Yazar "Delice, Elif" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim
    Öğe
    An integrated framework for non-traditional machining process technology selection in healthcare applications
    (Strojarski Facultet, 2022) Delice, Elif; Tozan, Hakan; Karadayı, Melis Almula; Harnicarova, Marta; Turan, Başak
    In spite of continuous progress in technical advancement, the conventional machining process became unsatisfactory in the healthcare field due to its disadvantages. This inadequacy lead researchers to consider using the application of nontraditional machining that can machine extremely hard and brittle materials into complicated shapes such as medical devices and implants in healthcare. In this study, the three most popular nontraditional machining process technologies: Laser Beam Machining, Water Jet Machining, and Electrocautery are evaluated to determine the most appropriate technology using the Health Technology Assessment based Multi-criteria Decision-Making framework. HTA is organized evaluation of effects and properties of health technology that enables the application of systematic skills to solve a health problem. HTA's main goal is to raise awareness of new health technologies among decision makers. For these reasons, the HTA core model that enables the production of HTA-related information was utilized.The comparison of selected technologies was carried out via integrating the HTA core model, Best Worst, and Evaluation Based on Distance from Average Solution methods. Finally, a comparison was made to find the most suitable technology to create the necessary infrastructure. As a result, evaluation scores were computed as 0,673; 0,538 and 0,500 for WJM, LBM, and EC, respectively.
  • Küçük Resim Yok
    Öğe
    Applications of data mining algorithms for customer recommendations in retail marketing
    (Nova Science Publishers, Inc., 2022) Delice, Elif; Polatlı, Lütviye Özge; Düzdar Argun, İrem; Tozan, Hakan
    In recent years, researchers have highlighted how large volumes of data can be transformed into information to determine customer behaviors, and data mining applications have become a major trend. It has become critical for organizations to use a tool for understanding the relationships between data to protect their marketplace by increasing customer loyalty. Thanks to data mining applications, data can be processed and transformed into information, and in this way, target audiences can be determined while developing marketing strategies. This chapter aims to increase the market share with products specific to the customer portfolio, introduce strategic marketing tools for retaining old customers, introduce effective methods for acquiring new customers, and increase the retail sales chart, based on purchasing habits of customers. The data set was collected under pandemic conditions during the COVID-19 process and analyzed to support retail businesses in their online shopping orientation. By examining the local customer base, it was assumed that the customer group would display similar behaviors in online or teleordering methods, customer identification and order estimation were made to follow an effective sales policy. Segmentation was performed with data mining applications, and the grouped data were separated according to their similarities. The data set consisting of demographic characteristics and various product information of the enterprise's customers were analyzed with Decision Tree and Random Forest, which are data mining methods, the best performing algorithm in the data set was selected by comparing the performance of the methods. As a result of the findings, appropriate suggestions were given to the business to determine the purchasing tendencies of the customers and to increase the level of effectiveness in sales-marketing strategies. In this way, materials were presented to assist the enterprise in developing strategies to increase the number of sales by taking faster and more accurate action by avoiding the time and expense that would be lost by the trial-error method.
  • Küçük Resim
    Öğe
    Public perception towards children's COVID-19 vaccination with natural language processing
    (İstanbul Medipol Üniversitesi, Fen Bilimleri Enstitüsü, 2022) Delice, Elif; Tozan, Hakan; Karadayı, Melis Almula
    At the end of 2019, Coronavirus disease manifested itself in Wuhan, China, turned into an epidemic, became a global problem, and caused many deaths. As a result of the strategies developed with the spread of the epidemic, parents and children have faced restrictions that deeply affect their lives. Especially in children, negative psychological effects such as depression and tension triggered by anxiety have been encountered due to the restriction of their social lives. Vaccines developed against COVID-19 are seen as a way to end or mitigate the pandemic. However, it is emphasized that the majority of the world's population should accept and have the vaccine in order to gain herd immunity. At this point, many debates arose about the vaccination of children with a sensitive bodies so that they can return to their normal lives, and the public's reaction was echoed through social media. As a consequence of the detailed literature search, to the best of our knowledge, no study specifically determined the judgments of families about getting their children vaccinated, that combined topic modeling which reveals the hidden meanings in the texts, and sentiment analysis techniques, that determine the emotional states of individuals. supported by Twitter data, and presented a model, generally, it was seen that the studies carried out analyses using topic modeling, as well as, sentiment analysis methods separately and they made inferences in this way. In this direction, by observing the deficiency in the literature, it is targeted to present a model that reveals the perception of parents about getting their children vaccinated, extracts the main themes, and determines the emotional changes about these topics, thus, it is intended to contribute to the literature in this area. Furthermore, it is hoped that the outputs acquired as a result of the implementation would shed light on the subject and guide future studies by presenting a model to the policymakers in the process of developing their strategies. Scope of work, with the support of the Octoparse web scraping tool, data was extracted from Twitter with the help of keywords determined in English in order to address the whole world, within the framework of the date of August 1, 2020-October 1, 2021, when the epidemic turned into a global problem and the discussions about vaccines intensified. Then, using the topic modeling and sentiment analysis techniques under the umbrella of NLP, main, sub-topics about parents' attitudes were revealed, also vaccine perceptions and attitudes were detected by performing sentiment analysis within the framework of the identified subjects. As a result, four topic clusters were determined: "the opinion of the need for the first dose of vaccination according to age", "the effectiveness of the first dose of vaccine", "the opinion of the need for vaccination of school-age children", and "the need for vaccination arising from the protection of unvaccinated children with only mask protection". With the processing of the determined topic clusters with sentiment analysis, it was determined that positive emotions were dominant, and three emotions, namely trust, expectation, and fear, came to the fore.