Skip to main content
NCBI home page
Elsevier - PMC COVID-19 Collection logoLink to Elsevier - PMC COVID-19 Collection
. 2022 Dec 22;36:101144. doi: 10.1016/j.imu.2022.101144

ParsBERT topic modeling of Persian scientific articles about COVID-19

Mohammad Dehghani a,∗, Fezzeh Ebrahimi b
PMCID: PMC9771580  PMID: 36573134

Abstract

Purpose

The COVID-19 pandemic has indisputably impacted every aspect of human life, and a host of studies have investigated its different aspects. This paper models the contents of Persian literature on COVID-19.

Method

This is a descriptive-exploratory study in which 815 articles were collected from the Magiran database. The articles were published before March 2022. The abstracts and titles were used in the modeling. The modeling was performed by combining the latent Dirichlet allocation (LDA) algorithm with ParsBERT.

Findings

Topic modeling indicated ten major topics, including medicine, psychology, humanities, politics, management, biology, economics, culture, engineering, and religion. The articles under the category of medicine had the largest cluster (42.3%), while engineering and religion had the smallest clusters (1.1% each).

Conclusion

The found topics in the created clusters have structural relationships. The COVID-19 effect on physical and mental health (medical and psychological topics) is the most crucial factor. These clusters provide evidence that COVID-19 affects all facets of human society at three levels: the individual, family, and society. Aside from the ten critical clusters in the humanities field, the utmost disorder is related to teaching and learning. For the first time, this research has presented a model of scientific communication in the field of COVID-19 based on the data collected from a Persian database – Magiran.

Keywords: Corona, COVID-19, LDA, ParsBERT, Topic modeling

1. Introduction

As a severe public health concern worldwide [1], COVID-19 has impacted the entire world, controlling which requires a more in-depth understanding of the disease. The global struggle with the pandemic has revealed how this virus has influenced every aspect of life today. The rapid and widespread spread of the virus has contributed to the development of Infodemia. Reviewing relevant studies can help achieve a more thorough understanding of the disease. Since most scientific results are published in journals, analyzing them can provide valuable information [2]. In a scientific paper, the abstract provides comprehensive information on the research. The abstract contains essential content and is written with much effort. Metadata such as title, author's name, year of publication, and place of publication are standard features that can be retrieved and accessed in most databases for similarity measurement. However, retrieving and analyzing abstracts from most databases is not easy. Topic modeling is an unsupervised machine-learning technique that identifies abstract topics in an extensive collection of documents. It is one of the abstract processing techniques discussed in the article.

Latent Dirichlet Allocation (LDA), Latent semantic Analysis (LSA), probabilistic LSA [3], and Non-Negative Matrix Factorization (NMF) are among the famous techniques for identifying latent topic patterns. The LDA technique has recently gained significant popularity in various fields [4,5]. The algorithm can discover the topics and their between-relationships based on the distinction of co-occurring vocabulary clusters. Topic modeling is an unsupervised machine-learning technique for finding abstract topics in large sets of documents. The case helps organize, realize, and summarize large sets of textual information and discover latent themes that differ among documents in a given collection. In recent years, different studies have employed natural language processing methods to analyze scientific productions. Language processing methods are generally divided into lexical analysis in bibliography and topic modeling.

The previous studies on lexical analysis in bibliometrics identified keywords as the main element. In these studies, the main tools used in SPSS were Hayward, Between-groups linkage & within-groups linkage algorithms, and clustering was conducted using about 50–100 keywords with the highest frequency. First, the co-occurrence matrix of the words was examined, followed by clustering [[6], [7], [8]]. Besides, the LDA algorithm was used in some studies to analyze abstracts by topic modeling. Accordingly, since our method shares some similarities with this method, some of these studies are reviewed below:

Authors in Ref. [9] used the LDA algorithm to extract and identify undesirable drug reactions. Authors in Ref. [10] identified and topic modeled the literature on teaching and learning during the COVID-19 pandemic in Scopus. In that study, six main crucial topics were recognized: the impact of COVID-19 on higher education institutions and the challenges ahead of them, the use of various instruments and educational strategies by these institutions, the teaching and learning experience of schools and school teachers, the impact of COVID-19 on the education of healthcare workers, the information related to COVID-19 and therapeutic strategies for patients, and students' mental health as a result of COVID-19 implications and e-learning. The topic modeling of Iranian publications on COVID-19 in the LitCOVID database indicated eight topic groups: prevention, treatment, diagnosis, prediction, case report, mechanism, transmission, and communal [6].

The LDA method was applied to PubMed data to explore the publications on COVID-19 concerning 25 topics [11]. The topic modeling of 154 articles on quantum mechanics collected from Phy.org indicated ten main topic clusters. The data on COVID-19 were also extracted and were topic modeled from Facebook [12]. In addition, 14 main research topics were identified by applying the LDA algorithm to the data extracted from PubMed. The most common topics were healthcare responses and clinical manifestations [13].

The novelties of this research can be outlined from several perspectives. Since there is no data receipt in the Magiran database, which is a comprehensive database of Persian scientific publications, the COVID-19 data provided here were collected by coding, making this paper the first work in topic modeling the COVID-19 case in Persian databases. The adopted approach is new and highlights meaning. The ParsBERT algorithm is a recent monolingual language model for vectorization of Persian texts with a significant advantage: It takes into account the semantics and meaning of Persian sentences. In earlier studies, the TF-IDF and Bag of Words models were used to apply vectorization, while here, the LDA and Pars-Bart algorithms were used for topic modeling. First, the dimensions were reduced using the SOM algorithm and visualization. The SOM algorithm covers and optimizes data density better than other algorithms by moving the neural units. In addition, another novelty is the simultaneous use of two aspects – titles and abstracts – to enhance the analysis; one of these two aspects (the abstract) has been used in earlier research.

In order to investigate the previous studies on COVID-19, this paper presents a topic modeling of scientific publications on COVID-19 using artificial intelligence and text mining technologies. Our results will not only show essential issues highlighted in these studies but also assist in predicting future research directions. We have extensively reviewed the Persian research on COVID-19 in the Magiran database to achieve the following objectives: to prepare and compile a comprehensive dataset of Persian scientific studies on COVID-19, implement the latest topic modeling techniques suitable for Persian, and thematically analyze and visualize the available data.

