'echo chamber' version
We noticed that using strictly the terms that policy makers of the UN use in the targets and indicators, that much of the research that did not use that specific terms was left out in the result set. (eg. "mortality" vs "deaths") To increase the recall, without reducing precision of the papers in the results, we added keywords that were obvious synonyms and antonyms to the existing 'strict' keywords. This was done based on the keywords that appeared in papers in the result set of version 2. This creates an 'echo chamber', that results in more of the same papers.