Yıldız Technical University Research Information System
Not a member yet
91324 research outputs found
Sort by
Aging-Based Weighting for Session Classification in User Behavior Analysis
No full textComprehending user behavior on e-commerce platforms is essential for augmenting customer interaction and refining recommendation algorithms. Clickstream data provides a significant resource for examining user navigation patterns; nevertheless, accurately describing and categorizing user sessions poses a challenge. This paper shows how to improve the accuracy of session-based classification using an embedding-based method combined with an aging-based weighting mechanism. The considered embedding methods, Word2Vec, Node2Vec, and LSTM Autoencoder, can turn session-based clickstream data into numbers. Furthermore, a dynamic weighting technique is introduced to emphasize recent interactions to improve classification performance. Our empirical assessment on an authentic e-commerce dataset reveals that the LSTM Autoencoder surpasses conventional embedding methods in capturing sequential dependencies. In addition, the age-based weighting technique markedly improves the classification accuracy, especially when used with deep learning models. A comparison of different classification algorithms, such as Random Forest, Logistic Regression, Gaussian Naive Bayes, and LSTM, shows that LSTM models are the best at finding correlations between events over time. The results also show the importance of temporal weighting in session-based clickstream analysis and provide a solid foundation for further research in behavioral analytics and personalized recommendation systems. This paper introduces an efficient method for clickstream-based user modeling that facilitates better user engagement in e-commerce systems
KENT KİMLİĞİNİ YANSITACAK VE İKLİM DEĞİŞİKLİĞİNE UYUM SAĞLAYABİLECEK ODUNSU BİTKİLER: BİLECİK
No full textCost and energy improvement comparison of using electronic expansion valve and vacuum insulation panel in domestic refrigerators
No full textThe paper aims to decrease energy consumption (EC) and costs for refrigerators across various climates. Although research on vacuum insulation panels (VIP) in refrigerators is limited, no studies explicitly examine the effects of VIPs and electronic expansion valves (EEV) on EC, thermal efficiency, cost, and insulation aging (for both VIPs and polyurethane) across different climate classes. This research is highly valuable because it compares multiple factors simultaneously. VIPs enhance efficiency by decreasing heat gain, thus increasing the coefficient of performance (COP), while EEVs adjust refrigerant flow to increase COP. However, VIPs are expensive. In this research, six configurations, including VIP and EEV, and without VIP or EEV, were tested on the same refrigerator across four climate classes. Moreover, a 5% annual increase in electricity cost is included. Results revealed that the minimum payback period was 2 years for an EEV-only in its vapor compression cycle with no-VIP. In the subnormal (SN) class, a full-VIP insulated cabinet and an EEV-only cabinet consumed the same energy (166 kWh/year). In the normal (N) class, the door VIP-only with an EEV reduced EC by 2% and shortened the payback period by 3 years. In the subtropical (ST) class, both full-VIP and EEV configurations consumed 221 kWh/year. In the tropical (T) class, full-VIP resulted in the lowest EC, making EEV technology ineffective. This study enables comparisons of cabinet configurations based on EC, cost, heat gain, aging, and EEV technology across climate classes
Generating landslide archive inventories for Türkiye using web scraping and natural language processing techniques
No full textLandslides are among the most frequent natural hazards that cause significant loss of life and serious economic damage worldwide. Although many inventories have been created using different approaches to understand landslide events, these inventories are rarely updated automatically or in real time. Traditional approaches are time-consuming and labor-intensive and are often limited in timeliness because of reporting delays. To address these challenges, we developed an automated approach that integrates web scraping, natural language processing (NLP), and geocoding techniques using digital media news sources in Türkiye to create a landslide archive inventory. Our algorithm verified 1727 of the 3051 news articles it captured between 1997 and 2024 as landslides and identified a total of 478 fatalities in 212 deadly incidents. A total of 66.5% of the landslides captured on the web were located at the neighborhood/village level, providing substantial spatial accuracy. This location accuracy also enabled risk estimation