2. Methodology

This research used descriptive-exploratory methods to explore 815 articles, which were indexed in the Magiran database, and published until March 2022. The keywords included Corona, Coronavirus, COVID, and COVID-19.

Two Python libraries, including BeautifulSoup and Request libraries, were used for data extraction. Table 1 shows translations of the titles, abstracts, keywords, and authors of the extracted articles. The data is accessible via https://github.com/mohamad-dehghani/CoPers.

Table 1.

The translated excerpts of data crawled from the magiran database.

Title Abstract Author Keywords
A Comparative Study of Iran and Singapore: Performance in Dealing with COVID-19 and Their Future The COVID-19 pandemic has become a major international concern of colossal proportions. The virus has been able to give states powers that would not normally be possible, imposing many restrictions. This paper is a comparative study of the performance of the Iranian and Singaporean states in dealing with the pandemic. It explores the political-social future of these two nations and the ultimate status of the disease. Mohsen Seifabadi Shafiee, Ali Dolat Abadi Iran, Singapore, Coronavirus, Future Study, State
Pregnancy Outcomes and Clinical Manifestations of COVID-19 in Pregnant Women: A Narrative Review The current epidemic of SARS-CoV-2 is intense and worrying worldwide, and the number of pregnant women with the virus is on the rise. In pregnant women with severe acute respiratory syndrome (SARS) and Middle East severe acute respiratory syndrome coronavirus (MERS), morbidity and mortality are higher than in non-pregnant women. Maryam Nikpour, Mahboobeh Darzipoor, … Pregnancy, COVID-19, SARS-CoV- 2, Narrative Review
Analysis of the News: on The Coronavirus Crisis A crisis, in its various forms, is considered a silent war. Rather than affecting the country's hardware, the coronavirus crisis damaged and disrupted its software system. A widespread crisis is a threat. Endangered societies will face new problems affecting people's lifestyles and social behavior. At the time of widespread crisis, endangered or vulnerable society will undergo fundamental changes and determine its distance from what has been before. This situation is critical because it affects the individual, social and human relations more than anything else and requires investigation. Hassan Bashir Coronavirus, Crisis, Society, Endangered, Society, Confidence, Building, CoronaPhobia, Rumor
Psychological Coping Strategies and Techniques to deal with Stress & Anxiety around Coronavirus (COVID-19) In early 2020, people were plagued by a common problem beyond color, race, social class, nationality, or geography: COVID-19. Today, anxiety, worry, and stress are increasing worldwide, and dozens of people witness death every hour. In the days when society is facing more than ever an unbalanced amount of misinformation about the disease and the state of the world, and the anxiety and stress of getting sick have become more annoying than getting the disease, the social and moral responsibility of psychologists and counselors demands that take this step. This paper mainly aims to identify the stress and anxiety associated with COVID-19 and provide practical solutions to managing them. Mohammad Khodayarifar, Mohammad Hassan Asayesh Stress, COVID-19, Coronavirus, Disease Anxiety, Corona Anxiety Syndrome
Evaluation of Mutation Spread in The SARS-CoV2 Genome The Coronaviridae family includes viruses that are the causative agents of respiratory infections. Human RNA viruses have the largest genome. Coronaviruses undergo elusive genetic changes through mutation during replication. A gene mutation is a permanent alteration in the nucleic acid sequence. If mutation occurs in large numbers, it changes the biological features of a species. Kamal Fakhredini, Hooreih Soleimanjahi, Razieh Sadat Banijamali, Taravat Bamdad1 SARSCoV2, COVID-19, Mutation, Coronavirus
Open in a new tab

After extraction, the data were cleaned, and the items with null values were eliminated. Finally, 718 articles were left. Next, topic modeling was conducted on the article titles and abstracts using Python 3.9 by combining the LDA and ParsBERT. First, tokenization was performed, and punctuations and stop words were removed for data pre-processing. Then, stemming and lemmatization were performed, and the words were converted into root states. Furthermore, the redundant white spaces were removed. To pre-process the data, two Hazm and Parsivar tools were employed. Since NLTK and SpaCy do not support Persian, Hazm and Parsivar were used instead. These are dedicated tools unique to Persian. Since each tool has unique advantages and disadvantages, both tools were utilized.

The data consisted of unstructured abstracts and titles, which should be converted into structured and computable data. Bag of Words and the TF-IDF model are the methods applicable for this purpose. In this study, however, a newer method was used to convert text and words into vectors based on the ParsBERT algorithm. This algorithm was used to make the feature matrix for each word with 768 dimensions, yielding a vector with 768 dimensions for each input (text) after averaging. These vectors were transferred to the LDA algorithm, which is one of the most famous algorithms for identifying hidden topic patterns. After using ParsBERT and LDA to model the topic, visualization was required to help better display the topics. Additionally, given that the ParsBERT algorithm gives each text a total of 768 vectors, and this dimension cannot be displayed, the SOM algorithm [14] was used for dimension reduction, and the 768-dimensional space was reduced to two dimensions. Applying the SOM, TSNE, UMAP, and PCA algorithms is a common technique for dimension reduction. Since the SOM algorithm covers and optimizes data density and compression better than other algorithms by shifting neural units, it was used in this article to reduce dimensions. In addition, it can be used to a greater extent for clustering words [15]. The ability of two-dimensional visualization is one of the most important features of SOMs, which is not used by other data analysis tools [16]. In this paper, the heatmap and network packages were also used to draw graphs and correlation diagrams.

2.1. Findings

In this study, 812 articles were initially retrieved from Magiran, which were used for analysis after cleaning 718 records. Fig. 1 shows the word cloud of abstracts and titles of articles at this stage.

Fig. 1.

Fig. 1

Open in a new tab

The word clouds of the article abstract and titles related to COVID-19.

In the next step, the LDA and ParsBERT algorithms were applied to topic modeling. The topics of the extracted documents weres divided into ten main clusters. Fig. 2 shows the frequency of the topics. The most popular research interests were in the humanities, psychology, and medical fields, accounting for 42.34%, 13.65%, and 11.28% of the entire collection.

Fig. 2.

Fig. 2

Open in a new tab

The frequency of clusters created by scientific productions in the field of COVID-19.

As shown in Fig. 3 , these ten clusters have nothing in common and overlap, indicating that the topics are distinctly segregated.

