About TextMed

What is TextMed?
Are you a commercial site?
How do you get your medical abstracts?
What are entities?
What are sources?
How can I locate a given entity?
Why does my search return no results?
What does "Coref./Ref." in entity pages mean?
What happens if I click the blue and orange "Coref./Ref." symbol?

What is TextMed?
TextMed is a search engine for medical entities: diseases, drugs, chemicals, organs and organisms.TextMed aims to identify relationships between these medical entities.TextMed uses natural language processing techniques to track medical entity references from the scientific literature, and a variety of statistical techniques to analyze the relationships between them. TextMed presents our analysis of roughly 15 million Medline/PubMed medical abstracts, including the latest abstracts analyzed as they arrive each day.

Are you a commercial site?
No.  TextMed is a research project lead by Steven Skiena at the Computer Science Department at SUNY Stony Brook. TextMed is an academic research project, not a commercial operation.  However, TextMed technology is available for licensing through the Research Foundation at Stony Brook University.

How do you get your medical abstracts?
TextMed licences these abstracts directly from the National Library of Medicine. We do not scrape any web pages for this part of the project.

What are entities?
Entities are such things as diseases, drugs, chemicals, organs and organisms. TextMed identifies interesting entities by analyzing medical abstracts. Each given entity gets its own page, where we identify other entities which which it shares significant relationships. Important associations are identified on each of three different time scales namely cumulative analysis, references over the last 30 days and references over the last 7 days.
Similar: What does "Coref./Ref." in entity pages mean?

What are sources?
Sources are the different medical journals where the abstracts were published in.
We maintain a list of sources with links to information pages for each journal.

How can I locate a given entity?
Type in the full name of your favorite entity into the search window and click the "Search!" button.
If you are lucky, you will go directly to the entity page you requested.
If your query matches multiple entities, you will be presented with a table of candidate entities.

Another approach is to track down your entity through our entity indices,
which are broken down by first letter and type.  Click the "Entities" button on the main page.

Our search engine is not (yet) competition for Google, so one final approach is to ask Google to find your query with the term TextMed added to the search.  Presumably we will be the top entry.

Why does my search return no results?
There are two possible explanations.
First, TextMed is not aware of your entity of interest.
All TextMed entities are extracted from online news sources.
If you want your own page on TextMed you must do something newsworthy!

The other possibility is that your entity is present but your search was inadequate. To do better, check out:
Similar: How can I locate a given entity?

What does "Coref./Ref." mean in entity pages?
References track the frequency with which we observe a given entity through our corpus of news articles.
Coreferences track the frequencies with which two given entities appear together in news articles.
In our graphic depiction, the length of orange bar is a function of the entity references while the length of the blue bar is a function of entity coreferences.
A higher ratio of blue implies a stronger relationship between the two entities.

What happens if I click the blue and orange "Coref./Ref." symbol?
Clicking on the Coref./Ref. bar gives links to recent articles in which both entities appear together.