An object-oriented implementation of an adaptive classification of job openings
Stephen W. Clyde, Jianping Zhang, Chih-Chung Yao · 2002
Automating job classification is challenging because it involves a large number of dynamic classes and features, concept drift uncertainty, and noisy data. We present a software solution to this problem that consists of an incremental learning subsystem and a job classifier. We also describe our design and implementation using object-oriented systems modeling, a complete object-oriented approach that supports analysis, specification and design, and has a smooth mapping to most oriented-object programming languages. Some experimental results and comparisons to other learning/classification algorithms are given. A production version of the software written in C/sup ++/ is performing with superior accuracy.>