Fig. 3.

Fig. 3

Open in a new tab

Ten Topics obtained using topic modeling.

Then, using the BOW method, a vector was created for the thousand most frequent words of each topic, and the correlation between them was determined (Fig. 4 ). As shown, biology best correlates with medicine, politics, and humanities, with a correlation of 0.66, and economics best correlates with management, with a correlation of 0.62.

Fig. 4.

Fig. 4

Open in a new tab

Correlation of topics.

As a next step, a graph was formed from the topics, with each topic's edge having a correlation of at least 45%. Fig. 5 shows the graph. It was determined by trial and error with a threshold of 45%. This graph determines important hubs for identifying more important topics and communications. Humanities, politics, and management topics are most connected with five edges, whereas biology and religion are least connected with one.

Fig. 5.

Fig. 5

Open in a new tab

The graph of topics.

The topics were found in ten different clusters (Fig. 6 ). The following parts will describe the topics, followed by the content analysis of the articles related to each topic.

Fig. 6.

Fig. 6

Open in a new tab

The topics identified by combining the LDA + ParsBERT algorithms.

Table 2 shows the frequency of repetition of the top five words in each topic.

Table 2.

The top five words of clusters.

Topic Topics Top 5 words translated into English
1 Medicine Respiratory, Infection, Nurses, Dentistry, Life
2 Psychology Anxiety, Stress, Excitement, Lived Experience, Psyche
3 Humanities Family, Rights, Responsibility, Life, Teaching
4 Politics Politics, Diplomacy, America, Democracy, Future
5 Management Industry, Tourism, Decision, Career, Capital
6 Biology Evolution, Protein, Immunity, Genetics, Security
7 Economics Price, Index, Hotel, Urban, Rural, Budget
8 Culture Media, Social, Rumor, Danger, Coronaphobia
9 Engineering Consumption, Electricity, Job Burnout, Industry, Masks
10 Religion Islam, Religion, Compassion, Death, Middle East
Open in a new tab

2.2. Topic 1. Medicine

The largest cluster, with 304 records, is related to the medical cluster, which includes diseases such as diabetes, infection, pneumonia, medical diagnosis, and control and treatment methods. This cluster also includes nurses at the forefront of diagnosing, treating, and caring for patients with the COVID-19 disease, who run a high risk of contracting the illness and exhibiting mental health symptoms, such as stress and anxiety, which causes occupational depression for this class. More research should be done in light of the massive wave of contamination, the spread and rapidity of this virus, the proportion of deaths brought on by this illness globally, and the development of medicine and vaccines. The symptoms of this disease have been reported in all body organs, such as the eyes, nose, brain, lungs, heart, digestive system, and kidney [17]. With the limitations of control and treatment methods, the most effective solution is to prevent the contagion of the virus.

2.3. Topic 2. Psychology

The second topic is psychology. In the previous cluster (medicine), COVID-19 affected people's physical health. However, this cluster demonstrates the damaging effects of COVID-19 on psychological well-being, impeding recovery and worsening patients' suffering while ill. The increased prevalence and mortality of COVID-19 disease and the lack of definitive treatment have also caused a massive wave of anxiety worldwide and decreased the quality of life [18,19].

Anxiety is a major symptom of COVID-19. Previous research has shown that anxiety and stress are the most common psychological symptoms of this disease, which is confirmed by the frequency of keywords in this research. This research's crucial components with high frequencies are resilience, social support, and maintaining coherence. In addition, previous research showed that these components play significant roles in the level of anxiety experienced by COVID-19 [20]. Social support, which relatives can provide, is one of the critical resources in reducing and controlling COVID-19-induced. Social support includes affection, attention, care, and assistance from individuals or groups such as family, friends, and others [21]. Resilience [22,23] and a sense of consistency [24,25] significantly reduce anxiety risk. People with high resilience have more adaptive behaviors in problem-solving situations and are more prepared in the face of adverse situations. The resilience mechanism reduces anxiety in such a way that its main components, such as self-confidence, personal competence, trust in instincts, positive acceptance of change, control, and spiritual effects, act as a shield under stressful situations. Resilience is also seen as the ability to maintain the biological-psychological balance in dangerous conditions and stability against injuries [26]. This may justify the interest in understanding the lived experiences of the people involved and their perceptions and connections between relationships and situations.

This anxiety engages all people, from children to older adults, but fear and anxiety are seen far more in older adults. The researchers know the underlying diseases as the reason for the weaker immune system and higher mortality in this group [18,27].

There is a relationship between the cognitive regulation of excitement and anxiety and stress in older adults [28,29]. Therefore, promoting the positive strategies of cognitive regulation of emotion and reducing the negative strategies of cognitive regulation of emotion by increasing distress tolerance will decrease the anxiety of COVID-19 in older adults.

2.4. Topic 3: Humanities

COVID-19 has highlighted the vitality of human and social behavior in controlling epidemics. The lack of effective treatment and vaccines means the world has relied on changing behaviors to prevent virus transmission. The behaviors involve individuals, groups, and societies operating in complex systems in different physical and social environments. Large-scale behavioral change requires behavioral, environmental, social, and systemic interventions to suppress epidemics. Social sciences and humanities noticeably contribute to solving the enormous challenges of the COVID-19 pandemic [30], which drastically changed social and educational interactions and disrupted education in many ways. Virtual teaching has imposed unknown problems and challenges on teachers, families, students, and educational centers [31].

China was the first country to shut down schools to limit the spread of this virus [29]. The closure of schools and other educational facilities then impacted 94% of other areas [32]. Their solution was to benefit from virtual education, which had advantages and disadvantages, both for the educational system and families. Virtual education helped increase the motivation for learning and students' enthusiasm [33,34].

Family is one of the most frequent and essential keywords in this cluster, with a significant role in making peace and teaching social skills. Research shows that the family positively affects teaching children social skills and interactions. In addition, as social peace has been affected by the virus in many countries, the family should also pay attention to it.

2.5. Topic 4: Politics

COVID-19 has led to unprecedented changes in society [35]. The disease was reported in almost all countries and regions, making the world adapt to a new definition of a normal situation. The world realized the necessity of quickly adapting to new ways of living and working. Many countries decided to close schools, colleges, and universities. This crisis creates a dilemma for policymakers between closing schools to reduce contact and save more lives and keeping them open, thus allowing workers to work and preserving the economy.

