ORMAE

    Contact Us
    Contact Us
    Menu

    Anomaly Detection

    TELECOM AND CALL CENTER

     

    Enhancing vulnerability management through Deep Learning, Random Forest and Clustering.

    About the customer

    An established e-commerce company that extensively used Data Science and related technologies to automate various aspects of its business to enhance revenues and customer satisfaction.

    Business Problem

    As the market leader in its space, the e-commerce firm is pro-active in observing patterns in its data. To ensure that any deviation from business as usual is noted well in advance, the management wanted to:

    • Develop an anomaly detection algorithm of its web-infrastructure system.
    • Identify and isolate the reasons for detecting violations in response times encountered by customers, and predict them at least five minutes ahead of time.

    Our Approach

    ORMAE's team identified the core aspects around which the solution had to be developed. They included:

    • Problem scale: Five applications ~200 server instances.
    • Pattern recognition.
    • Processing of unstructured data.
    • Identifying inconsistent response times.
    • Random Forest.
    • Deep learning.
    • Clustering.
    Get in Touch

    Have a project in mind?

    Looking for collaboration?
    Send an email to bdm@ormae.com
    for availability and enquires.

    The Result

    ORMAE's optimization algorithm delivered the required results (taking into account factors like cannibalization, post-promotion dip effects and others) to the FMCG client. They included:

    • Random Forest that predicted anomalies five minutes in advance and with 93% accuracy.
    • Improved customer experiences and system efficiencies.
    • Reduced customer churn rates.

    Get in Touch

    Have a project in mind?

    Looking for collaboration?
    Send an email to bd@ormae.com
    for availability and enquires.

    ×