Boolean Operators
Use Boolean operators ( AND , OR , NOT ) to combine multiple search terms in one query.
| Operator | Alias | Result |
| AND | && |
Using the AND operator to find documents that contain both of the words or phrases linked by the AND . Example: TI:(antenna AND radio) will find all documents whose title contains "antenna" and "radio" You can also use the alias TI:(antenna && radio) . |
| OR | || |
Use the OR connector to find documents that contain either or both of the words or phrases linked by OR . The OR connector is useful to link search words that are synonyms, antonyms, alternative spellings, or abbreviations. Example: TI:(antenna OR radio) will find all documents whose title contains "antenna" or "radio". You can also use the alias TI:(antenna || radio) . |
| NOT | ! |
Use the NOT connector to find documents in which a search word or phrase is to be excluded. Example: TI:(antenna NOT radio) will find all documents whose title contains "antenna" minus the documents whose title contains "radio". You can also use the alias TI:(antenna ! radio) . |
You can use multiple Boolean operators in one query, while possible it is easy to loose track of the
operator range and precedence. The operator range is determined by the use of parentheses (). For
the operator preference TotalPatent One® uses the following logic, OR has precedence, then NOT,
then AND. For example the query TI:(chocolate OR strawberry AND cake) :
| Query | Number of Results |
| TI:(chocolate OR strawberry AND cake) | 846 |
| TI:(chocolate AND cake) | 686 |
| TI:(strawberry AND cake) | 163 |
| TI:(chocolate AND strawberry AND cake) | 3 |
As can be observed from these results the OR operator has precedence. The query that is executed gives you all documents with "chocolate" OR "strawberry" AND "cake" in the title. There are 3 documents with "chocolate" AND "strawberry" AND "cake".
Tip: Instead of nesting complicated queries using Boolean operators (AND, OR, NOT) you can also create two individual queries and use the Compare queries functionality instead (see Compare Queries).
Caution: You can add multiple keywords to a single search, but spaces between keywords are interpreted as an AND operator unless you use double quotes (") surrounding the keywords.
Caution: Be advised to use wild-card, fuzzy or Boolean operators with restriction, too many in a single query will cause serious delays in the response time. Best practice is to split overcomplicated queries and combine the results via the history page.
Proximity Operators
Proximity operators ( NEARn , NEARs , PREn ) - used to combine multiple search terms in one query.
| Operator | Result |
| NEARn |
Use the NEARn connector to find documents with search words that appear within n words of each other. The value of n can be any number from 0 up to 255. Use NEARn to join words and phrases that express parts of a single idea or to join closely associated ideas. Proximity requests can only be done within the same object, make sure you use the parentheses () correctly. |
| NEARs |
Use the NEARs connector to find documents with search words that appear within the same sentence. You may also use NEARs when you wish a close relationship between words without specifying an exact proximity. Either word may appear first. Example: TAC: (antenna NEARs radio) will find all documents whose title, abstract or claims contain the search words "radio" AND "antenna" in the same sentence. Proximity requests can only be done within the same object, make sure you use the parentheses () correctly. Note: As you can see in the example above, keywords are individually highlighted, in this case "Radio" and "Antenna". Highlighting does not take proximity operators (NEARn, NEARs, PREn) into account. |
| PREn |
Use the PREn connector to find documents in which the first search word precedes the second by not more than the stated number of words. As with NEARn , both words must be in the same field and appear within n words of each other. The value of n can be any number from 0 up to 255. Example: TI:(radio PRE3 antenna) will find all documents whose title contains "radio" AND "antenna" where radio is the first word and they are maximally Proximity requests can only be done within the same object, make sure you use the parentheses () correctly |
Boost Operator
To boost a search term, use the caret (^) operator to give more value to a certain term compared to
another.
Boosting allows you to control the relevance of a document. In some queries one keyword is more important than the other. In that case you would like the results that contain the more relevant keyword at the top of the result list. This can be achieved by boosting. The syntax for keyword boosting is ^n where n represents a number indicating the relevance of one keyword compared to another. For example, cake^1 AND chocolate^5 searches documents where the terms "cake" AND "chocolate" occur in the full text, the results are sorted based on the most occurrences of "chocolate".
Without the boost operator, cake AND chocolate you will retrieve the same number of results. The relevance is now determined equally on both keywords leading to a different sorting order.
Please keep in mind stemming is used by TotalPatent One independent of phrase matching, this means in this example the plural forms "cakes" and "chocolates" are also part of the result.
Boosting is supported for phrases as well.
Note:
Document relevance is calculated based on:
- keyword occurrence per document - thirteen occurrences are more relevant than one.
- occurrence in the whole document set - a keyword occurring in 20.000 documents is less
relevant than a keyword occurring in only four documents. - document length - one occurrence in a short document can be equally relevant as five
occurrences in a long document. - section of the document - a keyword in a title will be more relevant than a keyword in a
description.
The result list is sorted according a weighted total score of all of these. Therefore it is not possible to fully predict the relevance sorting order beforehand. Boosting a keyword helps, but expanding and/or filtering your query helps even more.