(203) 809-9490 - Send a Message - Schedule an Inquiry Phone Call - on the Homepage
Legal Research Strategy: Using a NOT Boolean Operator
Why/When to Use
One use of a NOT boolean operator is to exclude results which are not relevant to your search when part way through reviewing the results, a certain type of repeated non-relevant result is observed. Then, a search is done with the same search terms as before plus the use of the NOT operator to remove the repeated non-relevant result observed. After the non-relevant results are excluded, completing the review of the remainder of the results will take less time. This can be a time saving technique when you’re trying to review many search results such as for a patent search or during e-discovery.
Example of Using a NOT Boolean Operator for E-discovery
Example of How to Use
For example, in the context of an e-discovery project, you are asked to do a search to determine the best instance of when two parties discuss suspected prior art relating to an invention used for speeding up curing timing of concrete with compound y.
You have 50,000 emails to search through. You’ve made a list of suspected keywords which will give you good results. After doing some test searches, you find that if you search “concrete” AND “smith” AND “cure” you are finding some good results because you find that the emails which include those key words discuss prior art. The emails have John Smith talking about the cure time of the concrete.
After a search with the three good keywords stated earlier you have 2,000 emails to go through, and when going through the first 50 of those emails you notice that one out of every two emails in that 2,000 set deals with a topic that is not relevant to your search. The non-relevant emails all tend to pertain to a different invention with compound x but do not mention compound y.
As such, you have found a sub-group of non-relevant results within your relevant results which you can exclude with a NOT operator, which will reduce your search time because you will not need to review the non-relevant subgroup. You will then use a NOT operator to exclude the results that mention compound x by searching “concrete” AND “smith” AND “cure” NOT “compound x.” This will exclude the non-relevant results and reduce the number of results you need to search through – hopefully down to 1,000 emails instead of 2,000 emails if each email about compound x also includes the words “compound x.”
Generic Representation of Using a NOT Boolean Operator
Restated Generically
The NOT operator will be used after you determine search terms which are giving you relevant search results, but the relevant results includes a subgroup of non-relevant results which all have a common keyword which is not in the relevant results. The NOT operator will be used to exclude the subgroup of non-relevant results from the relevant results. The time taken to redo the previous search – excluding non-relevant results – will be time efficient because the non-relevant are so numerous the exclusion of the non-relevant results will reduce the overall review time.
In the diagram above, the main group of relevant results was found with the search X AND Y. The subgroup of the non-relevant results, which was found by a partial review, is removed from the main group by adding a NOT operator to exclude the results with a keyword Q.
Possible Downside
You might exclude relevant search results that include the keyword causing exclusion – the NOT keyword. Therefore, try to find a keyword which is only in the non-relevant results and not in the relevant results.
Additionally, the above technique is best utilized if there is not the expectation that every document must be reviewed (such as for a patent search, where there are an extremely large amount of patents and all cannot be reviewed). If every document must be reviewed then you might be stuck looking at all of them.