participants in the TREC Web and Blog track diversity tasks), as well as to search engines professionals. In this workshop, we solicit submissions both on approaches and models for diversity, the evaluation of diverse search results, and on applications and presentation of diverse search results.
Topics:
- Modelling Diversity:
- Implicit diversification approaches
- Explicit diversification approaches
- Query log mining for diversity
- Learning-to-rank for diversification
- Clustering of results for diversification
- Query intent understanding
- Query type classification
- Modelling Risk:
- Probability ranking principle
- Risk Minimization frameworks and role diversity
- Evaluation:
- Test collections for diversity
- Evaluating of diverse search results
- Measuring the ambiguity of queries
- Measuring query aspects importance
- Applications:
- Product & review diversification
- Opinion and sentiment diversification
- Diversifying Web crawling policy
- Graph analysis for diversity
- Summarisation
- Legal precedents & patents
- Diverse recommender systems
- Diversifying in real-time & news search
- Diversification in other verticals (image/video search etc.)
- Presentation of diverse search results
While typing this up, I remembered the “little search engine that could” post (Going Head to Head with Google (and winning)). Are we really condemned to have to manage unforeseeable complexity or is that a poor design choice we made for search engines?
After all, I am not really interested in the entire WWW. At least for this blog I am interested in probably less than 1/10 of 1% of the web (or less). So if I had a search engine for all the CS/Library/Informatics publications, blogs, subject domains relevant to data/information, I would pretty much be set. A big semantic field and one that is changing, but not anything like search everything that is connected (or not, for the DeepWeb) to the WWW.
I don’t have an answer for that but I think it is an issue that may enable management of semantic diversity. That is we get to declare the edge of the map. Yes, there are other things beyond the edge but we aren’t going to include them in this particular map.