Add research workflow guide

README.md

@@ -1,3 +1,100 @@
-# patent-literature-search-guide
+# Searching Patent Literature for Academic Research: A Practical Guide
 
-A practical guide to searching patent literature for academic research, with a source-linked workflow using Yila https://yila.ai
+Patents are one of the largest and least used sources of technical information. Millions of documents describe inventions in chemistry, engineering, medicine, software and agriculture, often years before similar work appears in journals, and sometimes without ever appearing there at all. Many researchers ignore patents because they seem legalistic, hard to search and outside the normal scholarly record. That is a missed opportunity. Patent literature can reveal prior art, show where industry is investing, identify collaborators and competitors, and fill gaps in reviews of applied fields. This guide explains what patent documents contain, where to search them, how to build effective queries, how to read a patent efficiently and how to cite patents in academic work. It also describes how a source-linked research assistant such as [Yila](https://yila.ai) can help connect patent findings to the journal literature while keeping each claim tied to its source.
+
+## Why patents matter to researchers
+
+A patent grants its owner exclusive rights to an invention for a limited period, usually twenty years from filing, in exchange for public disclosure. That disclosure must be detailed enough for a skilled person to reproduce the invention. As a result, patents often include experimental examples, formulations, process parameters and drawings that are more specific than those in journal articles.
+
+For researchers, patents are useful in several ways:
+
+- **Prior art and novelty.** Before claiming that a method is new, check whether it has been patented or disclosed in a patent application.
+- **Technology landscapes.** Patent counts by field, applicant and country reveal where innovation is concentrated and how it changes over time.
+- **Applied detail.** Industrial patents can describe scale-up, manufacturing conditions and practical problems rarely discussed in academic papers.
+- **Commercial context.** Knowing who holds patents in an area helps assess translation potential and freedom-to-operate questions, although legal advice is needed for the latter.
+- **Systematic reviews in applied fields.** Engineering, materials science and pharmaceutical reviews can be incomplete without patents.
+
+## Anatomy of a patent document
+
+Most patent documents follow a standard structure.
+
+- **Bibliographic data:** publication number, application number, filing and priority dates, applicants, inventors and classification codes.
+- **Title and abstract:** often brief and sometimes deliberately vague.
+- **Description:** background, summary of the invention, detailed description and examples. This is where the technical substance lies.
+- **Claims:** numbered statements defining the legal scope of protection. Claims are precise but written in a specialised style.
+- **Drawings:** diagrams, flowcharts and figures referenced in the description.
+- **Citations:** prior art cited by the applicant or the examiner, which can lead you to related patents and papers.
+
+Patent applications are usually published eighteen months after the priority date, whether or not a patent is eventually granted. Granted patents are published again. Keep track of whether a document is an application or a grant, and of the family of related documents filed in different countries.
+
+## Where to search
+
+Several free databases cover most needs.
+
+- **Espacenet**, from the European Patent Office, covers more than 150 million documents worldwide with machine translation and family information.
+- **Google Patents** offers simple full-text search, quick access to PDFs and links to non-patent literature.
+- **WIPO Patentscope** covers international applications under the Patent Cooperation Treaty and many national collections.
+- **USPTO Patent Public Search** provides access to United States patents and applications.
+- **The Lens** combines patents with scholarly works, making it particularly helpful for linking patents to research papers.
+
+Commercial databases add curated indexing, analytics and better coverage of some regions, but free tools are adequate for most academic purposes.
+
+## Building effective queries
+
+Patent language differs from journal language. Applicants often use broad or unusual terms to widen protection: a "fastening means" rather than a "screw", or a "data processing apparatus" rather than a "computer". Keyword searches alone therefore miss many relevant documents.
+
+### Use classification codes
+
+The Cooperative Patent Classification (CPC) and the International Patent Classification (IPC) assign every patent to detailed technical categories. Searching by classification code captures documents regardless of wording. Find codes by searching a few known relevant patents and noting their classifications, or browse the classification scheme on Espacenet. Combine codes with keywords for precision.
