Yazar "Ortakci, Yasin" seçeneğine göre listele

Listeleniyor 1 - 6 / 6
  • Küçük Resim Yok
    Öğe
    An analysis of intelligent turkish text classification models for routing calls in call centers: a case study on the republic of turkiye ministry of trade call center
    (2024) Özdemir, Muammer; Ortakci, Yasin
    Call centers play a key role in the management of customer relationships in the modern business world. However, the growing demand for their services presents significant challenges, particularly in terms of staffing and handling increasing call volumes. This paper addresses these issues by presenting an AI-driven text classification framework tailored for the Republic of Turkiye Ministry of Trade Call Centre (MTCC), with the aim of automatically routing calls to relevant departments. Using a specific dataset of 20,000 phone call texts collected from the MTCC, the study employs TF-IDF, Word2Vec, and GloVe text vectorization techniques and applies various machine learning algorithms such as K-Nearest Neighbours, Naive Bayes, Support Vector Machines, Adaptive Boosting, Decision Tree and Random Forest for text classification. Through a comprehensive analysis, the study answers key research questions regarding optimal classifiers and vectorization methods. The proposed solution not only improves the efficiency of MTCC's call routing but also provides researchers with practical insights regarding Turkish text classification. The results indicate that a combination of the Random Forest classifier and Word2Vec text vectorization method is the optimal model that can manage to route calls in real-time.
  • Küçük Resim Yok
    Öğe
    Green AI-Driven Concept for the Development of Cost-Effective and Energy-Efficient Deep Learning Method: Application in the Detection of Eimeria Parasites as a Case Study
    (Wiley, 2024) Acmali, Suheda Semih; Ortakci, Yasin; Seker, Huseyin
    Although large-scale pretrained convolutinal neural networks (CNN) models have shown impressive transfer learning capabilities, they come with drawbacks such as high energy consumption and computational cost due to their potential redundant parameters. This study presents an innovative weight-level pruning technique that mitigates the challenges of overparameterization, and subsequently minimizes the electricity usage of such large deep learning models. The method focuses on removing redundant parameters while upholding model accuracy. This methodology is applied to classify Eimeria species parasites from fowls and rabbits. By leveraging a set of 27 pretrained CNN models with a number of parameters between 3.0M and 118.5M, the framework has identified a 4.8M-parameter model with the highest accuracy for both animals. The model is then subjected to a systematic pruning process, resulting in an 8% reduction in parameters and a 421M reduction in floating point operations while maintaining the same classification accuracy for both fowls and rabbits. Furthermore, unlike the existing literature where two separate models are created for rabbits and fowls, this article presents a combined model with 17 classes. This approach has resulted in a CNN model with nearly 50% reduced parameter size while retaining the same accuracy of over 90%.
  • Küçük Resim Yok
    Öğe
    Optimising customer retention: An AI-driven personalised pricing approach
    (Pergamon-Elsevier Science Ltd, 2024) Ortakci, Yasin; Seker, Huseyin
    Customer churn has become one of the most important challenges that telecom companies have to deal with. Churn cases not only cause revenue losses but also impose extra costs of finding new customers. To overcome this issue, they develop various strategies to retain their customers. In this regard, this study presents an integrated artificial intelligence -based model that can meet the expectations of these companies not only to profile the customer churn, but also to predict a service fee that is likely to be accepted by the customers. The model first identifies the customers who are likely to churn and then offers the customers a personalised service fee that is likely to be acceptable. In this study, the K -nearest neighbours, Decision Tree, Random Forest, and Support Vector Machine methods are adapted as classifiers for churn prediction, and regression models of the same methods are utilised to predict the most optimum personalisedservice fee for potential customer churns. Additionally, to reduce the cost of data collection for companies, the feature selection method is used to determine the most optimal feature subset in churn analysis and service fee prediction. The results show that the Random Forest method is superior to other methods in both churn and price predictions and has resulted in as much as a predictive accuracy of 94% and AUC of 98%. The outcome of this comprehensive analysis using four artificial intelligence methods over three diverse telecom datasets, suggests that the proposed personalisedpricing model in the telecom sector could prevent the churn and increase the profitability by up to 36%. In addition, the model based on SVM suggests that it is possible to reduce the number of required data to be collected by as much as 20%. As the robustness and generalisation ability of the models has been demonstrated over three diverse data sets, it can be further adapted in different sectors.
