The Crystallographer's Assistant
Abstract
instances. The Crystallographer ’s Assistant is based upon a case-based reasoning approach, and involves, as a first step, creating a database (from both existing experiment notebooks and on-going experiments), of about 1000 examples of crystallography experiments. These examples will provide us with both successful as well as failed experiments, and will be used both by RL as well as the case-based reasoner. Given the significant complexity and weak theory of the relationships between the features of the experiments, a case-based approach is being taken for similarity assessment. Experiential data concerning how the domain experts define pairs of cases to be similar and different will be used to guide the indexing and selection of cases. The result will be an experimenter’s assistant which, given the results of the latest set of experiments, will remind the user of previous experiments with similar conditions and make suggestions based upon what was done in both cases
Cite
Text
Gopalakrishnan et al. "The Crystallographer's Assistant." AAAI Conference on Artificial Intelligence, 1994.Markdown
[Gopalakrishnan et al. "The Crystallographer's Assistant." AAAI Conference on Artificial Intelligence, 1994.](https://mlanthology.org/aaai/1994/gopalakrishnan1994aaai-crystallographer/)BibTeX
@inproceedings{gopalakrishnan1994aaai-crystallographer,
title = {{The Crystallographer's Assistant}},
author = {Gopalakrishnan, Vanathi and Hennessy, Daniel N. and Buchanan, Bruce G. and Subramanian, Devika},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {1994},
pages = {1451},
url = {https://mlanthology.org/aaai/1994/gopalakrishnan1994aaai-crystallographer/}
}