Authors :
Chou-Cheng Chen
Volume/Issue :
Volume 7 - 2022, Issue 11 - November
Google Scholar :
https://bit.ly/3IIfn9N
DOI :
https://doi.org/10.5281/zenodo.7439942
Abstract :
PubMed database stores more than 34 million
citations and abstracts of biomedical literature published
in approximately 30,000 Journals. “Best Match” function
of the PubMed query website helps users sort selected
match publications, but it does not provide suggested
journals for users to submit manuscripts. A previous study
showed Jane is a website tool with an elegant algorithm
that provides journal suggestions and finds related articles
for users; however, it merely finds journals that have been
indexed from its work. This study thus creates
CatchStudy, which is a windows form tool to help users
currently extract information from PubMed and find
suitable journals fitting the scope of the user. CatchStudy
is based on PubstractHelper and adds a new approach to
give a fitted score for each abstract, thereby sorting
abstracts, and it can significantly help the user extract
information from PubMed more quickly than “Best
Match” of PubMed (p < 0.05). CatchStudy also provides
suitable journals for user submission by calculating the
sum of the fitted scores in abstracts that are published in
the same journal. Users can also provide their title and
abstract from their manuscript, and CatchStudy selects
highly related nouns to query. CatchStudy uses these
selected highly related nouns from using natural language
processing (NLP) to token parts of speech and sorts by
weight, and then queries PubMed for retrieving abstracts.
Suitable journals are provided for the user by summing up
fitted scores in abstracts. Comparison between Jane and
CatchStudy shows that CatchStudy provides more
journals for the user as reference to submit their
manuscript. The software can be download from
https://drive.google.com/drive/folders/1h65rIt8Udz2KJc3
Ass4IE46BJ4E9Qa_t. To the best of my knowledge,
CatchStudy provides another efficient method to extract
information from querying PubMed and retrieving
suitable journals for submission of the user’s manuscript.
Keywords :
Text Mining, Fitted Abstracts, Suitable Journals, Natural Language Processing.
PubMed database stores more than 34 million
citations and abstracts of biomedical literature published
in approximately 30,000 Journals. “Best Match” function
of the PubMed query website helps users sort selected
match publications, but it does not provide suggested
journals for users to submit manuscripts. A previous study
showed Jane is a website tool with an elegant algorithm
that provides journal suggestions and finds related articles
for users; however, it merely finds journals that have been
indexed from its work. This study thus creates
CatchStudy, which is a windows form tool to help users
currently extract information from PubMed and find
suitable journals fitting the scope of the user. CatchStudy
is based on PubstractHelper and adds a new approach to
give a fitted score for each abstract, thereby sorting
abstracts, and it can significantly help the user extract
information from PubMed more quickly than “Best
Match” of PubMed (p < 0.05). CatchStudy also provides
suitable journals for user submission by calculating the
sum of the fitted scores in abstracts that are published in
the same journal. Users can also provide their title and
abstract from their manuscript, and CatchStudy selects
highly related nouns to query. CatchStudy uses these
selected highly related nouns from using natural language
processing (NLP) to token parts of speech and sorts by
weight, and then queries PubMed for retrieving abstracts.
Suitable journals are provided for the user by summing up
fitted scores in abstracts. Comparison between Jane and
CatchStudy shows that CatchStudy provides more
journals for the user as reference to submit their
manuscript. The software can be download from
https://drive.google.com/drive/folders/1h65rIt8Udz2KJc3
Ass4IE46BJ4E9Qa_t. To the best of my knowledge,
CatchStudy provides another efficient method to extract
information from querying PubMed and retrieving
suitable journals for submission of the user’s manuscript.
Keywords :
Text Mining, Fitted Abstracts, Suitable Journals, Natural Language Processing.
