نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشگاه علوم پزشکی شهید صدوقی
2 گروه علم اطلاعات و دانش شناسی، دانشکده علوم تربیتی، دانشگاه اصفهان
3 اداره کل کتابخانههای عمومی استان یزد، یزد
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Purpose: Diverse research in information literacy necessitates analyzing the topics of these studies to gain a clear and comprehensive understanding of this area. The current research aims to apply topic modeling to published scientific productions related to health information literacy using the PubMed database.
Method: This study employed a quantitative approach with an applied focus, utilizing text-mining techniques. Scientific publications in information literacy were extracted from the PubMed database using the MeSH term "information literacy" [Majr] without any time constraints. A search on August 5, 2024, yielded 8407 records from 1519 journals and books. Subsequently, the abstracts and titles of the articles were saved in text format and then converted into a structured Excel format for analysis. After removing null records, 6811 records with abstracts were used for analysis. The process involved tokenization, removal of punctuation and stop words, stemming, and conversion of text data into numerical vectors to apply machine learning techniques. Finally, topic modeling was performed using the Latent Dirichlet Allocation (LDA) algorithm. After data cleaning, the abstracts and titles of these articles were analyzed and topic modeled using the Pandas, PyLDAvis, sklearn, PyLDAvis, numpy, Setuptools, NLTK, Gensim, Wordcloud, and Seaborn libraries.
Findings: Analysis of the retrieved articles using the TF-IDF algorithm revealed that the terms "patients," "mental," "mental health," "information," and "care" had the highest term frequency-inverse document frequency weights.
Using Latent Dirichlet Allocation, seven thematic clusters were identified, including "Online Health Information Seeking and Digital Health Literacy"; "Impact of Health Literacy on Decision-Making"; "Readability of Patient Education Materials"; "Health Literacy in the COVID-19 Pandemic"; "Mental Health Literacy"; "Oral Health Literacy"; and "Communication in Healthcare."
In terms of the percentage of research productions in the field of information literacy, it was found that the topic of "Mental Health Literacy" had the highest percentage with 22%, followed by "Impact of Health Literacy on Decision-Making" with 19%. On the other hand, "Health Literacy In The COVID-19 Pandemic" had the lowest percentage of scholarly output with only 2%. The growth trend of scientific production in each of the extracted topics showed that the highest growth rate was observed in the topic cluster " Communication in Healthcare," followed by the " Online Health Information Seeking and Digital Health Literacy " topic.
Conclusion: The extracted thematic clusters from the scientific productions on information literacy demonstrated good coherence and strong thematic relationships; therefore, this research can significantly contribute to researchers in improving scientific production in the field of health information literacy.
کلیدواژهها [English]
ارسال نظر درباره این مقاله