Retrieval of very large numbers of items in the Web of Science: An exercise to develop accurate search strategies

Abstract

The Web of Science interface counts at most 100,000 retrieved items from a single query. If the query results in a dataset containing more than 100,000 items the number of retrieved items is indicated as >100,000. The problem studied here is how to find the exact number of items in a query that leads to more than 100,000 items. One way to achieve this objective is presented. The retrieval of the entire scientific production from the United States in a specific year (2007) is counted and an advanced search strategy is designed. Different sections of items can be retrieved using the Source field of the database. A Boolean statement was created with the aim of eliminating overlapped sections and improving the accuracy of this search strategy.