Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore techniques for handling machine data in this comprehensive talk from ODSC West 2015. Dive into the challenges of working with highly time-series-centric and volatile datasets with high signal-to-noise ratios. Learn about flexible and versatile tools for tackling machine data, including Hidden Markov Models, anomaly detection, and smoothing techniques. Discover real-world applications, the reasoning behind technique selection, and how to apply these methods to practical examples. Follow along with code examples and gain insights into model parameters, fit functions, and general sequence analysis. Ideal for data scientists and analysts looking to expand their skills in processing and analyzing machine-generated data.
Syllabus
Introduction
Sequences of Actions
Objectives
Hidden Markov Model
Anomaly Detection
Identification Relevance
Hidden Markov Models
Why Markov Models
Markov Models
Perfect Recall
Hidden States
Dependency
Forward Backward
Forward Forward
Viterbi Algorithm
Sequence
Welch Algorithm
Smooth Out Time Series
Smoothie Techniques
Smoothie in Time Series
Techniques
Trend similarity
Problem landscape
Prereq
Download Prereq
Code Examples
Ontology
Documentation
Model Parameters
Fit Function
General Sequence
Taught by
Open Data Science