Multiple-Goal Search Algorithms and Their Application to Web Crawling

Abstract

The work described in this paper presents a new framework for heuristic search where the task is to collect as many goals as possible within the allocated resources. We show the inadequacy of traditional distance heuristics for this type of tasks and present alternative types of heuristics that are more appropriate for multiple-goal search. In particular we introduce the yield heuristic that estimates the cost and the benefit of exploring a subtree below a search node. We present a learning algorithm for inferring the yield based on search experience. We apply our adaptive and non-adaptive multiple-goal search algorithms to the web crawling problem and show their efficiency.

Cite

Text

Davidov and Markovitch. "Multiple-Goal Search Algorithms and Their Application to Web Crawling." AAAI Conference on Artificial Intelligence, 2002. doi:10.5555/777092.777202

Markdown

[Davidov and Markovitch. "Multiple-Goal Search Algorithms and Their Application to Web Crawling." AAAI Conference on Artificial Intelligence, 2002.](https://mlanthology.org/aaai/2002/davidov2002aaai-multiple/) doi:10.5555/777092.777202

BibTeX

@inproceedings{davidov2002aaai-multiple,
  title     = {{Multiple-Goal Search Algorithms and Their Application to Web Crawling}},
  author    = {Davidov, Dmitry and Markovitch, Shaul},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2002},
  pages     = {713-718},
  doi       = {10.5555/777092.777202},
  url       = {https://mlanthology.org/aaai/2002/davidov2002aaai-multiple/}
}