  • Küçük Resim Yok
    Öğe
    Revolutionary text clustering: Investigating transfer learning capacity of SBERT models through pooling techniques
    (Elsevier - Division Reed Elsevier India Pvt Ltd, 2024) Ortakci, Yasin
    Large Language Models (LLMs), one of the most advanced representatives of neural networks, have revolutionized the field of natural language processing. Among the many applications of these models, text clustering is gaining increasing interest. In particular, the fact that LLMs digitize text more semantically and contextually than existing methods in the literature has led LLMs to produce more successful results with clustering algorithms. However, since these models are not specifically designed for text clustering, they can lead to processing times that exceed acceptable runtime thresholds. To address this challenge, the Sentence-BERT (SBERT) model has been proposed as a solution, offering the ability to accurately measure text similarity by transforming entire texts into dense, fixed-size vectors. SBERT has been integrated into various LLMs, resulting in the creation of diverse SBERT model variants. This study aims to assess the transfer learning capabilities of SBERT models in the context of text clustering. Furthermore, it investigates the influence of CLS (classification token), mean, and max pooling techniques on the performance of these models. In this direction, we applied these pooling techniques to DistilBERT, DistilRoBERTa, ALBERT, and MPNET based SBERT models and compared their performance on different corpora. The results show that there is no clear superiority among the SBERT models. However, the mean pooling emerged as the most effective method in 13 out of 16 text clustering tasks. This finding underscores the high compatibility of the mean pooling technique with SBERT models.
  • Küçük Resim Yok
    Öğe
    SmartEscape: A Mobile Smart Individual Fire Evacuation System Based on 3D Spatial Model
    (Mdpi, 2018) Atila, Umit; Ortakci, Yasin; Ozacar, Kasim; Demiral, Emrullah; Karas, Ismail Rakip
    We propose SmartEscape, a real-time, dynamic, intelligent and user-specific evacuation system with a mobile interface for emergency cases such as fire. Unlike past work, we explore dynamically changing conditions and calculate a personal route for an evacuee by considering his/her individual features. SmartEscape, which is fast, low-cost, low resource-consuming and mobile supported, collects various environmental sensory data and takes evacuees' individual features into account, uses an artificial neural network (ANN) to calculate personal usage risk of each link in the building, eliminates the risky ones, and calculates an optimum escape route under existing circumstances. Then, our system guides the evacuee to the exit through the calculated route with vocal and visual instructions on the smartphone. While the position of the evacuee is detected by RFID (Radio-Frequency Identification) technology, the changing environmental conditions are measured by the various sensors in the building. Our ANN (Artificial Neural Network) predicts dynamically changing risk states of all links according to changing environmental conditions. Results show that SmartEscape, with its 98.1% accuracy for predicting risk levels of links for each individual evacuee in a building, is capable of evacuating a great number of people simultaneously, through the shortest and the safest route.
  • Küçük Resim Yok
    Öğe
    VRArchEducation: Redesigning building survey process in architectural education using collaborative virtual reality
    (Pergamon-Elsevier Science Ltd, 2023) Ozacar, Kasim; Ortakci, Yasin; Kucukkara, Muhammed Yusuf
    Architectural education requires students to work as a group under the supervision of a teacher in the same physical environment since they need interaction to learn how to do a set of measurements in practice. However, a number of obstacles, such as pandemic situations, weather conditions, and the crowdedness of working sites, do not allow them to be and work together in a physical environment. Existing digital solutions, such as online and distance education, do not provide the required immersive and collaborative learning environment to assist students to practice in the same virtual environment. Therefore, the main aim of this study is to develop an immersive architectural educational environment, named VRArchEducation, using Virtual Reality (VR). Specifically, we design and implement a system that allows students and teachers to enjoy direct and simultaneous interaction through their virtual avatars regardless of their current physical location. Furthermore, the proposed VRArchEducation will enable users to have hands-on immersive experience while performing a set of measurement tasks. VRArchEducation presents four virtual fundamental measurement tools: water level hose, plumb, measurement tape, and a sketching board. Using these tools, students can perform a building survey process together in the VRArchEducation system just like in a traditional class environment. We conduct a user study to evaluate the system's effectiveness, success, accuracy, and usability. The proposed VRArchEducation has a great potential to be used in architectural education as an alternative to the traditional environment.(c) 2023 Elsevier Ltd. All rights reserved.