Machine Learning Tools in Seeq скачать в хорошем качестве

Machine Learning Tools in Seeq

Summary This video provides an overview of Seeq's machine learning anomaly detection tools, focusing on Isolation Forest and Self-Organizing Map algorithms. It explains the theory behind each method, demonstrates their use within Seeq, and compares their strengths and limitations for different types of industrial data analysis scenarios. Chapters 00:00 - Introduction Overview of Seeq's machine learning tools and the agenda for the video. 00:30 - Isolation Forest Overview Explains the Isolation Forest algorithm and its application for detecting outliers in multidimensional datasets. 01:35 - Isolation Forest Visual Example Walkthrough of a visual example showing how Isolation Forest isolates anomalies. 02:09 - Self-Organizing Map Overview Introduction to the Self-Organizing Map (SOM) algorithm, including how it detects anomalies using neural network-based clustering. 03:14 - Self-Organizing Map Visual Example Visual demonstration of SOM anomaly detection and explanation of clustering. 03:48 - Using Isolation Forest in Seeq Live demonstration of setting up and configuring Isolation Forest in Seeq, including key parameter options. 06:43 - Isolation Forest Results & Signal Output Reviewing results, interpreting output, and using the Isolation Forest output as a signal for further analysis. 07:55 - Using Self-Organizing Map in Seeq Live demonstration of configuring and running the Self-Organizing Map in Seeq, including advanced parameters. 11:44 - Self-Organizing Map Results & Signal Output Interpreting SOM results, reviewing output as a signal, and comparing to Isolation Forest. 12:53 - Pros and Cons of Each Method Summary of the advantages and disadvantages of Isolation Forest and Self-Organizing Map algorithms for anomaly detection.