SlideShare a Scribd company logo

WSO2 Machine Learner - Product Overview

•
•
WSO2
WSO2

WSO2 Machine Learner takes data one step further, pairing data gathering and analytics with predictive intelligence: this helps you understand not just the present, but to predict scenarios and generate solutions for the future.

1 of 28
Download to read offline
  WSO2  Machine  Learner   1.1.0        
WSO2  Analy+cs  Pla/orm       WSO2  Analy5cs  Pla8orm  uniquely  combines  simultaneous  real-­‐ 2me  and  batch  analysis  with  predic2ve  analy2cs  to  turn  data   from  IoT,  mobile  and  Web  apps  into  ac5onable  insights     2  
WSO2  Analy+cs  Pla/orm   3  
WSO2  Advantages   4  
Highly  Pluggable  Architecture   5  
Toolboxes  for  Extensibility   6   +   Toolboxes  =     Industry  or  domain  specific  analy7cs   Toolboxes:     •  Fraud  and  Anomaly  Detec+on-­‐    Supports  fraud  and  anomaly  detec7on  through  sta7c    rules,  Markov   chains,  and  scoring.   •  GIS  Data  Monitoring  -­‐  Can  take  any  data  stream  tagged  with  geographical  loca7ons  and  support   visualiza7ons  of  that  data  in  a  map.   •  Ac+vity  Monitoring-­‐  Lets  users  correlate  events  related  to  the  same  transac7on  in  order  to  visualize,   analyze,  and  write  queries  on  top  of  those  ac7vi7es.  
Edge  Analy+cs-­‐Mobile  and  IoT  Streams   7   Event  correla2on/filtering  available  at  the  edge  
High  Level  Languages   •  For  both  batch  and  real-­‐7me,  we  provide  structured  ,  SQL-­‐like  query  languages.   •  No  Java  programming  is  required   •  Lowers  the  adop7on  entry  point.   •  Batch  analy7cs  relies  on  SparkSQL.   •  Real  Time  analy7cs  implemented  through  WSO2  owned  solu7on  Siddhi   8  
Real+me  analy+cs  with  Siddhi   •  ThroRling  &  Blacklis7ng  users   define  stream  RequestStream  (  correla7onID  string,  serviceID  string,userID  string,  tear   string,  requestTime  long,  ...  )  ;   define  table  BlacklistedUserTable(userID  string,7me  long,requestCount  long);     from  RequestStream[tear==‘BRONZE’]#window.7me(1  min)   select  userID,  requestTime  as  7me,  count(correla7onID)  as  requestCount   group  by  userID   having  up  requestCount  >  5   insert  into  BlacklistedUserTable  ;   9  
Batch  Analy+cs  with  Spark  SQL     create temporary table product_data using carbonanalytics options (schema …) create temporary table products using carbonanalytics options (schema …) insert into products select product_name from product_data group by … 10  
Case  Studies   1
Smart  Home   •  DEBS  (Distributed  Event  Based  Systems)  is  a  premier  academic   conference,  which  post  yearly  event  processing  challenge  ( hRp://www.cse.iitb.ac.in/debs2014/?page_id=42)     •  Smart  Home  electricity  data:  2000  sensors,  40  houses,  4  Billion  events   •  We  posted  fastest  single  node  solu7on  measured  (400K  events/sec)   and  close  to  one  million  distributed  throughput.     •  WSO2  CEP  based  solu7on  is  one  of  the  four  finalists  (with  Dresden   University  of  Technology,  Fraunhofer  Ins7tute,  and  Imperial  College   London)   •  Only  generic  solu7on  to  become  a  finalist   12  
Healthcare  Data  Monitoring   •  Allows  to  search/visualize/analyze  healthcare  records  (HL7)    across  20  hospitals  in   Italy   •  Used  in  combina7on  with  WSO2  ESB   •  Custom  toolbox  tailored  to  customer’s  requirement  (  to  replace  exis7ng  system)         •      13  
Cloud  IDE  Analy+cs   •  Custom  solu7on  created  in  partnership  with  Codenvy  to  bring  analy7cs  to  Codenvy   management  team  and  its  customers   •  Developed  in  less  than  a  month,  with  a  custom  plug-­‐in  to  MongoDB.   •  Deployed  in  the  codenvy.com  plamorm.   14  
Addi+onal  Customers  Use  Cases     •  Cisco  (BAM  +  CEP)  -­‐  OEM,  Healthcare,  Parking  Monitoring  (see  Solu7on  paRerns  based   approach  to  rapidly  create  IoE  solu7ons  across  industries,     •  hRp://us14.wso2con.com/videos/#Coumara-­‐Radja   •  Used  by  a  Large  Scale  IoT  System  Provider  for  use  cases  including  Vehicle  tracking,    Smart   City,  Building  Monitoring  (CEP)   •  See  “Internet  of  Big  Things:  The  Story  of  Pacific  Controls,  hRp://us14.wso2con.com/videos/#Sajaad-­‐Chaudry”     •  Transac7on  Monitoring  in  a  Large  Bank  (CEP)   •  Knowledge  Mining  and  tracking  Prospec7ve  Customers  through  Natural  Language  data   sources  (CEP)   •  CEP  Embedded  in  edge  Devices     •  See  WSO2Con  2013  -­‐  Keynote:Emerging  Founda7ons  of  Next-­‐Genera7on  Business  Systems   hRps://www.youtube.com/watch?v=7CyG3JKUxWw   •  ThroRling  and  Anomaly  Detec7on  by  Group  of  Telecom  Companies     15  
WSO2  Machine  Learner    (Technical  Overview)     1
WSO2  Machine  Learner   17  
Overview   18   o  Open source Machine Learning (ML) tool o  Scalable way to perform machine learning o  Visually explore uploaded data sets o  Support for various machine learning algorithms o  Metrics to evaluate and compare built ML models. o  Ability to export ML models o  Extensions for real-time predictions o  REST API to expose all features i.e. ML jobs are scriptable
Func+onality   19   o  Manage and explore your data o  Analyze the data using machine learning algorithms o  Build machine learning models o  Compare and manage generated machine learning models o  Predict using the built models
Manage  Data  set   20   o  Supported data sources o CSV/TSV files from local file systems. o Files from HDFS. o Tables from WSO2 Data Analytics Server o  Supports data set versioning. o Version data collected overtime from the same data set o  Generate models from the different versions. o  Manage datasets based on projects ,users.
Pre-­‐process  &  Explore  Data   21   o  Find key details from feature set o  Scatter plots to understand relationship between feature set o Supported graphs: o Scatter plots, Parallel sets,Trellis charts, Cluster diagram, Histogram o  Missing value handling with mean imputation and discard
Analysis  with  ML  Algorithm   22   o  Supports deep learning o  Supports supervised and unsupervised learning. o  Includes algorithms for numerical prediction, classification and clustering. o  Supports anomaly detection algorithm. o  Supports recommendation with Collaborative Filtering Recommendation Algorithm
Analysis  with  ML  Algorithm   23   o  Includes algorithms for numerical prediction, classification and clustering. Numerical prediction Linear Regression, Ridge Regression, Lasso Regression Classification Logistic Regression, Naive Bayes, Decision Tree, Random Forest and Support Vector Machines Clustering K-Means
Model  Evalua+on  &  Comparison   24   o  Evaluate generated models based on metrics o Accuracy o Area under ROC curve o Confusion Matrix o Predicted vs. Actual graphs o Feature importance o  Compare models generated from different analysis. o  Set fractions for training data
Integra+on  of  ML  Models   25   o  Models can be used via main transaction flow (WSO2 ESB) or data analysis flow (WSO2 CEP) o  Supports PMML for interoperability.
Deployment  Op+ons   26   o  Stand alone mode o  With external Spark Cluster o  With WSO2 DAS as external Spark Cluster
Run  Yourself  or  let  WSO2  Run  it  for  you   27   Self-Hosted •  Your operations team maintains the deployment with production support from WSO2 WSO2 Managed Cloud •  WSO2 Operations team runs the deployment in a dedicated environment in AWS datacenter of your choice •  Includes monitoring, backups, patches, updates •  Financially backed SLA on uptime and response time
Thank  You!   Download  WSO2  Machine  Learner  at:     h]p://wso2.com/products/machine-­‐learner/  

