References

Hausner, Elke et al. Prospective comparison of search strategies for systematic reviews: an objective approach yielded higher sensitivity than a conceptual one. Journal of Clinical Epidemiology, Volume 77, 118 - 124

Paynter, Robin A. et al. A prospective comparison of evidence synthesis search strategies developed with and without text-mining tools. Journal of Clinical Epidemiology,2021 https://www.jclinepi.com/article/S0895-4356(21)00085-8/abstract

Scells, H et al. A comparison of automatic Boolean query formulation for systematic reviews. Information Retrieval Journal 2021;1.https://www.springerprofessional.de/en/a-comparison-of-automatic-boolean-query-formulation-for-systemat/18527538

Text mining, R etc: texts, training

Book: Text Mining with R: https://www.tidytextmining.com/index.html

Deduplicating in R: https://revtools.net/deduplication.html#locating-duplicates

- And for in order to do that, you also need to import your data with the SynthesisR package.

- this is how you install a package: https://www.tutorialgateway.org/install-r-packages/

http://videolectures.net/Top/Computer_Science/Text_Mining/

http://www.nltk.org/

Course units of interest

http://rgcl.wlv.ac.uk/masters-and-phd-studies/

https://www.cs.bham.ac.uk/research/groupings/natural-language-processing/

ICASR and its conferences

Evidence Synthesis Hackathon’s online Evidence Synthesis & Meta-Analysis in R Conference (#ESMARConf), to be held on January 21st and 22nd 2021. https://www.eshackathon.org/events/2021-01-ESMAR.html

Welcome to Text Mining with R

• https://www.tidytextmining.com/

• find the book

An Introduction to Text Processing and Analysis with R

Safe exclusions

‒NOT should be used carefully

‒There are methods that can be used e.g.

1.Relevant topic A

2.(excluded topic NOT A).ti.

3.1 NOT 2

‒But this relies on A being consistently expressed

If we want to find studies of asthma in adults we could experiment with the following approach which focuses on removing children:

1.Search results

2.((children or child or boy or boys or girl or girls or infant or infants or baby or babies or teenager*) NOT (adult* or men or women or patient*)).ti.

3.1 NOT 2

Text mining/machine learning: clinical trial registries. Examples

Rybinski, M.; Xu, J.; Karimi, S. (2020): Clinical trial search: Using biomedical language understanding models for re-ranking. In J Biomed Inform 109, p. 103530. DOI: 10.1016/j.jbi.2020.103530.

Mayer, C. S.; Huser, V. (2020): Computerized monitoring of COVID-19 trials, studies and registries in ClinicalTrials.gov registry. In PeerJ 8, e10261. DOI: 10.7717/peerj.10261.

Lanera C, Minto C, Sharma A, Gregori D, Berchialla P, Baldi I. Extending PubMed searches to ClinicalTrials.gov through a machine learning approach for systematic reviews. J Clin Epidemiol. 2018 Nov;103:22-30. doi: 10.1016/j.jclinepi.2018.06.015. Epub 2018 Jul 5. PMID: 29981872.