+
+### Expand vocabulary
+
+List synonyms, broader terms and functional descriptions. Include British and American spellings, technical and lay terms, and chemical names alongside trade names. Truncation and proximity operators help capture variants.
+
+### Follow citations and families
+
+Once you find a highly relevant patent, look at the documents it cites and the documents that cite it. Family members filed in other countries may include different claims or examples. Citation chaining is often more productive than repeated keyword searches.
+
+### Search by applicant and inventor
+
+Searching for leading companies, universities and inventors in your field can reveal portfolios that keyword searches miss.
+
+## Reading a patent efficiently
+
+Patents can be long. A practical approach is:
+
+1. **Read the abstract and look at the drawings** to grasp the invention quickly.
+2. **Read the first independent claim** to understand the core of what is protected.
+3. **Skim the background section** for the applicant's view of prior art and the problem being solved.
+4. **Focus on the examples** in the detailed description, which contain experimental data and parameters.
+5. **Check the legal status** if relevance to current practice matters: whether the patent was granted, is in force or has lapsed.
+
+Remember that patents are written to secure rights, not to report findings objectively. Examples are selected to support the claims, data are rarely subjected to peer review and negative results are absent. Treat patent data as evidence of what was disclosed, not as validated findings.
+
+## Using patents in literature reviews
+
+In a systematic or scoping review, decide in advance whether patents are in scope and justify the decision. If they are included, report the databases, search strings, classification codes, date limits and number of records retrieved, just as you would for bibliographic databases. Deduplicate patent families so that one invention is not counted multiple times. Describe how you appraised patents, recognising that standard quality tools for studies do not apply directly.
+
+Patent analyses can also be a separate component of a review, mapping activity by year, country and applicant to complement the academic evidence. Such landscapes are useful for showing the gap between research and commercial development.
+
+## Citing patents
+
+Most citation styles have formats for patents. A typical citation includes the inventor or applicant, title, patent or publication number, issuing office and date. For example, in APA style: Inventor surname, initials (Year). *Title of patent* (Patent No. US 1,234,567). Issuing office. Cite the specific document you read and note if it is an application rather than a granted patent.
+
+## Common pitfalls
+
+- **Relying on titles and abstracts alone.** These are often uninformative; the claims and examples matter.
+- **Ignoring non-English documents.** Much patent activity occurs in China, Japan and Korea. Machine translation in Espacenet and Google Patents is adequate for screening.
+- **Confusing applications with grants.** Many applications are never granted.
+- **Assuming patents prove efficacy.** A patent shows that an invention was claimed and disclosed, not that it works as described.
+- **Drawing legal conclusions.** Questions about infringement or freedom to operate require professional advice.
+
+## Using Yila alongside patent databases
+
+Patent searches tell you what was disclosed; journal literature often tells you what was independently tested. Linking the two is time-consuming. [Yila](https://yila.ai) is a research assistant that answers questions with citations linked to the underlying sources. After identifying key technologies in patents, you can ask Yila which published studies have evaluated them, then open each cited paper to check methods and findings.
+
+Because each statement remains attached to its citation, you can keep a clear record of which claims come from patents and which from peer-reviewed studies. This distinction matters when writing a review or assessing whether a technology has robust evidence behind it. The assistant speeds up the connection between the two literatures; your judgement determines how to weigh them.
+
+## Summary
+
+Patent literature is a rich, underused source of technical detail and evidence of innovation. Free databases such as Espacenet, Google Patents, Patentscope and The Lens make it accessible to any researcher. Effective searches combine classification codes, varied vocabulary and citation chaining. Read patents strategically, focusing on claims and examples, and treat their data with appropriate caution. When patents are part of a review, report search methods transparently and deduplicate families. Combined with tools that keep every claim tied to its source, patent searching can strengthen literature reviews and help you see the full picture of research and development in your field.