Intellectuals of international relations usually use the term new order for the conditions that emerge after a new full-fledged war [36]. The effects of the COVID-19 epidemic on society are comparable to those of a major war. The epidemic has changed many laws, regulations, customs, habits, lifestyles, occupations, and structures. In the first stage, the COVID-19 crisis is a health crisis that threatens the lives and well-being of people worldwide. However, the current crisis is a major source of economic and political turmoil and concerns over regional security and threatens people's health and lives.

The diplomatic world is mainly affected by these variations due to the cancellation of meetings, conferences, and other significant events. New variants of the coronavirus have intensified the requirement for greater coordination across geographic and political boundaries and more vital collaboration across science, policymaking, and society. Using global health diplomacy, vaccine diplomacy, and science diplomacy provides an opportunity to review the global health dynamics to promote development, health security, justice, and health equity.

The novel field of health policy and systems research (HPSR) can help apply the modernist approach. HPSR adoption not only stimulates new thinking toward a transformative renaissance in health systems but also strengthens the synergies between policymakers, scientists, and communities to establish fundamental reforms in the existing sections, disciplines, and frames, especially in fragile countries with few resources. The application and performance of Global Health Diplomacy (GHD) have become very critical in this context. It enables several stakeholders to contribute to humanity's more extensive health needs, strengthen more potent interdisciplinary approaches, and promote negotiations created and managed by the global policy environment for health.

America, Europe, the West, and Union are some of the crucial keywords in this cluster. They demonstrate the global health community's unity in the fight against this disease. In times of crisis, international cooperation is more than imperative. Diplomacy is adapted based on necessity. The World Bank and the International Monetary Fund (IMF) will hold their spring meetings online for the first time. The United Nations, the European Union, and other organizations have moved from conference rooms to online spaces, which has caused a profound change in how diplomacy is applied. Digital diplomacy has been demonstrated to be successful in the world of diplomacy. It has helped governments and officials operate more efficiently in terms of time and resources while maintaining relationships.

2.6. Topic 5: Management

COVID-19 has imposed multiple challenges on most organizations and institutions with its rapid spread and global epidemic. Healthcare organizations faced many strategic and management challenges during the COVID-19 pandemic. The management of these indispensable organizations requires decision-making and planning to overcome, control, and restrain this crisis. Every manager must access correct and relevant information and their correct analysis for making correct and on-time decisions. Therefore, in the face of problems, managers may fail to deal with the massive amount of information and make correct decisions due to information overload. They need some tools to facilitate decision-making and problem-solving. The keywords, including clustering, text analysis, intelligence, text mining, phenomenology, sentiment analysis, social network analysis, neural network, meta-synthesis, and decision-making, are tools for managers to find knowledge, make decisions and improve leadership, which can be seen in this cluster.

Social capital is another critical topic in this cluster. Social capital results from the impact of social institutions, human relations, and norms on the quality and quantity of social interactions, and it significantly impacts the economy, industry, and development of different countries. Organizational promotion and empowerment are directly and indirectly related to social capital [37]. Nursing is one of the most vital occupations affected by social capital because nurses have direct contact with patients and clients [33,38].

Also, social capital is one of the critical and influential factors in occupational burnout [33,38] and organization-induced work stress [39]. Social capital can improve nurse performance by reducing occupational burnout and stress, hence the need to reinforce different dimensions of social capital among nurses. There is also evidence that sports [40] and tourism [41] directly relate to social capital. During the pandemic, governments took several measures to promote domestic tourism, such as removing travel restrictions, following the protocol in sports clubs, and restoring the trust of travelers.

2.7. Topic 6. The Biology

Biologists and virologists have raised concerns about COVID-19 complications. The identification of the virus's genome sequence has progressed significantly compared to the last efforts, but the disease's unpredictable behavior and constant mutations have made it difficult for biologists and virologists to recognize this virus. Biologists are trying to recognize this complex virus by investigating the biological basis of the new virus and the resulting evolutionary development of COVID-19. The global sequencing of the genomes of severe acute respiratory syndrome COVID-19 (SARS-CoV-2) has continued to discover new genetic variants, which are the key to finding its primary evolutionary history and tracking its global spread over time. The progenitor mutations and their offshoots have produced many dominant strains of COVID-19 that have spread episodically over time.

There are still different viewpoints about the origin and evolution of this virus. Biologists agree that SARS and SARS-CoV-2 have many similarities, but their characteristics differ. The differences in infection period, transmission, spread speed, and clinical features are more visible. There are also 380 different amino acids between SARS-CoV-2 and other coronaviruses, including the SARS-causing virus and SARS-like strains, which indicates the divergence feature of SARS-CoV-2 in comparison with other coronaviruses. Because of this, the coronavirus that causes COVID-19 is also referred to as SARS-CoV2 in many academic sources. Coronaviruses are classified into four groups based on their ability to infect respiratory and digestive systems and their genetic structure: alpha coronaviruses, beta coronaviruses, gamma coronaviruses, and delta coronaviruses [42]. Researchers have made significant progress in recognizing and describing the emerging SARS-CoV2 and working extensively on treatment methods and antiviral vaccines. Identifying this virus's severe reproduction rate and structure is significant for discovering a way for specific treatment or prevention due to its high transmission power and epidemic feature.

2.8. Topic 7: Economics

Besides threatening health, social, and sanitary sectors, COVID-19 has also become a threat to global economics. The COVID-19 pandemic has led to unprecedented changes in society and the economy. This crisis has changed the way people work in their daily lives. These changes, including those on security, oil, and tourism, have not equally affected all social groups. This virus has affected all groups, regardless of wealth, urbanity, or rurality [43]. Taking social distancing measures to control the virus, such as working from home and closing schools, has put an incredible burden on families [44].

