Text Mining for Search Strategy Development

PubVenn

PubVenn allows you to explore PubMed visually by generating Venn diagrams. You can enter any multi-term search to generate a Venn diagram that allows you to view the size of the citation set for each term as well as how those sets interact.

Availability: Free online

Data Source: PubMed

Tool Type: Visualisation tool

Features: Provides example PubMed search strategy and links to relevant publications. Selecting the 'Expanded subjects' checkbox will include other relevant terms from PubMed.

Import Formats: Free text

Export Formats: Save as .png

Known Limitations: PubMed data only

URL: https://pubvenn.appspot.com/

Also try Search Workbench: https://searchworkbench.info/

PubMed PubReMiner

Enter a free text search into the PubReMiner tool and it will query PubMed for relevant results.
The tool analyses these results and provides tables that rank the frequency of words in the title and abstract of the articles and also relevant MeSH headings. Other ranked tables include journals in which your query is published the most and authors which are most active in related fields.

Availability: Free online

Data Source: PubMed

Tool Type: Text frequency

Features: Self selection of elements to display in the results. Direct connection to PubMed

Import Formats: Free text

Export Formats: Save results as a .txt file

Known Limitations: Limited to PubMed

URL: http://hgserver2.amc.nl/cgi-bin/miner/miner2.cgi

Coremine

Coremine Medical is a product of the PubGene Company designed to be used by anyone seeking information on health, medicine and biology. It is ideal for those seeking an overview of a complex subject while allowing the possibility to "drill down" to specific details. Search results are presented in a dashboard format comprised of panels containing various categories of information ranging from introductory sources to the latest scientific articles. Coremine presents search results as a graphic network that describes relationships discovered through text-mining.

Availability: Register for free account

Data Source: PubMed

Tool Type: Visualisation, clustering, text frequency, relationship networks

Features: File upload, Hyperbrowser, search history, alerts

Import Formats: Free text, tab delimited file can be uploaded

Export Formats: None

URL: https://www.coremine.com/medical/

Instructions: https://www.coremine.com/medical/help.html

MeSH on Demand

The MeSH on Demand tool uses the NLM Medical Text Indexer to identifiy MeSH vocabulary in submitted text (paste up to five pages). The results are displayed in a list as well as highlighted in the pasted text (also defines term frequency within the text). Links to similar PubMed related citations are included.

Availability: Free online

Data Source: PubMed

Tool Type: Text analyser

Features: Suggests MeSH vocabulary found in similar, related citations as well as the submitted text.

Import Formats: Free text

Export Formats: Text file

Known Limitations: Non English text needs to be translated first

URL: https://meshb.nlm.nih.gov/MeSHonDemand

Yale MeSH Analyser

Availability: Free online

Data Source: PubMed

Tool Type: Comparison

Features: Easy to view comparison of article indexing

Import Formats: Free text (type or paste)

Export Formats: Excel, HTML table

Known Limitations: Query with PubMed IDs only - maximum 20

URL: http://mesh.med.yale.edu

HelioBLAST

HelioBLAST is a free service provided by HelioText. The HelioBLAST text similarity engine finds text records that are similar to the submitted query. Your query is searched against the citations (abstract and titles) in Medline/PubMed and the top matching articles are returned in the results

Availability: Free online

Data Source: PubMed

Tool Type: Word analysis

Features: 'Implicit keywords' - additional word frequency analysis tool

Import Formats: Free text - copy and paste

Export Formats: No export function but links to article records in PubMed

URL: https://helioblast.heliotext.com/

Carrot2

Carrot2 is an Open Source Results Clustering Engine that can automatically organise search results into topics. Carrot2 can query PubMed and allows boolean searching.

Availability: Free online, download

Data Source: PubMed

Tool Type: Clustering

Features: Simple online interface

Import Formats: Free text

Export Formats: View on screen

Known Limitations: No export options

URL: https://search.carrot2.org/#/search/web