Browsing by Author "Amankossova A."

Now showing 1 - 3 of 3
  • Thumbnail Image
    ItemOpen Access
    AN EVALUATION OF UNSUPERVISED OUTLIER DETECTION METHODS FOR UNIVARIATE TIME SERIES DATA IN FINANCIAL TRANSACTIONS
    (СДУ хабаршысы - 2023, 2023) Amankossova A.; Turan C.
    Abstract. An essential problem in finance application areas is identifying abnormal subsequences in time series data. Despite the wide range of outlier detection algorithms, no substantial research has been conducted to thoroughly investigate and assess the various methodologies, particularly in the financial industry. This study focuses on comparing and contrasting the outcomes of various unsupervised algorithms. The findings reveal that the Local Outlier Factor technique outperforms the other methods in terms of precision, recall, and Fl-score. The research provides valuable insights for financial institutions and businesses looking to improve their identification of abnormalities systems and highlights the importance of choosing the appropriate unsupervised outlier detection method for financial transaction data. The outcomes of this study can be used to inform future research and development in the area of financial unusual case detection.
  • Thumbnail Image
    ItemOpen Access
    Automating banking sector monitoring procedures for exceptional situations
    (2023) Amankossova A.
    The need to detect anomalous events and react to them immediately in real time is becoming increasingly important in the banking sector. The main objective of this thesis is to propose the development of a real time alert notification system that uses outlier detection algorithms to discover unexpected trends in the key performance indicators of a financial industry. In order to enable real-time monitoring of data streams and notify users of anomalous occurrences as they happen, the system will take advantage of the capabilities of cloud computing and big data technologies. The proposed system will be evaluated against traditional outlier identification techniques. The efficacy of the outlier detection algorithms for the banking dataset is assessed using precision, recall, and Fl score measurements. The approaches of sending alerts are evaluated, with the strengths and weaknesses of each method taken into account. This thorough evaluation approach aims to emphasise the advantages and disadvantages of the suggested system as well as identify potential areas for improvement. The suggested system will allow users to take proactive action to lessen the consequences of abnormal events, reduce the risk of costly downtime and other adverse effects.
  • Thumbnail Image
    ItemOpen Access
    IMPLEMENTATION OF A REAL-TIME ALERT-NOTIFICATION SYSTEM FOR DATA MONITORING IN THE FINANCIAL INDUSTRY
    (СДУ хабаршысы - 2023, 2023) Amankossova A. ; Turan C.
    Abstract. An alert notification system that operates in real-time is crucial for data monitoring, as it enables financial institutions to detect and address possible compliance issues promptly, preventing them from escalating into more significant problems. In the event that there are issues with the data, we may not realise that some parameter values are missing. In such instances, automatic alerts are unquestionably the most effective system. The use of an automated system to notify the responsible parties will benefit any organization's digitalization and reduce the burden on employees who must monitor manually. The article provides an examination and contrast of real-time alert automation techniques implemented through the programming languages of R and Python. The analysis encompasses the problems and limitations of the different methods. This study contributes to the field of automated notifications in the financial industry by evaluating different methods.