at the neighborhood/village level. A comparison with the manual national inventory revealed moderate agreement, with F1 scores ranging from 0.434 to 0.552 in the ± 1 and ± 7 daytime windows, respectively. The automated method not only captures spatial and temporal patterns of landslides but also extracts key attributes such as location, number of fatalities, and triggering factors (i.e., natural and anthropogenic). Our study demonstrates the potential of web-based automated approaches to complement traditional landslide inventories by providing near-real-time and verifiable data. Finally, we suggest adopting common reporting standards for natural hazard digital newspapers so that this approach can be applied globally
Determination of arsenic species in seafood samples using a combination of HPLC and ICP-MS after ultrasonic assisted sample preparation
No full textIn this study, an analytical method was applied for the separation and determination of inorganic and organic arsenic species (As+3, As+5, AsB, MMA) in anchovy, haddock, mussels and prawns using HPLC coupled with ICP-MS. Efficient separation of the arsenic species on the HPLC-ICP-MS system was achieved using 50 mM (NH4)2CO3 at pH 9.50 diluted in 1 % methanol and 0.50 mM EDTA as mobile phase. The arsenic species were extracted from the samples by means of ultrasonically assisted sample preparation. The LOD/LOQ values of the detection system were 0.07/0.22 ng/mL, 0.12/0.41 ng/mL, 0.06/0.22 ng/mL and 0.03/0.11 ng/mL for arsenobetaine (AsB), As+3, methylarsonic acid (MMA) and As+5, respectively. The relative standard deviations calculated for the lowest calibration standards ranged between 1.01 and 7.88 %, verifying good precision for replicate measurements. The accuracy of the method was validated by spike recovery experiments, with recorded recoveries in the range of 85-117 %.The total arsenic content of the extracts was determined by direct ICP-MS analysis. A certificated reference material (NIST-1573A) was analyzed for the total As concentration. The method was successfully applied for the qualitative and quantitative determination of AsB, As+3, MMA and As+5 in the seafood samples ranging from 6.0 to 5700 ng/g. The carcinogenic and non-carcinogenic risk for haddock and anchovies was low, whereas the carcinogenic risk for prawns was relatively high but below the threshold value, according to the risk assessment of the samples
PEGylated isothiocyanate-functionalized zinc(II) phthalocyanine exhibits cell-type dependent photodynamic activity in 2D and 3D tumor models
No full textThis study reports the synthesis and characterization of an asymmetric zinc(II) phthalocyanine (5) containing three tetraethyleneglycol monomethyl ether groups and one isothiocyanatophenoxy group at its periphery. The isothiocyanate unit was selected to ensure selective bioconjugation under mild reaction conditions and to reduce side product formation, while tetraethyleneglycol monomethyl ether groups were incorporated to increase solubility and tailor photophysical and photochemical properties relevant to photodynamic therapy applications. The compound showed a singlet oxygen quantum yield (ΦΔ) of 0.38, confirming efficient photosensitizer performance. Photodynamic activity was evaluated across multiple cancer cell lines in both 2D monolayer and 3D spheroid cultures. In 2D models, compound 5 produced pronounced light-dependent cytotoxicity accompanied by increased intracellular ROS. Cell-death profiles varied among cancer types, with FaDu cells showing the highest sensitivity under the tested conditions, consistent with differences in cellular susceptibility to compound 5–mediated PDT. In 3D spheroids, efficacy was reduced, in line with known limitations of PDT in compact tumor-like structures, including restricted light propagation, oxygen gradients, and limited compound penetration. Minimal phototoxicity in non-malignant fibroblasts under the same conditions suggests preferential photodynamic activity in the tested cancer models. Overall, these results support the PDT potential of compound 5 and highlight the influence of cellular context and 3D architecture on treatment responses
Adverse childhood experiences and fear of happiness: serial mediation by belongingness and dating anxiety