Tourism is one of the things that have impacted the economy during the COVID-19 era. Medical tourism is a dimension of tourism that contributes to the sustainable development and dynamism of the country's economy. One of the long-term quarantine consequences during the COVID-19 era is the financial shock and failure that plagued most people in society [45]. According to the report of the International Monetary Fund and the World Bank, in 2021, most economies in different geographical regions will have a harmful gross domestic product (GDP) or a deficient gross domestic product (GDP) in the best situation, and most businesses and companies will go bankrupt or struggle with serious challenges. Millions of jobs will disappear, and more families will be exposed to poverty and insecurity worldwide. Crime, violence, and insecurity will be the direct consequences of the pandemic [36]. Other consequences can include decreased growth rate of the global economy, unemployment (case study: Tabriz city), declined oil demand and a sharp drop in the oil price, closure of production centers and labor force reduction, and decline in the stock, oil, and coin markets. These consequences severely impact individuals and society. Our results also highlighted that the greatest impact is related to economic factors.

2.9. Topic 8: Culture

A crisis is a form of a silent war, one that is not like other wars but can lead to total destruction like an actual war. Rather than annihilating bodies, the destruction targets minds and morale. A that is widespread is a risk, and a risky society faces new problems that can affect its lifestyle and social behaviors. In widespread crisis conditions, a risky society determines its distance from what it was before through fundamental changes. COVID-19 has led to a severe international crisis and a risky global society. Its consequences are reflected in its indisputable impacts on media and culture.

The results demonstrate the variety and multiplicity of communication perspectives in this area. The methods of informing, news management, trust building, social and mental health, rumors and fake news, corona phobia, and the significance of media literacy are just a few of the broad themes that have attracted attention from media, communication, and cultural points of view.

The impacts of COVID-19 on culture can be addressed from three angles: The first perspective is that of conformists, who think that society's culture should not be altered. The second point of view is that of nonconformists, who believe that society's culture will face structural changes. Critical rethinking holds that cultural policies and shared beliefs have encountered difficulties and will no longer be the same in the post-corona era. However, evidence shows that the media play a significant role in promoting health [46]. The global information epidemic and Infodemia have turned social media, television networks, and virtual media into tools for improving control and prevention methods, reducing COVID-19 complications, and spreading authentic ct information internationally. Numerous actions have been taken since the outbreak of the epidemic, including raising awareness among the family, the general public, businesses and trades, workers, and employees. In addition, non-face-to-face training, audio and television, educational media, and other methods have been used. As a communication channel between governments, health institutions, and people, the media provides a window for accurate information, sound scientific facts, government decisions, and public reactions [46].

2.10. Topic 9: Engineering

Engineers can play a significant role in solving the many dimensions of the COVID-19 crisis. Their proceedings include producing related medical equipment, food products, pharmaceutical raw materials, anti-virus and industrial and domestic disinfectants, and vaccine production.

This virus has also affected the petrochemical and polymer industries. After the pandemic, the petrochemical industry increased the production of some petrochemical and polymer products to solve the COVID-19 crisis. One of the most important measures is coordinating to increase the production of some key products to tackle the spread of COVID-19.

The chemical engineering community faces various problems as this disease spreads rapidly. There are a variety of different issues that need to be addressed, including expediting the production of essential medical equipment, food products, antiviral raw materials (API), disinfectants, and health products, as well as participating in research, developing, and increasing the scale of vaccine production and successful drugs in order to obtain the necessary permits. Chemical engineers have multiple options in dealing with COVID-19 challenges: using quantum processes and pattern-building in the design of the new molecules to make specific COVID-19 medications and achieve a better understanding of epidemic patterns.

Other frequently used terms in this cluster include artificial intelligence and information technology. The speed of transmission and spread of the coronavirus is so great globally that it requires the employment of new technologies of the fourth industrial revolution related to artificial intelligence.

The record speed of developing a vaccine was primarily thanks to advances in artificial intelligence that helped researchers analyze large amounts of data about the coronavirus. Using artificial intelligence can increase the speed of detection. Artificial intelligence and machine learning have achieved great success in medical imaging and diagnosis, as well as the development of new treatment approaches and understanding of development patterns due to their high extraction capabilities. In addition to all these cases, artificial intelligence can prevent the further spread of this disease by reducing the contact of medical staff with patients and preventing the continuation of this sequence. This issue is one of the reasons for the international interest in using automated programs to perform diagnostic and therapeutic tests. Currently, artificial intelligence speeds up the testing process by simulating the environment and variables.

2.11. Topic 10: Religion

Religion and religiosity are also affected by COVID-19 disease, so we cannot ignore its effects and consequences on religion and its place in society. The relationship between COVID-19 and religion can be discussed from two perspectives: how COVID-19 affects religion and how religion affects COVID-19.

COVID-19 has had a significant impact on society's religious field. The stress induced by COVID-19 has influenced people's religious beliefs [47]. The prohibition of religious gatherings and the closure of holy places and rituals worldwide, including churches, mosques, blessed monuments, Umrah, and other options, helped restrain the spread of this disease. Several researchers have investigated the effects of religion and spirituality on relieving pain and speeding up treatment. According to empirical evidence, religion often helps believers cope with stress, particularly by regulating their emotional state [48,49]. Religion may effectively manage stress as it can explain stressful events as a source of hope and a sense of belonging to a religious community.

The findings of research regarding the COVID-19 epidemic conducted in Malaysia's main religious groups (Christians, Muslims, and Buddhists) have shown that religious expression, disease belief, and stress regulation during epidemic quarantine are verifiable in all three groups [50].

Religion is significant in various religions and rituals, including Protestantism [51] and Christianity [47], as well as Islam, Catholicism, and Buddhism [52]. Freud holds that religion has a negative effect on health. The researchers have discussed the relationship between COVID-19 and apocalyptic discussions and describe COVID-19 as the white death and an apocalyptic sign. Authors in Ref. [53] call COVID-19 a "Minor Death" because of its depriving and informing role. The priority and significance of saving people's lives from death have taken precedence over religious beliefs and practices and caused the closure of sacred rituals and places.

The Middle East keyword has the highest frequency in this cluster. Zoroastrianism, Judaism, Christ, and Islam all emerged in the Middle East, which is also where the three continents of Africa, Asia, and Europe meet. The Middle East is the cradle of civilization and has seen the emergence of all four major religions. Its geographical situation has made it possible to exchange different opinions [54].

Sympathy or compassion is one of the most frequent keywords in this cluster, which is one of the most critical topics in various religions and rituals. Different religions have recommended sympathy with different expressions [55]. Sympathy is the third place among the four spiritual authorities in Buddhism that leads to salvation since the essence of Buddha's instructions is according to natural law (dharma) or moral order. According to the German philosopher Arthur Schopenhauer, sympathy has a Buddhist flavor and considers the same metaphysical unity as the basis of the sympathy principle.

