10.6084/m9.figshare.7552208.v1 Ömer Faruk Akmeşe Ömer Faruk Akmeşe Hasan Erbay Hasan Erbay Hakan Kör Hakan Kör Derin Öğrenme ile Görüntü Kümeleme IMISC 2019 Deep learning Image clustering Business Information Systems 2019-01-06 18:30:20 Journal contribution https://imisc.figshare.com/articles/journal_contribution/Derin_O_g_renme_ile_Go_ru_ntu_Ku_meleme/7552208 <div> <div> <div> <div> <p>Title </p> <p>Image Clustering by Deep Learning </p><p><br></p> <p>Abstract </p> <p>The digital data we obtain is not always in the form of a table. The data may be composed of video and audio recordings, plain texts and pictures. Data is converted to perform data mining operations on data in different forms. Successful results are obtained in recognition, classification and clustering of images using machine learning methods. For this, a large number of object images should be given to the computer and learning should be provided. Thus, the computer will begin to establish its own cognition. This is called supervised learning and requires a lot of data. Today, a large number of images have been uploaded to cloud-based systems for sharing or storage. In this study, deep learning method was used for the clustering of the available images. Technically, the image is sent to Google servers and passed through a deep neural network earlier trained here. In deep learning, artificial neural networks are used and give excellent results in terms of accurate results in image aggregation. Image recognition by artificial neural networks mimicking the working principle of human brain, image classification and clustering can be done easily with computer. </p><p><br></p><p> </p><div> <div> <div> <div> <p>Editor: H. Kemal İlter, Ankara Yıldırım Beyazıt University, Turkey<br> Received: August 19, 2018, Accepted: October 18, 2018, Published: November 10, 2018 </p> <p>Copyright: © 2018 IMISC Akmeşe et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. </p> </div> </div> </div> </div> </div> </div> </div> </div>