The document discusses the importance of machine learning (ML) in business operations, emphasizing its application to predictive maintenance, security threat detection, and customer churn analysis. It explains ML processes, types, and workflows alongside potential use cases in IT, security, and analytics, while highlighting the capabilities of Splunk's ML tools. The document also provides guidance on getting started, including software installation and exploration of data for fitting predictive models.

2. Copyright © 2016 Splunk Inc.
Machine Learning
Andrew Phillips
Sr. Sales Engineer

3. 3
Disclaimer
During the course of this presentation, we may make forward looking statements regarding future
events or the expected performance of the company. We caution you that such statements reflect our
current expectations and estimates based on factors currently known to us and that actual events or
results could differ materially. For important factors that may cause actual results to differ from those
contained in our forward-looking statements, please review our filings with the SEC. The forward-looking
statements made in the this presentation are being made as of the time and date of its live presentation.
If reviewed after its live presentation, this presentation may not contain current or accurate information.
We do not assume any obligation to update any forward looking statements we may make.
In addition, any information about our roadmap outlines our general product direction and is subject to
change at any time without notice. It is for informational purposes only and shall not, be incorporated
into any contract or other commitment. Splunk undertakes no obligation either to develop the features
or functionality described or to include any such feature or functionality in a future release.

4. Copyright © 2016 Splunk Inc.
Why do we need ML?

5. 5
ML in Everyday life

6. Copyright © 2016 Splunk Inc.
Historical Data Real-time Data Statistical Models
DB, Hadoop/S3/NoSQL, Splunk Machine Learning
T – a few days T + a few days
Why is this so challenging using traditional methods?
• DATA IS STILL IN MOTION, still in a BUSINESS PROCESS.
• Enrich real-time MACHINE DATA with structured HISTORICAL DATA
• Make decisions IN REAL TIME using ALL THE DATA
• Combine LEADING and LAGGING INDICATORS (KPIs)
Splunk
Security Operations Center
Network Operations Center
Business Operations Center

7. Copyright © 2016 Splunk Inc.
What is ML?

8. 8
ML 101: What is it?
• Machine Learning (ML) is a process for generalizing from examples
– Examples = example or “training” data
– Generalizing = building “statistical models” to capture correlations
– Process = ML is never done, you must keep validating & refitting models
• Simple ML workflow:
– Explore data
– FIT models based on data
– APPLY models in production
– Keep validating models
“All models are wrong, but some are useful.”
- George Box

9. 9
Types of Machine Learning
1. Supervised Learning: generalizing from labeled data

10. 10
Types of Machine Learning
2. Unsupervised Learning: generalizing from unlabeled data

11. 11
Types of Machine Learning
3. Reinforcement Learning: generalizing from rewards in time
Leitner System Recommender systems

12. Copyright © 2016 Splunk Inc.
ML Use Cases

13. 13
IT Ops: Predictive Maintenance
1. Get resource usage data (CPU, latency, outage reports)
2. Explore data, and fit predictive models on past / real-time data
3. Apply & validate models until predictions are accurate
4. Forecast resource saturation, demand & usage
5. Surface incidents to IT Ops, who INVESTIGATES & ACTS
Problem: Network outages and truck rolls cause big time & money expense
Solution: Build predictive model to forecast outage scenarios, act pre-emptively & learn

14. 14
Security: Find Insider Threats
Problem: Security breaches cause big time & money expense
Solution: Build predictive model to forecast threat scenarios, act pre-emptively & learn
1. Get security data (data transfers, authentication, incidents)
2. Explore data, and fit predictive models on past / real-time data
3. Apply & validate models until predictions are accurate
4. Forecast abnormal behavior, risk scores & notable events
5. Surface incidents to Security Ops, who INVESTIGATES & ACTS

15. 15
Business Analytics: Predict Customer Churn
Problem: Customer churn causes big time & money expense
Solution: Build predictive model to forecast possible churn, act pre-emptively & learn
1. Get customer data (set-top boxes, web logs, transaction history)
2. Explore data, and fit predictive models on past / real-time data
3. Apply & validate models until predictions are accurate
4. Forecast churn rate & identify customers likely to churn
5. Surface results to Business Ops, who INVESTIGATES & ACTS

16. 16
Summary: The ML Process
Problem: <Stuff in the world> causes big time & money expense
Solution: Build predictive model to forecast <possible incidents>, act pre-emptively & learn
1. Get all relevant data to problem
2. Explore data, and fit predictive models on past / real-time data
3. Apply & validate models until predictions are accurate
4. Forecast KPIs & notable events associated to use case
5. Surface incidents to X Ops, who INVESTIGATES & ACTS
Operationalize