Schopenhauer points to the deep relationship of COVID-19 between philosophy and Buddhism [56]. From the point of view of Islam, compassion and empathy with others are serious commands in political ethics, and their absence is thought of as carelessness and irresponsibility towards difficulties. Therefore, observing it reduces problems at all times, especially during the COVID-19 crisis.

The research also shows that higher spiritual intelligence and compassion have performed better in those suffering from COVID-19 [57].

Finally, we decreased the dimensions and converted the vectors from the 768-dimensional space to the two-dimensional space for building a two-dimensional vector and an illustration of the vector produced by ParsBERT using the SOM algorithm [14] (Fig. 7 ).

Fig. 7.

Fig. 7

Open in a new tab

Conversion to two-dimensional space using the SOM algorithm.

3. Discussion and conclusions

In this study, the scientific outputs concerning COVID-19 from the Magiran database were topic modeled using the LDA and ParsBERT algorithms. In the first step, we applied the required pre-processing to the data set by writing the code after data extraction. In the next step, we converted the keywords into vectors using ParsBERT and categorized the words using the LDA probabilistic method as unsupervised learning. Finally, we transformed the 768-dimensional vectors into two dimensions and displayed them using SOM. In addition, there were limitations in this research. The data had to be extracted by coding, which was both complicated and challenging. Since there were fewer than 1000 articles, applying topic modeling was difficult. Moreover, there are few tools and technologies available to analyze Persian texts. In addition, topic modeling can be applied using other algorithms on the data set of this research for future research.

We can conclude that COVID-19 is a complex multidimensional reality with a multifaceted impact on our life in three main dimensions: individual, family, and society. The results show that the medicine and psychology clusters are more prominent than others. An individual's physical and mental health impacts the health of their family and society. Considering the importance of health, these two issues responsible for physical and mental health are the foundation and root of all clusters. Health is a single and inseparable whole and consists of complex physical, mental, spiritual, and social dimensions. Any change in each of these dimensions leads to its frequency and disease. Furthermore, this role has become more prominent with the emergence of the coronavirus. Psychology forms the next cluster, which indicates the mental aspect of the damage associated with the disease in addition to its physiological implications. Health and illness are fundamentally influenced by people's attitudes toward life, the world, others, and death, as well as the behaviors that result from these attitudes. Discussing the psychological effects of this viral disease on people's mental health at different levels of society is critical. Accordingly, psychology is a separate cluster and is not combined with the medical cluster.

COVID-19 also impacts families. The family is the door to society, and the foundation of many instructions, skills, and social relationships lies within the family. The pandemic aggravated unemployment, resulting in financial and psychological problems and miscommunication among family members. It can also raise divorce rates and lead to family-related conflicts and violence, hostility between spouses, and marital problems. The COVID-19 crisis has rapidly evolved from a medical and psychological issue to a societal issue that is also biological, political, economic, and cultural. The preventive and control measures of the coronavirus crisis, which require a comprehensive, systematic, and multidimensional approach and knowledge of its origin, have led to the emergence of biological clusters. Politics forms another cluster in this study. Communication at the local and global levels has exceeded time and place in the contemporary era, so the interactions and communications shattered the boundaries, mainly involving governments and policymakers in controlling and managing.

The COVID-19 pandemic has caused unprecedented economic changes, altering how people work, and influencing different sectors, from the oil industry to tourism. The influence sphere of media highlights the significance of the culture cluster. Media channels are sources of information about governments, health institutions, and the media, contributing to the formation of scientific facts, decisions of governments, and public reactions [46]. Therefore, culture can be seen as a bridge connecting the individual with society, which can act as a double-edged sword: it can complement the administrative policies of crisis managers or heighten the level of executive involvement in a crisis and deepen the disaster.

Considering the issues raised in the clusters above, the existence of the management cluster demonstrates how management, strategy, and correct decision-making are crucial to overcoming, controlling, and restraining this crisis. Religion could be considered a subset of humanities. Religion and spirituality's effects on healing, reducing patient pain, and speeding up the treatment process have been placed in separate clusters. Empirical evidence shows that religion can also successfully deal with and manage stress. Additionally, the correlation between topics confirms the connection between topics, including biology with medicine, politics with humanities, and economics with management, thus proving the validity of our research. The COVID-19 pandemic has affected almost all vital dimensions. The clusters obtained in this research also confirm this virus's global spread and pandemic. Policymakers and scientists must work together to control and reduce the effects of this crisis. For future research, other algorithms can be applied to topic modeling using the data set of this research.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

N/A.

Abbreviations:

LDA

Latent Dirichlet Allocation

BOW

Bag of Word

HPSR

health policy and systems research

GHD

Global Health Diplomacy

IMF

International Monetary Fund

GDP

harmful gross domestic product

NLP

natural language processing

Multimedia Appendix 1.

The articles are available in the Magiran scientific database at the address of https://www.magiran.com/, but it is not possible to save and download the data. The saved dataset can be downloaded at:

https://github.com/mohamad-dehghani/CoPershttps://github.com/mohamad-dehghani/CoPers.

