Text Mining Approach for Trend Tracking in Scientific Research: A Case Study on Forest Fire

Scientific studies are increasing day by day with the development of technology. Today, more than 171 billion academic records are made available to researchers via the Web of Science database, which is frequently followed by the scientific community, and is where records of articles, proceedings, and books in many different fields are kept. More than 40 thousand studies are reached when a search is made for research on forest fires in the relevant database. It is unfeasible to examine and read so many publications and understand what topics are important in the relevant field, what is trending, or whether there is a difference between the subjects studied based on years and/or regions/countries. The most effective and scientific method of deriving information from such large and unstructured data is text mining. In this study, text mining is used to reveal where the research on forest fires in the Web of Science database concentrates, which study topics have emerged, how an issue's level of importance changes over the years, and which topics different countries focus on. Therefore, the s of approximately 32 thousand articles published in English were collected and analyzed based on the country of the authors and the published years. Over 600 words in the s were indexed for each article and their importance was calculated according to inverse document frequency. A size reduction was made to determine the main concepts of the articles by using the singular value decomposition and a total of 29 different concepts were found. Among these, important concepts can be mentioned such as damage to vegetation and species affected, post-fire actions, fire management, and post-fire structural changes. Considering all the articles, studies on soil, fuel (biofuel), treatment, emissions, and species were found to be important. The results we have obtained in this study are by no means a summary of the research carried out in the field;they do, however, allow statistical due diligence concerning, for example, which subjects are important in the relevant field, the determination of increasing and decreasing trending topics, which countries attach importance to in the same research, and so on. Thus, it will function as be a guide in terms of the direction, timing, and budget allocation of research plans in a specific area in the future.

Does fleet standardization matter on profitability and financial policy response of airlines during COVID-19 pandemics in the U.S?

This study tries to find the contribution of the fleet standardization index of airline companies through the financial values of airlines in the COVID-19 pandemic crisis. One of the driving forces of this research is the stagnation in the aviation industry, along with the cessation of all international flights to prevent worse pandemic conditions. The main contribution of the study is clarifying the effect of fleet variety and its financial inferences about fleet standardization and crisis management relationship under inoperative service considerations. Data Envelopment Analysis (DEA) is used as a complementary method for standardization index calculations. According to the results, while the positive contribution of the fleet standardization index to operational revenues was observed, there was no sign of the effects of fleet standardization on the accounting performance of companies. Besides, fleet standardization enabled us to measure the response of Low-Cost Carrier (LCC) and Full-Service Carrier (FSC) airlines in crises. In this study, nine airline companies were selected from both trend representatives, and the results were presented for the U.S. It is found that fleet standardization is related to the adopted flight network type and profitability of each company directly. Results obtained revealed the relationship between fleet standardization and financial relations under inoperative conditions positively. In addition, inferences were made about the effects of the operational policies adopted by the companies.