17. Copyright © 2016 Splunk Inc.
ML with Splunk

18. 18
Splunk built-in ML capabilities
kmeans cluster
outliers / anomalies / anomalydetection
predict

19. 19
Machine Learning in Splunk ITSI
Adaptive Thresholding:
• Learn baselines & dynamic thresholds
• Alert & act on deviations
• Manage for 1000s of KPIs & entities
• Stdev/Avg, Quartile/Median, Range
Anomaly Detection:
• Find “hiccups” in expected patterns
• Catches deviations beyond thresholds
• Uses Holt-Winters algorithm

20. 20
Splunk User Behavior Analytics (UBA)
• ~100% of breaches involve valid credentials (MandiantReport)
• Need to understand normal & anomalous behaviors for ALL users
• UBA detects Advanced Cyberattacks and Malicious Insider Threats
• Lots of ML under the hood:
– Behavior Baselining & Modeling
– Anomaly Detection (30+ models)
– Advanced Threat Detection
• E.g., Data Exfil Threat:
– “Saw this strange login & data transfer
for user mpittman at 3am in China…”
– Surface threat to SOC Analysts

21. 21
ML Toolkit & Showcase – DIY ML
• Splunk Supported framework for building ML Apps
– Get it for free: https://splunkbase.splunk.com/app/2890/
• Leverages Python for Scientific Computing (PSC) add-
on:
– Get it for free: refer to Splunkbasefor your OS version
ê https://splunkbase.splunk.com/app/2881/ to /2884/
– Open-source Python data science ecosystem
– NumPy, SciPy, scitkit-learn, pandas, statsmodels
• Showcase use cases: Predict Hard Drive Failure, Server
Power Consumption, Application Usage, Customer
Churn & more

22. 22
Standard algorithms out of the box:
Clustering: DBSCAN, KMeans, Birch, SpectralClustering
Regression: LinearRegression, RandomForestRegressor, ElasticNet, Ridge, Lasso
Classification: LogisticRegression, RandomForestClassifier, SVM, Naïve Bayes
(GaussianNB, BernoulliNB)
Transformation: PCA, KernelPCA, TFIDF Vectorizer, StandardScaler
Text Analytics: TF-IDF
Feature Extraction: FieldSelector (e.g. Univariate, ANOVA, K-best, etc.)
Implement one of 300+ algorithms by editing Python scripts

23. 23

24. Copyright © 2016 Splunk Inc.
Building ML Apps

25. 25
Analysts Business Users
1. Get Data & Find Decision-Makers
2
IT Users
ODBC
SDK
API
DB Connect
Look-Ups
Ad Hoc
Search
Monitor
and Alert
Reports /
Analyze
Custom
Dashboards
GPS /
Cellular
Devices Networks Hadoop
Servers Applications Online
Shopping Carts
Analysts Business Users
Structured Data Sources
CRM ERP HR Billing Product Finance
Data Warehouse
Clickstreams

26. 26
2. Explore Data, Build Searches & Dashboards
• Start with the Exploratory Data Analysis phase
– “80% of data science is sourcing, cleaning, and preparing the data”
– Tip: leverage ITSI KPIs – lots of domain knowledge
• For each data source, build “data diagnostic” dashboard
– What’s interesting? Throw up some basic charts.
– What’s relevant for this use case?
– Any anomalies? Are thresholds useful?
• Mix data streams & compute aggregates
– Compute KPIs & statistics w/ stats, eventstats, etc.
– Enrich data streams with useful structured data
– stats count by X Y – where X,Y from different sources
– Build new KPIs from what you find

27. 27
3. Fit, Apply & Validate Models
• ML SPL – New grammar for doing ML in Splunk
• fit – fit models based on training data
– [training data] | fit LinearRegression costly_KPI
from feature1 feature2 feature3 into my_model
• apply – apply models on testing and production data
– [testing/production data] | apply my_model
• Validate Your Model (The Hard Part)
– Why hard? Because statistics is hard! Also: model error ≠ real world risk.
– Analyze residuals, mean-square error, goodness of fit, cross-validate, etc.
– Take Splunk’s Analytics & Data Science Education course

28. 28
4. Predict & Act
• Forecast KPIs & predict notable events
– When will my system have a critical error?
– In which service or process?
– What’s the probable root cause?
• How will people act on predictions?
– Is this a Sev 1/2/3 event? Who responds?
– Deliver via Notable Events or dashboard?
– Human response or automated response?
• How do you improve the models?
– Iterate, add more data, extract more features
– Keep track of true/false positives