No full textAdverse childhood experiences (ACEs) constitute the foundation of many problems in an individual’s life. Among these are a diminished sense of belonging and heightened anxiety, particularly in the context of romantic relationships. Individuals who suffer from anxiety in romantic relationships frequently worry about being rejected or feeling inadequate. Under this emotional strain, people could doubt chances to capture happiness. In this study, the serial mediation effects of belongingness and dating anxiety were examined in the ACEs - fear of happiness link. The sample consists of 346 participants (M = 22.82, SD = 3.37). The mediating role was assessed using the bootstrap method through structural equation modeling (SEM) to verify the presence of mediation. SEM indicated that belongingness and dating anxiety were significant mediators in the association between ACEs and fear of happiness. According to the results, ACEs may increase fear of happiness, which may arise as a consequence of low belongingness and high dating anxiety. As a result, happiness may be hindered by ACEs as well as by poor connection with other people and the degree of anxiety they experience about establishing and maintaining romantic relationships. Based on these findings, it is recommended to develop early intervention programs and support services for individuals with ACEs. These programs can focus on enhancing feelings of belongingness and reducing dating anxiety, thereby mitigating fear of happiness
Radio frequency identification and real-time security of LoRa devices with deep learning on embedded GPU
No full textIn this paper, we propose the creation of unique Radio Frequency (RF) fingerprints by using communication signals for physical layer security of LoRa devices and real-time security control with these fingerprints. To achieve the goal, we constructed a 250 GB dataset, comprising 350,000 samples, by decoding raw LoRa communication data with Software Defined Radio and Embedded Linux and labeling each decoded ID with the corresponding raw LoRa signal. Utilizing the dataset, we conducted training in deep-learning models Convolutional Neural Networks, Temporal Convolutional Networks, Long Short-Term Memory, and Gated Recurrent Units, on a GPU server infrastructure. The CNN model demonstrated the best performance among the deep-learning models, achieving an accuracy rate of 99.8%. However, over time, we observed a decline in performance due to variations in the characteristics of electronic components. To ensure the stability of the system, we have determined that it is essential to retrain the model using the data acquired during secure communication. Low-power devices have limited packet transmissions and may not allocate a sufficient amount of data for retraining purposes. Hence, we performed minimum data analysis to retrain the system at regular intervals and found that even one packet per hour is sufficient for the devices. © 2017 Elsevier Inc. All rights reserved
Artificial Synaptic Device and Chaotic Oscillator Implementation Using a Novel Floating Memtranstor Emulator
No full textThe memtranstor is a memory element that establishes a direct relationship between charge and magnetic flux through nonlinear magnetic effects and is classified as the fourth memory element after the memristor, memcapacitor, and meminductor. This paper discusses the design of a floating emulator integrating a newly introduced memory element called memtranstor. The proposed memtranstor circuit employs four differential voltage current conveyors (DVCCs), one analog multiplier, three grounded capacitors, and four grounded resistors. The PSPICE simulation results are done using the 0.18-μm CMOS technology parameter to confirm the functionality of the suggested floating emulator circuit. By altering the parameters in the models, a variety of simulations are done including memory effect simulations, Monte Carlo simulations, pinched hysteresis loops using various DC control voltages and frequencies, temperature variation, and output voltage noise simulations. To demonstrate the potential applications of the proposed memtranstor, its artificial synaptic plasticity and its role in a memtranstor-based chaotic oscillator are validated through example simulations
On the soliton solutions of the stochastic Schrödinger-Hirota equation with Kerr law and spatio-temporal dispersion
No full textThis study intends to attain the optical soliton solutions of the stochastic Schrödinger-Hirota equation with Kerr law and spatio-temporal dispersion in Itô calculus. The Schrödinger-Hirota equation is employed to describe daily life problems in dispersive optical fibers and nonlinear optics. The new Kudryashov scheme and a subversion of the new extended auxiliary equation scheme are implemented to achieve this purpose. As a result, the stochastic model's bright and dark soliton solutions are successfully derived by allocating appropriate parameter values. For the first time, the effects of noise impact on the solitons' dynamics are examined in detail and are depicted by various graphs. It is expected that the results and graphics obtained in this paper will shed light on researchers to fill the gaps in this branch