Wos Kapalı Erişimli Yayınlar

Permanent URI for this collectionhttps://hdl.handle.net/11727/10753

Browse

Now showing 1 - 16 of 16

Acoustic Scene Classification Using Spatial Pyramid Pooling With Convolutional Neural Networks
(2019) Basbug, Ahmet Melih; Sert, Mustafa; 0000-0002-7056-4245; AAB-8673-2019
Automatic understanding of audio events and acoustic scenes has been an active research topic for researchers from signal processing and machine learning communities. Recognition of acoustic scenes in the real life scenarios is a challenging task due to the diversity of environmental sounds and uncontrolled environments. Efficient methods and feature representations are needed to cope with these challenges. In this study, we address the acoustic scene classification of raw audio signal and propose a cascaded CNN architecture that uses spatial pyramid pooling (SPP, also referred to as spatial pyramid matching) method to aggregate local features coming from convolutional layers of the CNN. We use three well known audio features, namely MFCC, Mel Energy, and spectrogram to represent audio content and evaluate the effectiveness of our proposed CNN-SPP architecture on the DCASE 2018 acoustic scene performance dataset. Our results show that, the proposed CNN-SPP architecture with the spectrogram feature improves the classification accuracy.
Anomaly Detection in Smart Home Environments using Convolutional Neural Network
(2021) Ercan, Naci Mert; Sert, Mustafa
The use of smart devices in home environments has been increasing in recent years. The wireless connection of these devices to the internet enables smart homes to be built with less cost and hence, recognition of activities in home environments and the detection of possible anomalies in activities is important for several applications. In this study, we propose a new method based on the changepoint representation of sensor data and variable-length windowing for the recognition of abnormal activities. We present comparative analyses with different representations to demonstrate the efficacy of the proposed scheme. Our results on the WSU performance dataset show that, the use of variable-length windowing improves the anomaly detection performance in comparison to fixed-length windowing.
Applications of Deep Learning Techniques to Wood Anomaly Detection
(2022) Celik, Yaren; Guney, Selda; Dengiz, Berna; Xu, J; Altiparmak, F.; Hassan, MHA; Marquez, FPG
Wood products and structures have an important place in today's industry. They are widely used in many fields. However, there are various difficulties in production systems where wood raw material is under many processes. Some difficulty and complexity of production processes result in high variability of raw materials such as a wide range of visible structural defects that must be checked by specialists on line or of line. These issues are not only difficult and biased in manual processes, but also less effective and misleading. To overcome the drawbacks of the manual quality control processes, machine vision-based inspection systems are in great of interest recently for quality control applications. In this study, the wood anomaly has been detected by using deep learning. As it will be a distinction-based method on image processing, the Convolution Neural Network (CNN), which is one of the most suitable methods, has been used for anomaly detection. In addition, it will be tried to obtain the most suitable one among different CNN architectures such as ShuffleNet, AlexNet, GoogleNet for the problem. MobileNet, SqueezeNet, GoogleNet, ShuffleNet among considered methods show promising results in classifying normal and abnormal wood products.
A Component Based Model Developed for Machine Tool Selection Decisions
(2020) Ic, Yusuf Tansel; Yurdakul, Mustafa; AAI-1081-2020
Machine tools are widely used in manufacturing sectors; such as automotive industry, metal cutting industry, aerospace industry etc. Purchase of a machine tool is a long-term capital investment decision and requires a high initial investment cost. Machine tool producers offer a wide-ranging types and models of machine tools. On the other hand, expectations and requirements of the manufacturing companies differ depending on the parts produced and their strategic objectives. High stiffness, rigidity, metal cutting performance, surface finish and low tolerance range are common expectations from machine tools. This paper aims to develop a technical evaluation model to help manufacturing companies in their machine tool purchasing decisions. In the proposed model, first components used in machine tools are analyzed and based on this analysis a technical evaluation model is developed. The application of the developed model is illustrated by making a selection among nine different machine tool alternatives.
