Faster conceptual blending predictors on relational time series
Tan, Terence K.
Darken, Christian J.
MetadataShow full item record
Tasks at upper levels of sensor fusion are usually concerned with situation or impact assessment, which might consist of predictions of future events. Very often, the identity and relations of target of interest have already been established, and can be represented as relational data. Hence we can expect a stream of relational data arriving at our agent input as the situation updates. The prediction task can then be expressed as a function of this stream of relational data. Run-time learning to predict a stresm of percepts in an unknown and possibly complex environment is a hard problem and especially so when a serious atttempt needs to be made even on the first few percepts. When the percepts are relational (logical atoms), the most common practica technologies require engineering by a human expert and so are not applicable. We briefly describe and compare several approaches which do not have this requirement on the initial hundred precepts of a benchmark domain. The most promising approach extends existing approaches by a partial matching algorithm inspired by theory of conceptual blending. This technique enables predictions in novel situations where the original approach fails, and significantly improves prediction performance overall. However an implementation, based on backtracking, may be too slow for many implementations. We provide an accelerated approximate algorithm based on best-first and A* search, which is much faster than the initial implementation.
Showing items related by title, author, creator and subject.
Anastasopoulos, Achilles (Monterey, California. Naval Postgraduate School, 1997-03);Relational, hierarchical, network, functional, and object oriented databases support its corresponding query language, SQL, DL/I, CODASYL-DML, DAPLEX, and OO-DML, respectively. However, each database type may be accessed ...
Reactivation of the relational interface in M_x001B_p2_x001B_sDBMS and implementation of the EWIR database Scrivener, Donna N.; Edwards, Renell D. (Monterey, California. Naval Postgraduate School, 1996-06);The primary Department of Defense source for technical parametric performance data on non-communications emitters is the Electronic Warfare Reprogramming Database (EWIRDB). Data representation in the EWIRDB is via disjointed ...
Grenseman, Paul D. (1988);The relational database model has become the most popular and widespread database model. Most current database systems are based upon or related to -he relational model. However, the relational model is beset with ...