df14 building machine learning systems with apex
DESCRIPTION
Slide deck from the Dreamforce 20134 talk "Building Machine Learning Systems with Apex". Includes links to github code repository and contact details for speakers.TRANSCRIPT
![Page 1: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/1.jpg)
Building Machine Learning Systems in ApexJen Wyher
Technical Architect
@jenwyher
Paul Battisson
Technical Architect
@pbattisson
![Page 2: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/2.jpg)
Safe HarborSafe harbor statement under the Private Securities Litigation Reform Act of 1995: This presentation may contain forward-looking statements that involve risks, uncertainties, and assumptions. If any such uncertainties materialize or if any of the assumptions proves incorrect, the results of salesforce.com, inc. could differ materially from the results expressed or implied by the forward-looking statements we make. All statements other than statements of historical fact could be deemed forward-looking, including any projections of product or service availability, subscriber growth, earnings, revenues, or other financial items and any statements regarding strategies or plans of management for future operations, statements of belief, any statements concerning new, planned, or upgraded services or technology developments and customer contracts or use of our services. The risks and uncertainties referred to above include – but are not limited to – risks associated with developing and delivering new functionality for our service, new products and services, our new business model, our past operating losses, possible fluctuations in our operating results and rate of growth, interruptions or delays in our Web hosting, breach of our security measures, the outcome of any litigation, risks associated with completed and any possible mergers and acquisitions, the immature market in which we operate, our relatively limited operating history, our ability to expand, retain, and motivate our employees and manage our growth, new releases of our service and successful customer deployment, our limited history reselling non-salesforce.com products, and utilization and selling to larger enterprise customers. Further information on potential factors that could affect the financial results of salesforce.com, inc. is included in our annual report on Form 10-K for the most recent fiscal year and in our quarterly report on Form 10-Q for the most recent fiscal quarter. These documents and others containing important disclosures are available on the SEC Filings section of the Investor Information section of our Web site. Any unreleased services or features referenced in this or other presentations, press releases or public statements are not currently available and may not be delivered on time or at all. Customers who purchase our services should make the purchase decisions based upon features that are currently available. Salesforce.com, inc. assumes no obligation and does not intend to update these forward-looking statements.
![Page 3: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/3.jpg)
Jennifer WyherTechnical Architect at Mavens Consulting
@jenwyher
![Page 4: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/4.jpg)
Paul BattissonTechnical Architect at Mavens Consulting
@pbattisson
Summer ’14 Force.com MVP
@forcedotcomcast
![Page 5: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/5.jpg)
Mavens Consulting• Preferred Life Sciences implementation
partner for salesforce.com and Veeva• 60+ consultants located across North America
and Europe• 12 Mavens in attendance at #Dreamforce14,
speaking in 7 different technical sessions
![Page 6: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/6.jpg)
Use to introduce a
demo, video, Q&A, etc.
Baseline setting -Who has worked on a machine learning system before?
![Page 7: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/7.jpg)
What is Machine Learning?Autonomous vehicles
Spam filtering
Search engines
Data analysis
![Page 8: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/8.jpg)
Use to introduce a
demo, video, Q&A, etc.
“Field of study that gives computers the ability to learn without being explicitly programmed”
- Arthur Samuel, 1959
![Page 9: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/9.jpg)
Unsupervised
System determines classification parameters
and groups
![Page 10: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/10.jpg)
Supervised
You provide the system with some
guidance
$200k
$120k $100k $180k $110k
$???
![Page 11: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/11.jpg)
Why Apex?• Governor limits make it hard to do long
running or big jobs with apex• Showing the power of the platform
![Page 12: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/12.jpg)
K-Means Clustering
• Account targeting• Medical diagnosis aid• Data segmentation
“given a group of m differentdata points derive k clusters
of related items”
![Page 13: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/13.jpg)
The Algorithm
• Initialize K centroids• Assign each training example to it’s “nearest” centroid
• Reset the centroid as the mean of all assigned examples
• Repeat until the centroid is fixed
![Page 14: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/14.jpg)
The Algorithm
• Initialize K centroids• Assign each training example to it’s “nearest” centroid
• Reset the centroid as the mean of all assigned examples
• Repeat until the centroid is fixed
![Page 15: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/15.jpg)
How we thought it would work
![Page 16: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/16.jpg)
How it does work
![Page 17: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/17.jpg)
Demo
![Page 18: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/18.jpg)
The Need For Speed• Chained Batches
– Batches creating batches
• Speedier loops– Remove around 90% of CPUTime– See http://goo.gl/mR5GZe
• JSON serialize/deserialize and attachments– Quick and effective way of storing data– Attachments have much larger limit (around 10x the amount of data)
• Running totals (stateful batch)– Saves repeated loops
• Javascript Remoting for charting– Loading so many attachments destroys heap size– Use remoting to load attachments for display asynchronously
![Page 19: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/19.jpg)
Future Ideas• Recommendation Engines
– Content– Products/services
• Neural Networks– Lots of number processing– Chaining will be key
• Real time sites recommendations– Think Amazon recommendations
![Page 20: Df14 Building Machine Learning Systems with Apex](https://reader035.vdocuments.net/reader035/viewer/2022081413/547da9cdb4af9f18128b473d/html5/thumbnails/20.jpg)
@jenwyher
@pbattisson
@mavens
https://github.com/pbattisson/DF14-Building-Machine-Learning-Systems-With-Apex