References

  • 1.Ahmad A.R., Murad H.R. The impact of social media on panic during the COVID-19 pandemic in Iraqi Kurdistan: online questionnaire study. J Med Internet Res. 2020;22(5) doi: 10.2196/19556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Abdulmajid A.H. A Survey of authored library and information articles published in Farsi journals of Humanities and Social Sciences. J Libr Stud Inf Organ. 2008;19(3):55–70. [Google Scholar]
  • 3.Albalawi R., Yeap T.H., Benyoucef M. Using topic modeling methods for short-text data: a comparative analysis. Front.Artif Intell. 2020;3 doi: 10.3389/frai.2020.00042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Maier D., Waldherr A., Miltner P., et al. Applying LDA topic modeling in communication research: toward a valid and reliable methodology. Commun Methods Meas. 2018;12(2–3):93–118. doi: 10.1080/19312458.2018.1430754. [DOI] [Google Scholar]
  • 5.Chen D., Zhang R., Zhao H., Feng J. A bibliometric analysis of the development of ICD-11 in medical informatics. J Healthc Eng. 2019;2019:1–12. doi: 10.1155/2019/1649363. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Danesh F., Dastani M., Ghorbani M. Retrospective and prospective approaches of coronavirus publications in the last half-century: a Latent Dirichlet allocation analysis. Libr Hi Technol. 2021;39(3):855–872. doi: 10.1108/lht-09-2020-0216. [DOI] [Google Scholar]
  • 7.Danesh F., Ghavidel S. Coronavirus: scientometrics of 50 Years of global scientific productions. Iran J Med Microbiol. 2020;14(1):1–16. doi: 10.30699/ijmm.14.1.1. [DOI] [Google Scholar]
  • 8.Makkizadeh F., Bigdeloo E. Drawing and analyzing the conceptual network of scientific products in the field of covid-19. HSR. 2021;17(2):142–149. [in Persian] [Google Scholar]
  • 9.Joshi C., Attar V.Z., Kalamkar S.P. Advances in data and information sciences. Springer Singapore; 2022. An unsupervised topic modeling approach for adverse drug reaction extraction and identification from natural language text; pp. 505–514. [DOI] [Google Scholar]
  • 10.Vijayan R. Teaching and learning during the COVID-19 pandemic: a topic modeling study. Educ Sci. 2021;11(7):347. doi: 10.3390/educsci11070347. [DOI] [Google Scholar]
  • 11.Gupta A., Aeron S., Agrawal A., Gupta H. Trends in COVID-19 publications: streamlining research using NLP and LDA. Frontiers in Digital Health. 2021;3 doi: 10.3389/fdgth.2021.686720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kim H., Song* J. Analysis of articles related to quantum mechanics by using topic modeling: focus on comparing LDA, CTM, STM models. New Physics: Sae Mulli. 2020;70(5):493–502. doi: 10.3938/npsm.70.493. [DOI] [Google Scholar]
  • 13.Älgå A., Eriksson O., Nordberg M. Analysis of scientific publications during the early phase of the COVID-19 pandemic: topic modeling study. J Med Internet Res. 2020;22(11) doi: 10.2196/21559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kohonen T. Self-organized formation of topologically correct feature maps. Biol Cybern. 1982;43(1):59–69. doi: 10.1007/bf00337288. [DOI] [Google Scholar]
  • 15.Tache A., Mihaela G., Ionescu R.T. Proceedings of the 16th conference of the European chapter of the association for computational linguistics: main volume. Association for Computational Linguistics; 2021. Clustering word embeddings with self-organizing maps. Application on LaRoSeDa - a large Romanian sentiment data set. [DOI] [Google Scholar]
  • 16.Emami M. Islamic Azad University; Ghods branch: 2013. Self-organizing neural networks. Bachelor thesis (B.S.) of Computer major. Faculty of Engineering. [In Persian] [Google Scholar]
  • 17.Begun J.W., Jiang H.J. 2020. Characteristics of the covid-19 pandemic health care management during covid-19: insights from complexity science. Innovation in care delivery. Published online October. [DOI] [Google Scholar]
  • 18.Thyrian J.R., Kracht F., Nikelski A., et al. The situation of elderly with cognitive impairment living at home during lockdown in the Corona-pandemic in Germany. BMC Geriatr. 2020;20(1) doi: 10.1186/s12877-020-01957-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Fardin M.A. COVID-19 and anxiety: a review of psychological impacts of infectious disease outbreaks. Arch Clin Infect Dis. 2020:15. [In Persian] [Google Scholar]
  • 20.Eyni S., Ebadi M., Torabi N. Developing a model of corona anxiety in students based on optimism and resilience: the mediating role of the perceived social support. EBNESINA. 2020;22(1):17–28. [In Persian] [Google Scholar]
  • 21.Chadwick K.A., Collins P.A. Examining the relationship between social support availability, urban center size, and self-perceived mental health of recent immigrants to Canada: a mixed-methods analysis. Soc Sci Med. 2015;128:220–230. doi: 10.1016/j.socscimed.2015.01.036. [DOI] [PubMed] [Google Scholar]
  • 22.Poudel-Tandukar K., Chandler G.E., Jacelon C.S., Gautam B., Bertone-Johnson E.R., Hollon S.D. Resilience and anxiety or depression among resettled Bhutanese adults in the United States. Int J Soc Psychiatr. 2019;65(6):496–506. doi: 10.1177/0020764019862312. [DOI] [PubMed] [Google Scholar]
  • 23.Wood B. Role of resilience in buffering the effect of work-school conflict on negative emotional responses and sleep health of college students. J Occup Health Psychol. 2019;18(4):384–394. [Google Scholar]
  • 24.Hutchinson A.-M.K., Stuart A.D., Pretorius H.G. The human pursuit of well-being. Springer Netherlands; 2011. The relationships between temperament, character strengths, and resilience; pp. 133–144. [Google Scholar]
  • 25.Antonovsky A. Intergenerational networks and transmitting the sense of coherence. Life-span.Dev Psychol Intergener Relat. 2014;211 [Google Scholar]
  • 26.Connor K.M., Davidson J.R.T. Development of a new resilience scale: the connor-davidson resilience scale (CD-RISC) Depress Anxiety. 2003;18(2):76–82. doi: 10.1002/da.10113. [DOI] [PubMed] [Google Scholar]
