Research talk:Automated classification of article importance/Work log/2017-04-06

From Meta, a Wikimedia project coordination wiki

Thursday, April 6, 2017[edit]

Today I'll figure out what items I need to flag in WPMED, then flag those and train a new classifier that also uses that information. Then I'll probably dive into other WikiProjects.

Properties[edit]

I adapted my Python script so that it would also let me know what types of relationships existed between Low-importance items and their respective parent. Turns out, unsurprisingly when you think about it, that the ones we're interested in are all "instance of" relationships. That's information I already should have in my dataset, all I need to do is store the classes these items are an instance of.

Once the data is sufficiently wrangled, we get the follow table of articles that are not Low-importance, but instances of things that are in the table I created created yesterday:

Title Rating Reason
Curie Institute, Warsaw Mid architectural structure
National Institute on Deafness and Other Communication Disorders Mid research institute
Cochrane (organisation) Mid nonprofit organization
International Orthoptic Association Mid organization
Schola Medica Salernitana Mid educational institution, research institute
Marcus Welby, M.D. Mid television series
National Institute of Environmental Health Sciences Mid research institute
History of cancer High aspect of history
Orphanet Mid website
History of psychiatric institutions Mid aspect of history
Mucosal Immunology (journal) Mid scientific journal, periodical literature
Royal College of Physicians Mid organization
National Health Service (England) Mid health system, government agency
National Human Genome Research Institute Mid research institute
Tavistock and Portman NHS Foundation Trust Mid clinic
National Institute on Alcohol Abuse and Alcoholism Mid research institute
Health Advocate Mid business enterprise
World Health Academy Mid organization
Elizabeth Blackwell Medal Mid award
Patient UK Mid website
National Center for Immunization and Respiratory Diseases High research institute
PRA Health Sciences Mid business enterprise
History of herbalism Mid aspect of history
Royal College of Psychiatrists Mid health association
London Ambulance Service Mid NHS trust
PubMed Mid website
Medicines and Healthcare products Regulatory Agency Mid government agency
Libby Zion Law Mid statute
National Institute of Mental Health Mid research institute
By Reason of Insanity Mid television film
Gray's Anatomy Mid book
Shout Gladi Gladi Mid film
Royal College of Physicians of Edinburgh Mid organization, building
In silico clinical trials Mid software
Psychoactive Substances Act 2016 Mid Act of Parliament of the United Kingdom
History of tuberculosis Mid aspect of history
Planned Parenthood Mid nonprofit organization
Bristol-Myers Squibb Mid business enterprise
Royal College of Physicians and Surgeons of Canada Mid organization
History of penicillin Mid aspect of history
DSM-5 Mid book
House of Numbers: Anatomy of an Epidemic Mid film
Translators Without Borders High charitable organization
History of poliomyelitis Mid aspect of history
History of neuraxial anesthesia Mid aspect of history
Sunovion Mid business enterprise
National Institute of Allergy and Infectious Diseases Mid medical research centre
Acorns Children's Hospice Mid organization
Aerospace Medical Association Mid organization
History of medicine High aspect of history
Sushruta Mid human
National Institutes of Health Mid government agency
History of general anesthesia High aspect of history
History of malaria Mid aspect of history
Accountable care organization Mid organization
United States Public Health Service Mid government agency
Annals of Epidemiology Mid scientific journal
Apothecaries Act 1815 Mid Act of Parliament of the United Kingdom
Bahrain Red Crescent Society Mid organization
Harvard Medical School Mid medical school
Médecins Sans Frontières Mid organization
Breast Cancer Alliance Mid business enterprise
Brighton Collaboration Mid organization
British Heart Foundation Mid organization
Caitlin Raymond International Registry Mid hospital
Canadian Thoracic Society Mid organization
Cancer survivor Mid group of humans
Aretaeus of Cappadocia Mid human
Cytotherapy (journal) Mid scientific journal
9-1-1 High emergency telephone number
Epidemiology and Infection Mid scientific journal
European Organisation for Rare Diseases Mid organization
Fields Virology Mid book
Disability-adjusted life year Mid unit of measurement
Health and History Mid scientific journal
Healthcare Information For All Mid organization
Centers for Disease Control and Prevention High government agency
History of pathology Mid aspect of history
History of psychosurgery Mid aspect of history
History of smallpox Mid aspect of history
History of tracheal intubation High aspect of history
How Doctors Think Mid book
Hussaini Blood Bank Mid organization
International Journal of Epidemiology Mid scientific journal
Irish Thoracic Society Mid organization
John E. Fogarty International Center Mid government agency
Journal of Clinical Epidemiology Mid scientific journal
National Institute of Neurological Disorders and Stroke Mid government agency
Lambda Beta Society Mid honor society
National Cancer Institute High research institute
Macquarie University Hospital Mid hospital
Master of Surgery Mid master's degree
National Board for Respiratory Care Mid business enterprise
National Center for Research Resources Mid research institute
National Institute of Arthritis and Musculoskeletal and Skin Diseases Mid research institute
National Institute of Biomedical Imaging and Bioengineering Mid organization
National Institute of General Medical Sciences Mid research institute
National Institute on Drug Abuse Mid government agency
National University Health System Mid health system
National public health institutes Mid organization
Occupational and Environmental Medicine Mid scientific journal
PAMI Mid government agency
Partial hospitalization Mid hospital
Perelman Center for Advanced Medicine Mid hospital
International Red Cross and Red Crescent Movement Mid international organization
International Classification of Primary Care Mid book
Relief of Fevers Mid book
Roy J. and Lucille A. Carver College of Medicine Mid medical school
History of hypertension Mid aspect of history
History of poison Mid aspect of history
Stop TB Partnership Mid organization
TEDMED Mid business enterprise
The Natural History of Alcoholism Revisited Mid book
World Health Organization High organization
Toxin and Toxin-Target Database Mid database
Understanding Lung Sounds Mid book
University of Pennsylvania School of Nursing Mid educational institution
University of Pittsburgh Graduate School of Public Health Mid educational institution
University of Utah School of Medicine Mid medical school
National Center for Biotechnology Information Mid organization
Johns Hopkins Bloomberg School of Public Health Mid research institute
Johns Hopkins School of Medicine Mid medical school
University of Texas Medical Branch Mid research institute, hospital, public educational institution of the United States
A Clinical Lesson at the Salpêtrière Mid painting