Introduction
Prior Art Search is a crucial part of the patent examination process. Its purpose is to identify previously registered patents and scientific literature in related fields of invention. One of the main challenges in this process is designing search queries that provide accurate and relevant results without overwhelming the examiner with irrelevant patents or returning too few results. This article discusses how to structure search queries effectively using Boolean operators (AND, OR, NOT) and other techniques such as Truncation and Wildcards. Common mistakes in query design and how to avoid them will also be covered.
A Technical Topic for the Search
For this article, we assume the search topic is Ethanol Production from Biomass, which is a significant area in renewable energy technologies. The goal of the prior art search in this case is to identify existing patents related to ethanol production processes from starch, cellulose, and other biomass sources.
1. Combining Keywords with Boolean Operators:
In patent databases, Boolean operators can be used to combine different keywords. These operators help refine your query and get more accurate results.
1.1 AND
When you use AND, only patents containing all of the search terms will be returned. This operator helps narrow your search.
Example:
("ethanol production" AND "starch conversion")
In this example, only patents containing both "ethanol production" and "starch conversion" will be returned.
1.2 OR
OR is used to include patents that contain at least one of the search terms. This operator broadens your search to include a variety of relevant results.
Example:
("ethanol production" OR "biofuel production")
This search will return patents that contain either "ethanol production" or "biofuel production."
1.3 NOT
NOT is used to exclude certain words from your search. This operator helps remove irrelevant results.
Example:
("ethanol production" AND "starch conversion") NOT "corn"
This query returns patents that contain "ethanol production" and "starch conversion," but excludes patents related to "corn."
2. Building Multilayered Queries
Multilayered queries, also known as complex queries, are combinations of multiple Boolean operators that allow for more refined, targeted searches. This approach enables you to design precise searches and quickly obtain relevant information.
Example:
("ethanol production" AND ("starch conversion" OR "cellulose hydrolysis")) AND (C12P 7/00 OR C12N 9/00)
This query searches for patents related to "ethanol production," with either "starch conversion" or "cellulose hydrolysis," while also filtering for CPC classifications C12P 7/00 or C12N 9/00, which are related to biological processes.
3. Common Mistakes in Query Design
There are several common mistakes made when designing queries that can lead to inefficient searches. Below are some of the most common errors and how to avoid them:
3.1 Overuse of OR
Excessive use of OR can result in an overwhelming number of irrelevant patents that clutter your search results.
Solution: Use AND more frequently to narrow the search and make it more precise.
3.2 Improper Use of NOT
Improper use of NOT can reduce the number of relevant results, as it excludes patents that might be pertinent to the search.
Solution: Be cautious when using NOT. Ensure that the words you're excluding are genuinely irrelevant to your search.
3.3 Ignoring Synonyms and Variants
Sometimes, limiting your search to a specific set of terms can lead to missing out on relevant patents. It’s important to consider synonyms and different variations of technical terms.
Solution: Use variations and synonyms. For example, "ethanol production" could also be searched as "alcohol production" or "bioethanol production."
4. Using Truncation and Wildcards
Truncation and Wildcards are powerful techniques that allow you to expand your search to include multiple forms or variations of a word.
4.1 Truncation
Truncation allows you to search for a root word and its variations by using a specific symbol (e.g., "*"). This symbol represents any suffix added to the root word.
Example:
"biofuel*"
This search will return results for words like "biofuels," "biofueling," and "biofuelled."
4.2 Wildcard
Wildcard characters, often represented by a question mark (?), are used to replace one or more characters in a word. This helps refine your search and catch different spellings.
Example:
"cellul?se"
This query will return results for both "cellulose" and "cellulase."
Conclusion:
Search query engineering for prior art searches requires careful thought and attention to detail. By combining Boolean operators, using Truncation and Wildcards, and building multilayered queries, you can refine your searches and obtain more accurate results. Being aware of common mistakes and learning how to avoid them will significantly improve the quality of your search and help identify relevant patents efficiently.