  • 27.Meng H., Xu Y., Dai J., Zhang Y., Liu B., Yang H. The psychological effect of COVID-19 on the elderly in China. Psychiatr Res. 2020;289:1–3. doi: 10.1016/j.psychres.2020.112983. 10.1016%2Fj.psychre s.2020.112983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mertens G., Gerritsen L., Duijndam S., Salemink E., Engelhard I.M. Fear of the coronavirus (COVID-19): predictors in an online study conducted in March 2020. J Anxiety Disord. 2020;74 doi: 10.1016/j.janxdis.2020.102258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Zhang P., Wang M., Wang Y., Wang Y., Li T., Zeng J., Gong Y. Risk factors associated with the progression of COVID-19 in elderly. Diabetes Res Clin Pract. 2021;171(2):1–10. doi: 10.1016/j.diabres.2020.108550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Michie S., West R. Behavioural, environmental, social, and systems interventions against covid-19. BMJ. 2020 Jul 28;370:m2982. doi: 10.1136/bmj.m2982. PMID: 32723732. [DOI] [PubMed] [Google Scholar]
  • 31.Maggio L.A., Daley B.J., Pratt D.D., Torre D.M. Honoring thyself in the transition to onlinee teaching. Academic Medicline. 2018;93(8):1129–1134. doi: 10.1097/ACM.0000000000002285. [DOI] [PubMed] [Google Scholar]
  • 32.United Nations . 2020. Policy brief:education during COVID-19 and beyond. [Google Scholar]
  • 33.Ashoori J., Kajbaf M., Manshaee G., Talebi H. Comparison of the effectiveness of web-based, cooperative learning and traditional teaching methods in achievement motivation and academic achievement in the biology course. ijvlms. 2014;5(2):25–34. [In Persian] [Google Scholar]
  • 34.Mosayebi Ardakani M., Rezapour Mirsaleh Y., Behjati Ardakani F. The problems and challenges of virtual education in elementary school during the outbreak of coronavirus. J Educ Stud. 2021;7(27):65–79. [In Persian] [Google Scholar]
  • 35.Coughlin C.G., Sandel M., Stewart A.M. Homelessness, children, and COVID-19: a looming crisis. Pediatrics. 2020;146(2) doi: 10.1542/peds.2020-1408. [DOI] [PubMed] [Google Scholar]
  • 36.Institute for Political and International Studies- Impact of Coronavirus Pandemic on International Relations 2021. https://ipis.ir/en/subjectview/640958/Impact-of-Coronavirus-Pandemic-on-International-Relations Retrieved June 4, 2022, from.
  • 37.Malloy D.C., Hadjistavropoulos T., McCarthy E.F., et al. Culture and organizational climate: nurses' insights into their relationship with physicians. Nurs Ethics. 2009;16(6):719–733. doi: 10.1177/0969733009342636. [DOI] [PubMed] [Google Scholar]
  • 38.Piri Y., Zeinali A. Relationship between perceived social support, social capital and quality of life with job burnout among nurses. Ijn. 2016;29(103):13–23. [In Persian] [Google Scholar]
  • 39.Sadeghi M. Investigating the relationship between social capital and stress caused by organizational factors with job satisfaction of employees of Golestan Medical Sciences Organization. Public policy in management. 2017;8:157–180. [In Persian] [Google Scholar]
  • 40.Mastromartino B., Wang J.J., Suggs D.W., Hollenbeck C.R., Zhang J.J. Dimensions of sense of membership in a sport fan community: factors, outcomes, and social capital implications. Communication & Sport. 2020;11 doi: 10.1177/2167479520956370. Published online September. [DOI] [Google Scholar]
  • 41.Zhao W., Ritchie J.R.B., Echtner C.M. Social capital and tourism entrepreneurship. Ann Tourism Res. 2011;38(4):1570–1593. doi: 10.1016/j.annals.2011.02.006. [DOI] [Google Scholar]
  • 42.Zeinoddini M. Reasons for the creation of the new coronavirus 2019 (SARS-CoV2): natural mutation or genetically laboratory manipulation-point of view. JRUMS. 2020;19(7):749–764. [In Persian] [Google Scholar]
  • 43.Mitra S. The implications of COVID -19 for rural India. IDR. 2020. March;25 https://idronline.org/the -implications -of-covid -19 -for -rural -india/ [Google Scholar]
  • 44.Del Boca D., Oggero N., Profeta P., Rossi M. Women's and men's work, housework and childcare, before and during COVID-19. Rev Econ Househ. 2020;18(4):1001–1017. doi: 10.1007/s11150-020-09502-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Brooks S.K., Webster R.K., Smith L.E., et al. The psychological impact of quarantine and how to reduce it: rapid review of the evidence. Lancet. 2020;395(10227):912–920. doi: 10.1016/s0140-6736(20)30460-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Thomas S.L., Bestman A., Pitt H., et al. Young people's awareness of the timing and placement of gambling advertising on traditional and social media platforms: a study of 11–16-year-olds in Australia. Harm Reduct J. 2018;15(1) doi: 10.1186/s12954-018-0254-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Rigoli F. The link between COVID-19, anxiety, and religious beliefs in the United States and the United Kingdom. J Relig Health. 2021;60(4):2196–2208. doi: 10.1007/s10943-021-01296-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Pargament K.I. Guilford Press; 2001. The psychology of religion and coping: theory, research, practice. [Google Scholar]
  • 49.Pargament K.I., Raiya H.A. A decade of research on the psychology of religion and coping: things we assumed and lessons we learned. Psyke Logos. 2007;28(2):25. 25. [Google Scholar]
  • 50.Ting R.S.-K., Aw Yong Y.-Y., Tan M.-M., Yap C.-K. Cultural responses to covid-19 pandemic: religions, illness perception, and perceived stress. Front Psychol. 2021;12 doi: 10.3389/fpsyg.2021.634863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Kim M. The relationship between religion and society in the COVID-19 era: the case of protestantism in South Korea. Religions. 2021;12(12):1102. doi: 10.3390/rel12121102. [DOI] [Google Scholar]
  • 52.Park C., Kim K. Covid-19 and Korean Buddhism: assessing the impact of South Korea's coronavirus epidemic on the future of its Buddhist community. Religions. 2021;12(3):147. doi: 10.3390/rel12030147. [DOI] [Google Scholar]
  • 53.Navali M. The coronavirus as “minor death. J Phil Invest. 2020;14(31):163–168. doi: 10.22034/jpiut.2020.10996. [In Persian] [DOI] [Google Scholar]
  • 54.Sajedi A. The roots and causes to the growth of terrorism in the Middle East. J Polit Stud. 2016;9(33):139–160. [In Persian] [Google Scholar]
  • 55.Shojaaii A. The foundations of Buddhist ethics and the western models. Ethics. 2014;4(1):119–144. [In Persian] [Google Scholar]
  • 56.Kokabi Delavar F., et al. Sympathy in Buddhism and schopenhauer philosophy. Knowledge of Religions. 2014;5(3):137–153. [In Persian] [Google Scholar]
  • 57.Bagheri L., Safouraei S., Safouraei M. The resilience prediction of patients with corona disease based on life expectancy, religious coping style, spiritual intelligence and self-compassion. Quran Med. 2021;5(4):1–10. [In Persian] [Google Scholar]

Articles from Informatics in Medicine Unlocked are provided here courtesy of Elsevier

RESOURCES