Daphnet Freezing Recognition with Gait Data by Using Machine Learning Algorithms
(2020) Guney, Selda; Boluk, Busra
The aim of this study was to test the success of the data set obtained by a wearable health assistant developed for the symptom of freezing (FOG) in gait of Parkinson's patients and to increase the success of the system. The system was tested with different machine learning methods to measure the success of the wearable health assistant system. For all patients (ten patients), the highest success value was obtained and the mean sensitivity and specificity values of the system were calculated and compared with the results obtained in the literature review. In the literature, mean sensitivity and specificity were 73.1% and 81.6%, respectively; In this study, mean sensitivity and specificity were 91.9% and 71.14%, respectively. In order to better analyze the success of the system, two patients with successful and unsuccessful results were selected for the data set in line with the results obtained in the literature review. The success of the system was tested by using different machine learning methods on the data sets of two patients. Finally, the successes obtained by feature extraction methods were tried to be increased. Among the different machine learning methods on the data sets used for patient 8 and patient 3, the most successful method was obtained by combining the models (ensemble). The highest achievement value obtained by attribute extraction methods was obtained when PCA was applied. However, the success value obtained with raw data could not be increased. All results are tabulated and presented.
A Deep LSTM Approach for Activity Recognition
(2019) Guney, Selda; Erdas, Cagatay Berke; 0000-0002-0573-1326; 0000-0003-3467-9923; AAC-7404-2020
Since 1990s, activity recognition effectual field in machine learning literature. Most of studies that relevant activity recognition, use feature extraction method to achieve higher classification performance. Moreover, these studies mostly use traditional machine learning algorithms for classification. In this paper, we focus on a deep (Long Short Term Memory) LSTM neural network for feature free classification of seven daily activities by using raw data that collected from three-dimensional accelerometer. Based on the results, the proposed deep LSTM approach can classify raw data with high performance. The results show that the proposed deep LSTM approach achieved 91.34, 96.91, 88.78, 87.58 as percent classification performance in terms of accuracy, sensitivity, specificity, F-measure respectively.
Detecting COVID-19 from Respiratory Sound Recordings with Transformers
(2022) Aytekin, Idil; Dalmaz, Onat; Ankishan, Haydar; Saritas, Emine U.; Bagci, Ulas; Cukur, Tolga; Celik, Haydar; Drukker, K; Iftekharuddin, KM
Auscultation is an established technique in clinical assessment of symptoms for respiratory disorders. Auscultation is safe and inexpensive, but requires expertise to diagnose a disease using a stethoscope during hospital or office visits. However, some clinical scenarios require continuous monitoring and automated analysis of respiratory sounds to pre-screen and monitor diseases, such as the rapidly spreading COVID-19. Recent studies suggest that audio recordings of bodily sounds captured by mobile devices might carry features helpful to distinguish patients with COVID-19 from healthy controls. Here, we propose a novel deep learning technique to automatically detect COVID-19 patients based on brief audio recordings of their cough and breathing sounds. The proposed technique first extracts spectrogram features of respiratory recordings, and then classifies disease state via a hierarchical vision transformer architecture. Demonstrations are provided on a crowdsourced database of respiratory sounds from COVID-19 patients and healthy controls. The proposed transformer model is compared against alternative methods based on state-of-the-art convolutional and transformer architectures, as well as traditional machine-learning classifiers. Our results indicate that the proposed model achieves on par or superior performance to competing methods. In particular, the proposed technique can distinguish COVID-19 patients from healthy subjects with over 94% AUC.
IS THERE A WAY OUT FOR THE TURKISH ECONOMY?
(2019) Can, Ziya; 0000-0001-5919-4821; AAC-5504-2020
International trade is one of the most important issues of macroeconomics. Almost all international trade theories have tried to determine which country produces what, for which price it sells its products to whom. In a number of theories, these inequalities are based on the differences between factor endowment, while in some others they result from technological development or capital accumulation. However, very few theories have revealed that these differences bear the traces of colonial period. Financial liberalization and the spread of international capital movements by the end of colonialism were major developments in the second half of the 20th century. It is no coincidence that these developments occurred simultaneously. The financial revolution has led to a new kind of relationship between capital owner countries and the others. This is a sort of centre-periphery relationship. Those peripheral countries are the ones that have been affected by the hitches of last slowdown of the world economy. Just like Argentina, Turkey has encountered problems such as the higher inflation rate and the inevitable rise in interest rates, following the melting down in currencies in 2018. Again, in the same period, in contrast with the conventional economic views, current account deficit shrank depending not on the increase in exports, but the dramatic decline in imports instead. As a result of all these occurrences, there was a great loss of prosperity in the country. This study investigates whether it is possible for countries depending on the foreign capital like Turkey and Argentina to follow an independent policy from the fluctuations in the economic conjuncture or not. Is it possible to develop permanent policies that will eliminate dependency on foreign capital, instead of familiar ones such as targeting in inflation, exchange rate or current account deficit? Which has a higher cost? To develop and implement these policies, or to sway in every wind?
