TR-Dizin İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/396
Browse
5 results
Search Results
Article Forecasting the Consumer Price Index in Türkiye Using Machine Learning Models: A Comparative Analysis(Gazi Univ, 2025-09-01) Söylemez, İsmet; Ünlü, Ramazan; Nalici, Mehmet ErenThis study utilizes machine learning models to forecast Türkiye's Consumer Price Index (CPI), thereby addressing a critical gap in inflation prediction methodologies. The central research problem involves the forecasting of CPI in a volatile economic environment, which is essential for informed policymaking. The primary objective of this study is to evaluate the performance of three machine learning models, such as Decision Tree (DT), Random Forest (RF), and Support Vector Machine (SVM), in forecasting CPI over periods ranging from one to six months, utilizing data from 2012 to 2024. The study's unique contribution lies in the application of the \"SelectKBest\" method, which identifies the most relevant indices, thereby enhancing the efficiency of the models. An ensemble method, Averaging Voting, is also employed to combine the strengths of these models, producing more accurate and robust predictions. The findings indicate that while the RF model consistently generates the most accurate forecasts across all shifts, the SVM model demonstrates a particular strength in the domain of short-term predictions. The ensemble model demonstrates a substantial performance improvement, with a R2 value of 0.962 for one-month ahead of estimates and 0.956 for five-month forecasts. This combined approach has been shown to outperform individual models, offering a more reliable framework for CPI forecasting. The findings offer valuable insights for economic policymakers, enabling more precise and stable inflation predictions in Türkiye.Article Citation - WoS: 2Citation - Scopus: 3Magnetic Separation of Micro Beads and Cells on a Paper-Based Lateral Flow System(Gazi Univ, 2023-12-01) Farooqi, Muhammed Fuad; Icoz, KutayPaper based lateral flow systems are widely used biosensor platforms to detect biomolecules in a liquid sample. Proteins, bacteria, oligonucleotides, and nanoparticles were investigated in the literature. In this work we designed a magnetic platform including dual magnets and tested the flow of micron size immunomagnetic particles alone and when loaded with cells on two different types of papers. The prewetting conditions of the paper and the applied external magnetic field are the two dominant factors affecting the particle and cell transport in paper. The images recorded with a cell phone, or with a bright field optical microscope were analyzed to measure the flow of particles and cells. The effect of prewetting conditions and magnetic force were measured, and it was shown that in the worst case, minimum 90% of the introduced cells reached to the edge of the paper. The paper based magnetophoretic lateral flow systems can be used for cell assays.Article Citation - WoS: 2Machine Learning Based Network Intrusion Detection With Hybrid Frequent Item Set Mining(Gazi Univ, 2024-10-02) Firat, Murat; Bakal, Gokhan; Akbas, Ayhan; Bakal, MehmetWith the development and expansion of computer networks day by day and the diversity of software developed, the damage that possible attacks can cause is increasing beyond the predictions. Intrusion Detection Systems (STS/IDS) are one of the practical defense tools against these potential attacks that are constantly growing and diversifying. Thus, one of the emerging methods among researchers is to train these systems with various artificial intelligence methods to detect subsequent attacks in real time and take the necessary precautions. However, the ultimate goal is to propose a hybrid feature selection approach to improve the classification performance. The raw dataset originally enclosed 85 descriptor features (attributes) for classification. These attributes are extracted using CICFlowMeter from a PCAP file where network traffic is recorded for data curation. In this study, classical feature selection methods and frequent item set mining approaches were employed in feature selection for constructing a hybrid model. We aimed to examine the effect of the proposed hybrid feature selection approach on the classification task for the network traffic data containing ordinary and attack records. The outcomes demonstrate that the proposed method gained nearly 3% improvement when applied with the Logistic Regression algorithm on classifying more than 225,000 records.Article Citation - WoS: 1Citation - Scopus: 2Investigation of the Interaction of Adipose-Derived Mesenchymal Stem Cells With Ε-Polycaprolactone and EGG White Scaffolds(Gazi Univ, 2023-12-01) Oztel, Olga N.; Yilmaz, Hilal; Isoglu, I. Alper; Allahverdiyev, AdilThe development of three-dimensional (3D) cell culture models is becoming increasingly important due to their numerous advantages over conventional monolayer culture. This study aimed to examine the interaction of adipose tissue-derived mesenchymal stem cells (AD-MSCs) with scaffolds composed of e-polycaprolactone (e-PCL) and egg white. In our study, e-PCL and egg white scaffolds were produced from their monomers by tin octoate catalyzed and heat polymerization, respectively. Characterization of e-PCL was carried out by Gel Permeation Chromatography (GPC), Fourier Transform Infrared Spectrophotometry (FTIR), Proton Nuclear Magnetic Resonance (H-NMR), Differential Scanning Calorimetry (DSC) and Scanning Electron Microscopy (SEM). AD-MSCs labeled with red fluorescent CellTracker CM-DiI were cultured on egg white and e-PCL scaffolds for 12 days. Cell viability was determined using 3-(4.5Dimethylthiazol-2yl)-2.5-diphenyltetrazolium bromide (MTT) and nitric oxide (NO) level was evaluated for toxicity. The results showed that the number of AD-MSCs in the egg white scaffold increased periodically for 12 days compared to the other groups. Although the number of ADMSCs in the e-PCL scaffold increased until day 6 of the culture, the number of cells started to decrease after day 6. These results were associated with the toxic effect of lactic acid release on cells resulting from the decomposition of e-PCL scaffolds through catabolic reactions. Therefore, these results indicated that the egg white scaffold enhanced and maintained cell adhesion and cell viability more than the e-Polycaprolactone scaffold and could be used as a scaffold in tissue engineering studies involving stem cells.Article Citation - WoS: 2Citation - Scopus: 3Investigation of Hydrogen Diffusion Profile of Different Metallic Materials for a Better Understanding of Hydrogen Embrittlement(Gazi Univ, 2023-12-01) Kapci, Mehmet Fazil; Bal, BurakIn this study, hydrogen diffusion profiles of different metallic materials were investigated. To model hydrogen diffusion, 1D and 2D mass diffusion models were prepared in MATLAB. Iron, nickel and titanium were selected as a material of choice to represent body-centered cubic, facecentered cubic, and hexagonal closed paced crystal structures, respectively. In addition, hydrogen back diffusion profiles were also modeled after certain baking times. Current results reveal that hydrogen diffusion depth depends on the microstructure, energy barrier model, temperature, and charging time. In addition, baking can help for back diffusion of hydrogen and can be utilized as hydrogen embrittlement prevention method. Since hydrogen diffusion is very crucial step to understand and evaluate hydrogen embrittlement, current set of results constitutes an important guideline for hydrogen diffusion calculations and ideal baking time for hydrogen back diffusion for different materials. Furthermore, these results can be used to evaluate hydrogen content inside the material over expensive and hard to find experimental facilities such as, thermal desorption spectroscopy.