YAPAY SÍNÍR AĞLARI VE DERÍN ÖĞRENME KULLANARAK TÜRKÍYE''DEKÍ COVID-19 VAKALARI ÍÇÍN TAHMÍN MODELLERÍ

Hükumetler, bir pandemi salgını sırasında stratejik kararlar alırken, halk sağlığı ve ekonomi arasında bir ikilemle karşı karşıyadır. Özellikle salgın dönemlerinde hükumetler tarafından alınacak stratejik kararlar açısından vaka sayısını tahmin etmek ve belirtilen ikilemi yönetmek büyük önem taşımaktadır. Bugün neredeyse tüm ülkeler için önemli konulardan birisi de Covid-19 salgınıdır. Ne yazık ki, henüz Covid-19 için etkili bir aşı veya tedavi bulunamamıştır. Ayrıca, bu çalışmanın hazırlığı sırasında, Dünya Sağlık Örgütü tarafından dünya çapında toplam vaka sayısının on üç milyondan fazla olduğu bildirilmiştir. Böyle büyük bir salgınla başa çıkmak için çeşitli karantina önlemlerinin alınması gerekli olmuştur. Hükumetler tarafından alınan karantina önlemleri, ülkeleri ekonomik krizle karşı karşıya getirmiştir. Bu durum ekonomik belirsizlikler yaratmaktadır ve hükumetleri doğru ve en az zararlı stratejik kararlar almak için muazzam bir baskı altına sokmaktadır. Bu nedenlerle hükumetler, ani bir karar vermek yerine durumu adım adım gözlemleyerek Covid-19 için stratejik kararlar almayı tercih etmektedirler. Eğer pandemi vakalarının sayısı belirlenmiş bir zamandan önce tahmin edilebilirse, hükümetlerin halk sağlığı ve ekonomi ikilemini daha doğru bir şekilde yönetmeleri için önemli bir rehber olarak kullanılabilir. Bu nedenle, bu çalışmada 7 gün önceden Covid-19 vakalarını tahmin etmek için yapay sinir ağı (YSA) ve derin öğrenme (uzun-kısa süreli bellek, LSTM ağları) modelleri sunulmuştur. Önerilen modeller Türkiye'nin gerçek verileri üzerinde test edilmiştir. Sonuçlar LSTM modellerinin eğitim seti için hem kümülatif hem de yeni vaka tahminlerinde YSA modellerinden daha iyi performans gösterdiğini göstermiştir. Önerilen modellerin tüm veri seti üzerindeki performansları kıyaslandığında YSA ve LSTM algoritmalarının birbirleri ile rekabet edebilir sonuçlar verdiği gözlemlenmiştir. Ayrıca hem YSA hem de LSTM modellerinin kümülatif vaka tahmini performanslarının yeni vaka tahminlerinden daha iyi olduğu gözlenmiştir.Alternate abstract: Governments face a dilemma between public health and the economy while making strategic decisions on health during a pandemic outbreak. It is of great importance to forecast the number of cases in terms of strategic decisions to be taken by governments especially in outbreak periods and to manage the dilemma mentioned. One of the important issues today is the Covid-19 outbreak for almost all countries. Unfortunately, no effective vaccine or treatment has been found for Covid-19 yet. At the time of this study, however, it was reported that the total number of reported cases by the World Health Organization worldwide was more than thirteen million. Various quarantine measures have been necessary to deal with such a large epidemic. Quarantine measures taken by governments bring countries to face to face with the economic crisis. This creates economic uncertainties and puts governments under tremendous pressure to make accurate and least harmful strategic decisions. For these reasons, governments prefer to make strategic decisions for Covid-19 step by step observing the situation rather than making a sudden decision. If the number of pandemic cases could be predicted before a predetermined time, it would be used as an important guide for governments to manage public health and economic dilemma more accurately. Therefore, this study provides artificial neural network (ANN) and deep learning models (long-short term memory, LSTM networks) to forecast Covid-19 cases before 7-day. The proposed models were tested on real data for Turkey. The results showed that LSTM models performed better than ANN models in both cumulative cases and new cases on the training data set. Comparing the performance of the proposed models over the whole data set, it was observed that the ANN and LSTM algorithms gave competitive results. I addition, the cumulative case forecast performances of both ANN and LSTM models were observed to be better than the new case forecast.

Investigation of Breaking Points in the Airline Industry with Airline Optimization Studies Through Text Mining before the COVID-19 Pandemic

In this study, current literature in the field of airline optimization has been examined by the text mining method to understand trends and commercial threats in the airline industry. Prominent types of work and popular topics have been revealed to understand the importance of global events. This research summarizes trends and some important points relating to airline optimization. The results are striking. It analyzes studies conducted on behalf of aviation before the global COVID-19 pandemic. The economic contribution made by the aviation sector as well as the costs it suffers as a result of crisis situations are clearly explained. Reasons for differences in studies conducted by different countries in the field of aviation are also explained. This study is intended to give an idea of how the aviation sector shapes academic studies, how studies on aviation optimization could contribute in the future, and how the countries have addressed important challenges to the aviation industry in the past.