Ka Band Far Field Radio Link System Based on OAM Multiplexed Vortex Beams Collimated by a Paraboloidal Reflector
(2021) Hizal, Altunkan; Yildiz, Hayrullah
Electromagnetic vortex-waves (VW) have linear azimuthal-phase, orthogonality in azimuth and orbital-angular-momentum (OAM). The VW-pattern has a null along the vorticity-axis and the cone-half-angle (CHA) and the beam-width (BW) expands with the OAM mode number p. Here, we collimated all p-VW-beams into a radiation cone (Rcon) with a small CHA and BW using a paraboloidal reflector (PaR) fed by a VW uniform-circular-array (UCA). We multiplexed all the transmitted (TX) +/- p-modes, each modulated by a 16QAM symbol-vector. We receive (RX) the TX-signal by p(max) number of nonvortex PaR antennas placed on a small arc of the Rcon. The RX-signal is cast into the standard discrete-Fourier-transform (DFT) format, using the beam-collimation, the azimuthal-orthogonality and zero-padding. The demultiplexing is performed by IDFT. The UCA is designed at Ka-band using circular microstrip-patch-antennas (msPA). The +/- p-modes are TX'ed by orthogonally-polarized separate msPA's. The effects of coupling of +/- p-modes, the calibration inaccuracies and signal-to-noise-ratio (SNR) are simulated by the Monte-Carlo method. It was found that the SNR is very high and the far-field radio-link is feasible. The bit rate of the present OAM-16QAM radio-link is increased by a factor of 2 p(max).
Multiple Service Home Health Care Routing and Scheduling Problem: A Mathematical Model
(2020) Dengiz, Asiye Ozge; Atalay, Kumru Didem; Altiparmak, Fulya
The home health care routing and scheduling problem (HHCRSP) is an extension of the vehicle routing problem (VRP) that are scheduled and routed to perform a wide range of health care services. Nurses, doctors and/or caregivers provide these services at patients' home. In this study, a mathematical model for HHCRSP is presented. The model is extended to take into account additional characteristics and/or constraints based on specific services, patient needs. In the home health care (HHC) problems, services that must be performed simultaneously or within a convinced time are undoubtedly very important. Thus, we consider several numbers of services, skill requirements for the care workers and time windows. Generally, the main aim of the HHC problems is minimizing the travelling distance as well as maximizing the patients' satisfaction. Thus, the model in this study contains both of these objectives taking into account several measurements.
On computer-aided prognosis of septic shock from vital signs
(2019) Ogul, Hasan; Baldominos, Alejandro; Asuroglu, Tunc; Colomo-Palacios, Ricardo; AAC-7834-2020
Sepsis is a life-threatening condition due to the reaction to an infection. With certain changes in circulatory system, sepsis may progress to septic shock if it is left untreated. Therefore, early prognosis of septic shock may facilitate implementing correct treatment and prevent more serious complications. In this study, we assess the feasibility of applying a computer-aided prognosis system for septic shock. The system is envisaged as a tool to predict septic shock at the time of sepsis onset using only vital signs which are collected routinely in intensive care units (ICUs). To this end, we evaluate the performances of computational methods that take the sequence of vital signs acquired until sepsis onset as input and report the possibility of progressing to a septic shock before any further clinical analysis is performed. Results show that an adaptation of multivariate dynamic time warping can reveal higher accuracy than other known time-series classification methods on a new dataset built from a public ICU database. We argue that the use of computational intelligence methods can promote computer-aided prognosis of septic shock in hospitalized environment to a certain degree.
Order Acceptance and Scheduling Problem: A Proposed Formulation and the Comparison with the Literature
(2020) Bicakci, Papatya S.; Kara, Imdat
In classical scheduling problem, it is assumed that all orders must be processed. In the order acceptance and scheduling (OAS) problem, some orders are rejected due to limited capacity. In make-to-order production environment, in which the OAS problem occurs, accepting all orders may cause overloads, delay in deliveries and unsatisfied customers. Oguz et al. (2010) introduced the OAS problem with sequence-dependent setup times and release dates. In this paper, we propose a new mixed integer programming formulation with O(n(2)) decision variables and O(n2) constraints for the same problem. We conduct a computational analysis comparing the performance of our formulation with Oguz et al. (2010) formulation. We use the benchmark instances, which are available in the literature. We observe that our formulation can solve all the instances up to 50 orders in a reasonable time, while Oguz et al. (2010) formulation can solve only the instances with 10 orders in the same time limit.
