About Me

My Interests:

• I am interested in computer vision, end-to-end deep learning, generative models,
• I developed machine learning models, methods, especially time-series data analysis, fault analysis, deep learning, transfer learning.
• My overall future goal is to develop deep/machine learning algorithms.​
• Projects: https://github.com/omerbsezer

Publications:

Scholar Google Link: https://scholar.google.com.tr/citations?user=t-i73N4AAAAJ&hl=tr&oi=ao

ResearchGate Link: https://www.researchgate.net/profile/Omer_Sezer

Github Link: https://github.com/omerbsezer

Journal Papers:

• Algorithmic Financial Trading with Deep Convolutional Neural Networks: Time Series to Image Conversion Approach
  Omer Berat Sezer, Ahmet Murat Ozbayoglu
  (Journal) Applied Soft Computing, 2018
  Paper Links: [Elsevier] [ResearchGate] [BibTex]
• Context Aware Computing, Learning and Big Data in Internet of Things: A Survey
  Omer Berat Sezer, Erdogan Dogdu, Ahmet Murat Ozbayoglu
  (Journal) IEEE Internet of Things Journal, 2018
  Paper Links: [IEEE] [ResearchGate] [BibTex]
• Financial Trading Model with Stock Bar Chart Image Time Series with Deep Convolutional Neural Networks
  Omer Berat Sezer,  Ahmet Murat Ozbayoglu
  (Journal) Intelligent Automation & Soft Computing Journal, In-progress. Accepted.
  Paper Links: [Arxiv]

Conference Papers:

• MIS-IoT: Modular Intelligent Server Based Internet of Things Framework with Big Data and Machine Learning
  Aras Onal, Omer Berat Sezer, Ahmet Murat Ozbayoglu, Erdogan Dogdu
  (Conference) IEEE Big Data, 2018
  Paper Links: [IEEE]
• The Design and Architecture of Earth Orbiting Satellite Simulator (EOSS)
  Burak Ekinci, Yavuz Ozturk, Berk Yurttagul, Omer Berat Sezer, Coskun Celik, Murat Karahan
  (Conference) AIAA Space Ops, 2018
  Paper Links: [AIAA]
• A Deep Neural-Network Based Stock Trading System Based on Evolutionary Optimized Technical Analysis Parameters
  Omer Berat Sezer, Ahmet Murat Ozbayoglu, Erdogan Dogdu
  (Conference) Procedia Computer Science, Complex Adaptive System, 2017
  Paper Links: [ScienceDirect][ResearchGate][BibTex][Code-Github]
• An Artificial Neural Network-based Stock Trading System Using Technical Analysis and Big Data Framework
  Omer Berat Sezer, Ahmet Murat Ozbayoglu, Erdogan Dogdu
  (Conference) Proceedings of the SouthEast Conference, 2017
  Paper Links: [ACM][ResearchGate][BibTex][Code-Github]
• Weather data analysis and sensor fault detection using an extended IoT framework with semantics, big data, and machine learning
  Aras Onal, Omer Berat Sezer, Ahmet Murat Ozbayoglu, Erdogan Dogdu
  (Conference) IEEE Big Data, 2017
  Paper Links: [IEEE][ResearchGate][BibTex][code]
• An extended IoT framework with semantics, big data, and analytics
  Omer Berat Sezer, Ahmet Murat Ozbayoglu, Erdogan Dogdu, Aras Onal
  (Conference) IEEE Big Data, 2016
  Paper Links: [IEEE][ResearchGate][BibTex]
• Development of a smart home ontology and the implementation of a semantic sensor network simulator: An Internet of Things approach
  Omer Berat Sezer, Serdar Zafer Can, Erdogan Dogdu
  (Conference) IEEE International Conference on Collaboration Technologies and Systems (CTS), 2015
  Paper Links: [IEEE][ResearchGate][BibTex][Code-Github]
• A simple self-timed implementation of a priority queue for dictionary search problems
  Ali Muhtaroglu, Omer Berat Sezer
  (Conference) Adaptive Science & Technology, 2009
  Paper Links: [IEEE][BibTex]

Thesis:

• Implementation And Evaluation Of The Dependability Plane For The Dynamic Distributed Dependable Real Time Industrial Protocol (D3RIP)
  Omer Berat Sezer
  (M.Sc Thesis) Department of Electric and Electronics Engineering, Middle East Technical University, Ankara, Turkey, 2013
  Paper Links: [Metu-Library][BibTex]

• Zaman Serisi Verilerinin Derin Yapay Sinir Ağları ile Analizi ve Eniyilemesi: Finansal Tahmin Algoritmaları (Analysis and Optimization of the Time Series Data with Deep Artificial Neural Networks: Financial Estimation Algorithms)
  Omer Berat Sezer
  (Ph.D Thesis) Department of Computer Engineering, TOBB University of Economics and Technology, Ankara, Turkey, 2018
  Paper Links: [ResearchGate]

Certificates / Machine Learning Courses:

• “Deep Learning Specialization”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: 7VEJK7E5KH4C. (Five interconnected courses about  Computer Vision, Natural Language Processing, Speech Recognition: “Sequence Models”, “Convolutional Neural Networks”, “Structuring Machine Learning Projects”, “Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization”, “Neural Networks and Deep Learning”)
• “Sequence Models”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: VNWMURQ4YGDF. (Recurrent neural networks, backpropagation through time, different types of RNNs, language model, gated recurrent unit, LSTM, natural language processing, word embeddings, word2vec, sentiment classification, beam search, trigger word detection, speech recognition)
• “Convolutional Neural Networks”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: R79J7S9ZZYRS. (Computer vision, edge detection, padding, convolutions, pooling, Residual Nets, Inception network,  data augmentation, landmark detection, object detection, YOLO, one shot learning, face recognition, neural style transfer)
• “Structuring Machine Learning Projects”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: HHP5EABRTSLW. (Single number evaluation metric, train/dev/test distribution, avaidable bias, human level performance, error analysis, addressing data mismatch, transfer learning, multi-task learning, end-to-end deep learning)
• “Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: ZRGYXHMGFW9Z.  (Bias/variance, regularization, dropout, weight initialization, vanishing/explodient gradients, mini-batch, momentum, RMSprop, Adam optimization, learning rate decay, batch normalization, softmax, dl frameworks, tensorflow )
• “Neural Networks and Deep Learning”, Coursera, Prof. Andrew Ng, deeplearning.ai, Licence Number: UDTU7C54Z8D4.  (Logistic regression, shallow network and deep learning neural network implementation, backpropagation, forward propagation, vectorization, gradient descent, activation function)
• “Neural Networks for Machine Learning”, Coursera, Prof. Geoffrey Hinton, University of Toronto, Licence Number: 2XUGY5H26DXS. (Perceptron, forward and back propagation, convolutional nets, mini-batch, momentum, RMSprop, recurrent neural nets, LSTM, Hopfield nets, Boltzmann machines, RBM, Deep Belief Nets, Autoencoders)

Confirmation: https://www.coursera.org/account/accomplishments/certificate/\#LicenceNumber\#

e.g: https://www.coursera.org/account/accomplishments/certificate/ZRGYXHMGFW9Z