29. Copyright © 2016 Splunk Inc.
Demo

30. Copyright © 2016 Splunk Inc.
Next Steps

31. 31
Getting started
• Pre-requisite: you must be running Splunk 6.4.x
• Download and install the free ML Toolkit & Showcase!
– https://splunkbase.splunk.com/app/2890/
– https://splunkbase.splunk.com/app/2881/ to /2884/
• Speak to your local SE to discuss ways you could use ML
• Join our local User Group – we’ll be running ML workshops!
– http://www.meetup.com/splunk-melbourne/
• Contact me! (aphillips@splunk.com)

32. Copyright © 2016 Splunk Inc.
Q&A

33. Copyright © 2016 Splunk Inc.
Thank You

34. 34
Example Splunk SPL – Churn Use Case
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=0
| fit LogisticRegression Churn? from "CustServ Calls" "Day Mins" "Eve Mins" into
example_churn_model
| table *Churn*
| `confusionmatrix("Churn?","predicted(Churn?)")`
| listmodels
| summary example_churn_model
| deletemodel example_churn_model
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_churn_model"
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_churn_model"
| `confusionmatrix("Churn?","predicted(Churn?)")`
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_churn_model"
| `classificationstatistics("Churn?", "predicted(Churn?)")`
##### example training using logistic regression and random forest classifier in combination
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=0
| fit LogisticRegression "Churn?" from "CustServ Calls" "Day Mins" "Eve Mins" "Int'l Plan" "Intl
Calls" "Intl Charge" "Intl Mins" "Night Charge" "Night Mins" "VMail Plan" into "LogReg_churn"
| table *Churn*
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=0
| fit RandomForestClassifier "Churn?" from "CustServ Calls" "Day Mins" "Eve Mins" "Int'l Plan"
"Intl Calls" "Intl Charge" "Intl Mins" "Night Charge" "Night Mins" "VMail Plan" into "RF_churn"
| table *Churn*
##### example testing using logistic regression and random forest classifier in combination
| inputlookup churn.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply LogReg_churn as LogReg(Churn?)
| apply RF_churn as RF(Churn?)
| eval priorityscore(Churn?) = if('LogReg(Churn?)'="True.",10,0) + if('RF(Churn?)'="True.",100,0)
+ .1*'Day Charge'
| sort - priorityscore(Churn?)
| fields priorityscore(Churn?) *Churn?* "CustServ Calls" "Day Calls" "Day Charge" Phone State
| eval whattodo = if('priorityscore(Churn?)'>15, "Call them!", null())
| fieldformat "Day Charge" = "$".round('Day Charge')
| search "Churn?"="False."

35. 35
Example Splunk SPL – Malware Use Case| inputlookup firewall_traffic.csv
| inputlookup firewall_traffic.csv
| fit LogisticRegression used_by_malware from bytes_received bytes_sent dest_port dst_ip
has_known_vulnerability packets_received packets_sent receive_time serial_number
session_id src_ip src_port into example_firewall_traffic_model
| table *used_by_malware*
| `confusionmatrix("used_by_malware","predicted(used_by_malware)")`
| listmodels
| summary example_firewall_traffic_model
| deletemodel example_firewall_traffic_model
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_firewall_traffic_model”
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_firewall_traffic_model"
| `confusionmatrix("used_by_malware","predicted(used_by_malware)")`
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply "example_firewall_traffic_model"
| `classificationstatistics("used_by_malware", "predicted(used_by_malware)")`
##### example training using logistic regression and random forest classifier in combination
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=0
| fit LogisticRegression used_by_malware from bytes_received bytes_sent dest_port dst_ip
has_known_vulnerability packets_received packets_sent receive_time serial_number
session_id src_ip src_port into LogReg_used_by_malware
| table *used_by_malware*
| `confusionmatrix("used_by_malware","predicted(used_by_malware)")`
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=0
| fit RandomForestClassifier used_by_malware from bytes_received bytes_sent dest_port
dst_ip has_known_vulnerability packets_received packets_sent receive_time serial_number
session_id src_ip src_port into RF_used_by_malware
| table *used_by_malware*
| `confusionmatrix("used_by_malware","predicted(used_by_malware)")`
##### example testing using logistic regression and random forest classifier in combination
| inputlookup firewall_traffic.csv
| sample partitions=2 seed=1234
| search partition_number=1
| apply LogReg_used_by_malware as LogReg(used_by_malware)
| apply RF_used_by_malware as RF(used_by_malware)
| eval priorityscore(used_by_malware) = if('LogReg(used_by_malware)'="yes",10,0) +
if('RF(used_by_malware)'="yes",100,0) + if(has_known_vulnerability="yes",50,0)
| eval whattodo = if('priorityscore(used_by_malware)'>50, "Investigate!", null())
| fields whattodo priorityscore(used_by_malware) *used_by_malware* receive_time src_ip
serial_number session_id has_known_vulnerability
| sort whattodo