Previous research has motivated the idea of automatically determining when programmers are having difficulty, provided an initial algorithm (unimplemented in an actual system), an...
In task 1A of the BioCreAtIvE evaluation, systems had to be devised that recognize words and phrases forming gene or protein names in natural language sentences. We approach this ...
We present a method to learn visual attributes (eg.“red”,
“metal”, “spotted”) and object classes (eg. “car”,
“dress”, “umbrella”) together. We assume imag...
This paper studies on the mirror image learning algorithm for the autoassociative neural networks and evaluates the performance by handwritten numeral recognition test. Each of th...
The paper evaluates the eectiveness of learning for speeding up the solution of constraint satisfaction problems. It extends previous work (Dechter 1990) by introducing a new and ...