Yazar "Çelik, Mete" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    Anomaly Detection in Bitcoin Prices using DBSCAN Algorithm
    (2020) Dokuz, Ahmet; Çelik, Mete; Ecemiş, Alper
    Blockchain is an emerging technology which is also behind the Bitcoin digital money. Daily bitcoin transactions are increasing due tothe popular and widespread investments. The increase of Bitcoin related datasets and this increased big dataset requires novelapproaches and methods to analyze using data mining techniques. In addition, fluctuations and anomalies in the bitcoin prices couldmean a great deal to economists and discovering anomalies in bitcoin prices is important. In this study, anomaly detection in Bitcoinprices is performed based on the change of Bitcoin price difference and the change of Bitcoin price difference in percentage withrespect to previous day using 8-years of Bitcoin price dataset of the period of 2012-2019. First, the dataset is pre-processed andunnecessary columns are deleted. Then, 2 different datasets are created by using daily bitcoin prices, i.e. bitcoin price differencedataset and bitcoin price difference in percentage dataset. After that, for detecting anomalous price changes, DBSCAN algorithm andstatistical method are used, and the performance of the algorithms are evaluated. The results show that the DBSCAN algorithm andstatistical method successfully detects anomalies in bitcoin prices for both of the datasets. However, the DBSCAN algorithm performsbetter than the statistical method which could detect anomalies even they are close to the normal daily price changes. Also, in thisstudy, bitcoin price difference dataset and bitcoin price difference in percentage dataset are compared and the differences of theresults for both datasets and their reasons are explained.
  • Küçük Resim Yok
    Öğe
    Emerging and Vanishing Association Pattern Mining in Hydroclimatic Datasets
    (2018) Çelik, Mete; Dokuz, Ahmet Şakir; Çelik, Filiz Dadaşer
    Emerging and vanishing association patterns can be defined as association patterns whose frequencies (supports) get strongerand weaker over time, respectively. Discovering these patterns is important for several application domains such as financial andcommunication services, public health, and hydroclimatic studies. Classical association pattern mining algorithms do not considerhow the strengths of association patterns change over time. An association pattern can be defined as an emerging or vanishing patternwhen its support measure changes over time. In this paper, we focus on discovery of time evolving association patterns (i.e., emergingand vanishing association patterns) from datasets. To discover such patterns, a novel algorithm, named as Emerging and VanishingAssociation Pattern Miner (EVAPMiner) algorithm, was proposed. The proposed algorithm was evaluated using hydroclimatic datasetof Turkey. The analyses showed that the proposed algorithm successfully detects emerging and vanishing association patterns inhydroclimatic datasets.