Modern information retrieval systems are extremely different than their predecessors were. Natural language searching employs behind-the-scenes algorithms that parse users’ phrases and sentences and create input strings that database search engines can process, meaning that the search that a user inputs may bear little resemblance to the search that the database actually runs. In order to avoid confusing a novice searcher, some vendors choose to position the translated search on an “advanced search” or “details” page; other vendors consider their algorithms to be their competitive advantage and “black-box” the search details completely. While such practices do not often harm the casual searcher, they can present a frustrating or even impossible situation for a professional searcher or for someone who wishes to ensure that their search results are exhaustive. This talk will examine the outcomes of queries run using popular chemistry databases, attempting to peer into the “black boxes” beneath their attractive interfaces, and will conclude with suggestions that will help both occasional searchers and information professionals maximize retrieval of relevant results.

Artificial intelligence and chemical information retrieval: Who’s afraid of the big black box?

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.