Predicting Bank Return on Equity (ROE) using Neural Networks
(2021) Balci, Tolgay; Ogul, Hasan
Measuring the performance and profitability of the banking sector, which is the most important part of a country's financial system, is always important. Thanks to the performance measurement, banks can understand the competitive situation, their potential to grow, and the risk, and be more successful in sustaining their lives. This study is considered all state deposit money banks in Turkey. In the literature, using of artificial neural networks (ANN) in banking performance evaluation is rarely studied. Therefore, this paper aims to examine the possibility of ANN utilization for predicting return on equity of Turkey State Deposit Money Banks. The paper compares the accuracy percentages of optimization algorithms of ANN using eleven years quarterly data of six exogenous variables and eight endogenous variables as independent variables and the average return on equity from quarterly of all Turkey state deposit money banks as dependent variable. Given a number of recorded financial parameters, the task is to predict banks' performances using ANN computation methods and to compare prediction results with real results. To evaluate these methods, we built a data set from Banking Regulation and Supervison of Agency, The Banks Association of Turkey and banks' quarterly financial reports. According to all experimental results in optimization models were estimated with above % 80 accuracy. It is determined that the best optimization model is different for each bank.
Sustainable Transportation System Design
(2020) Colak, Melis; Utku, Irem Yaprak; Ozmisir, Deniz; Boz, Alican; Aydogdu, Tayfun; Didis, Mert Cem; Nadar, Emre
As reducing the carbon footprint became one of the topmost concerns of the firms, Company X Turkey has a goal of transforming all of its operations in environmentally sustainable manner. Therefore, they specified their main goal as reducing yearly carbon emission levels of the company by five percent calculated in key performance indicator. Although there are several causes of increased levels of carbon emission, since the control capability of the company is limited in other fields, this study focuses specifically on developing a strategy for reducing the carbon emission generated due to Company X's transportation system in Turkey. The aim of this study is to create a well-designed transportation network through the detection of CO2 emission causes. To lower route-based emission levels, more utilized use of cross-dock locations and alternative fuel usage is recommended, while routing is provided by integer programming. Improvement suggestions including fleet aerodynamics, tire pressure, optimal speed, and acceleration for fleets are constructed in a separate branch to decrease fleet-based carbon emissions in the system.
Understanding the Effect of Assignment of Importance Scores of Evaluation Criteria Randomly in the Application of DOE-TOPSIS in Decision Making
(2019) Ic, Yusuf Tansel; Yurdakul, Mustafa; 0000-0001-9274-7467; AGE-3003-2022
In conventional applications of hybrid DoE-TOPSIS technique in decision making problems, full factorial design layouts are generally used because of their ability to measure the effects of all possible combinations for evaluation factors. In a typical application, for a design layout, a number of replications are generated by assigning different sets of relative importance scores for evaluation factors. A TOPSIS score is then obtained for each experiment and replication pair. Regression analysis is finally applied to obtain a relationship with inputs (values of evaluation factors) and outputs (alternatives' TOPSIS meta-model scores). The key in conventional application of hybrid DoE-TOPSIS technique is generation of relative importance scores. Each set of scores can be assigned by a decision maker or generated randomly. This paper aims to determine whether using either of the two methods in determination of relative importance scores makes any difference in the ranking orders of alternatives.
Wi-Fi Based Indoor Positioning System with Using Deep Neural Network
(2020) Guney, Selda; Erdogan, Alperen; Aktas, Melih; Ergun, Mert
Indoor positioning is one of the major challenges for the future large-scale technologies. Nowadays, it has become an attractive research subject due to growing demands on it. Several algorithms and techniques have been developed over the decades. One of the most cost-effective technique is Wi-Fi-based positioning systems. This technique is infrastructure-free and able to use existing wireless access points in public or private areas. These systems aim to classify user's location according to pre-defined set of grids. However, Wi-Fi signals could be affected by interference, blockage of walls and multipath effect which increases error of classification. In this study Deep Neural Networks and conventional machine learning classifiers are utilized to classify 22 squared grids which represent locations. Five primary Wireless Access Points (WAPs) were mounted indoor environment and 177 secondary WAPs are observed by Wi-Fi module. Dataset was created with using five primary and 177 secondary WAPs. The performance of proposed method was tested using Deep Neural Networks and machine learning classifiers. The results show that Deep Neural Network present the best performance as compared to machine learning classifiers. 95.45% accuracy was achieved by using five primary WAPs and 97.27% accuracy was achieved by using five primary and 177 secondary WAPs together for Deep Neural Network.