Beam Search in Incremental Rule Learning
نویسنده
چکیده
This paper describes ICN, an incremental version of the CN2 rule learning system. Unlike other incremental rule learning systems which learn rules gradually, adding and removing conditions in a hill-climbing search, ICN learns or unlearns each rule “all at once,” using beam search as in CN2. In batch training and testing with the forest cover prediction problem, ICN performs nearly as well as CN2. ICN’s efficient incremental algorithm, however, allows it to learn from much more data. When trained and tested incrementally on the entire forest cover data set (581,012 instances), ICN’s performance exceeds that of the best known classifier for this problem.
منابع مشابه
HC-Search for Incremental Parsing
Standard incremental parsing algorithm employs a single scoring function and beam-search to find the best parse tree from an exponentially large search space. Inspired by recently proposed HC-search framework, we decompose the incremental parsing algorithm into two steps: first searching a set of high-quality outputs with beam-search, and second selecting the best output with a ranking model. W...
متن کاملAn Empirical Comparison of Hill-Climbing and Exhaustive Search in Inductive Rule Learning An Empirical Comparison of Hill-Climbing and Exhaustive Search in Inductive Rule Learning
Most commonly used inductive rule learning algorithms employ a hill-climbing search, whereas local pattern discovery algorithms employ exhaustive search. In this paper, we evaluate the spectrum of different search strategies to see whether separate-and-conquer rule learning algorithms are able to gain performance in terms of predictive accuracy or theory size by using more powerful search strat...
متن کاملSyntactic Processing Using Global Discriminative Learning and Beam-Search Decoding
This tutorial discusses a framework for incremental left-to-right structured predication, which makes use of global discriminative learning and beam-search decoding. The method has been applied to a wide range of NLP tasks in recent years, and achieved competitive accuracies and efficiencies. We give an introduction to the algorithms and efficient implementations, and discuss their applications...
متن کاملA Re-evaluation of the Over-Searching Phenomenon in Inductive Rule Learning
Most commonly used inductive rule learning algorithms employ a hill-climbing search, whereas local pattern discovery algorithms employ exhaustive search. In this paper, we evaluate the spectrum of different search strategies to see whether separate-and-conquer rule learning algorithms are able to gain performance in terms of predictive accuracy or theory size by using more powerful search strat...
متن کاملIncremental Dependency Parsing Using Online Learning
We describe an incremental parser that was trained to minimize cost over sentences rather than over individual parsing actions. This is an attempt to use the advantages of the two top-scoring systems in the CoNLL-X shared task. In the evaluation, we present the performance of the parser in the Multilingual task, as well as an evaluation of the contribution of bidirectional parsing and beam sear...
متن کامل