Combining fuzzy information from multiple systems (extended abstract)
PODS '96 Proceedings of the fifteenth ACM SIGACT-SIGMOD-SIGART symposium on Principles of database systems
A framework for expressing and combining preferences
SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Neural Networks: A Comprehensive Foundation
Neural Networks: A Comprehensive Foundation
Top-k selection queries over relational databases: Mapping strategies and performance evaluation
ACM Transactions on Database Systems (TODS)
RankSQL: query algebra and optimization for relational top-k queries
Proceedings of the 2005 ACM SIGMOD international conference on Management of data
Best-k queries on database systems
CIKM '06 Proceedings of the 15th ACM international conference on Information and knowledge management
Hi-index | 0.00 |
Currently, websites on the Internet serving structured data allow users to perform search based on simple equality or range constraints on data attributes. However, to begin with, users may not know what is desirable to them precisely, to be able to express it accurately in terms of primitive equality or range constraints. Additionally, in most websites, the results provided to users can be sorted with respect to values of any one particular attribute at a time. For the user, this is like searching for a needle in a haystack because the user's notion of interesting objects is generally a function of multiple attributes. In this paper, we develop an approach to (i) support a family of functions involving multiple attributes to rank the tuples, and (ii) improve the ranking of results returned to the user by incorporating user feedback (to learn user's notion of interestingness) with the help of a neural network. The user feedback driven approach is effective in modeling a user's intuitive sense of desirability of a tuple, a notion that is otherwise near impossible to quantify mathematically. To prove the effectiveness of our approach, we have built a middleware for an application domain that implements and evaluates these ideas.