The Lens gives you a number of ways to hone down a query to help you find exactly what you want. The easiest way is to use the normal search function and to refine your search using the filters available. Technically savvy users are also able to construct complex queries directly using the search syntax. The Structured Search page is another way to refine your search from the very start and gives you unique options for searching for text within specific fields of each document.
Using the Structured Search tools you can create specific search queries, filtering by dates, jurisdictions and document type. You can also alter default search parameters such as stemming, family grouping, query language and the whether to only include documents with full text.
Structured Search also allows you to limit your search terms to specific sections of documents. By searching for terms only within a given field, you can greatly reduce the number of extraneous results. For example, searching for an inventor’s name only within the Inventor field will eliminate documents that happen to mention the inventor in passing.
To begin a search on a topic, it is therefore often best to begin by searching within the Title, Abstract, or Claims to capture the broadest range of topical documents. Searching within Applicants may be useful for determining which entities hold which patent documents, but keep in mind that patents may change ownership and this change may not be recorded, and that patents held by a subsidiary will not appear in a search for patents held by the parent company.
The Structured Search page can be found here: https://lens.org/lens/structured-search.
You can get to the Structured Search page at any time by clicking the cog button next to the search button on the home page or next to the new search button on the search results page.
In the query field you can choose which section of each document you would like your search terms applied to. You can choose between the sections: Inventors, Owners (US), Title, Abstract, Claims, Applicants, Authors, Citation ID, Non Patent Citations, ORCID works citations, Lens ID, Publication Number, Filing Number as well as US and IPCR Classifications. You can add any number of these fields and combine them all using a boolean AND or OR predicate.
Selecting AND will mean a document must have at least one match for each term in the section it is being searched on to be added to the search results. Selecting OR will mean a document must have at least one of the terms match in the section it is being searched on to be added to the search results.
ORCID ID Citations
Once a user enters the name or ORCID ID of a researcher in the structured search, the lens extracts the article DOIs or PMIDs listed in the researcher’s ORCID profile (if that researcher has voluntary made his profile public) and check if there are matches to these article IDs in the Lens citation database. If a match is found, the Lens displays the patents that cite that scholarly work in the Search result page.
Ex. I searched for Laurel Haak
And here is the results I got. These five patents cite the works listed by Laurel Haak in her ORCID public profile. However, to evaluate which article is cited in which patent, it may be difficult using this search tool. We recommend you check PatCite app wherein you can enter a list of the article IDs and analyse the citing patents per each cited article and examine their family network.
You can use this section to search for only documents that were published or filed between a specific date range. You can select this date range using the in built calendar or enter it in year-month-day format (eg. 1969-07-21).
You can use this section to restrict your results to documents from selected jurisdictions, or to exclude selected jurisdictions.
You can use this section to limit your search to specific document types.
This section allows you to change some search parameters.
- The “Full Text” option will limit your search to only documents which have full text available.
- The “One doc per family” will group your results into simple families, showing only one result from each (this may disrupt the result count and slow down the “select all documents” feature on the search result page).
- Stemming of results can be turned on or off, by default it is turned on. Stemming is an algorithm which modifies your search so only the “stems” of words are checked. For example with stemming on the words: “run”, “runner” and “running” are all stemmed to the word “run” and treated the same in both the query and in target documents. With stemming turned on you will get more results at the cost of some accuracy, and when it is turned off you will have higher accuracy but might miss close results. Note that wildcard search terms may not work as expected for searches where Stemming is turned on. Wildcard terms are not stemmed and therefore may not match against the stemmed values in the search index. For example, “runn*” won’t match “runner”or “running” with Stemming turned on. Read more about stemming on wikipedia.
- Query language allows you to change the language your queries are entered in. This will have an effect on how the queries are interpreted (including how stemming is processed).