More Related Content

WSO2 Machine Learner - Product Overview

  • 1.   WSO2  Machine  Learner   1.1.0        
  • 2. WSO2  Analy+cs  Pla/orm       WSO2  Analy5cs  Pla8orm  uniquely  combines  simultaneous  real-­‐ 2me  and  batch  analysis  with  predic2ve  analy2cs  to  turn  data   from  IoT,  mobile  and  Web  apps  into  ac5onable  insights     2  
  • 6. Toolboxes  for  Extensibility   6   +   Toolboxes  =     Industry  or  domain  specific  analy7cs   Toolboxes:     •  Fraud  and  Anomaly  Detec+on-­‐    Supports  fraud  and  anomaly  detec7on  through  sta7c    rules,  Markov   chains,  and  scoring.   •  GIS  Data  Monitoring  -­‐  Can  take  any  data  stream  tagged  with  geographical  loca7ons  and  support   visualiza7ons  of  that  data  in  a  map.   •  Ac+vity  Monitoring-­‐  Lets  users  correlate  events  related  to  the  same  transac7on  in  order  to  visualize,   analyze,  and  write  queries  on  top  of  those  ac7vi7es.  
  • 7. Edge  Analy+cs-­‐Mobile  and  IoT  Streams   7   Event  correla2on/filtering  available  at  the  edge  
  • 8. High  Level  Languages   •  For  both  batch  and  real-­‐7me,  we  provide  structured  ,  SQL-­‐like  query  languages.   •  No  Java  programming  is  required   •  Lowers  the  adop7on  entry  point.   •  Batch  analy7cs  relies  on  SparkSQL.   •  Real  Time  analy7cs  implemented  through  WSO2  owned  solu7on  Siddhi   8  
  • 9. Real+me  analy+cs  with  Siddhi   •  ThroRling  &  Blacklis7ng  users   define  stream  RequestStream  (  correla7onID  string,  serviceID  string,userID  string,  tear   string,  requestTime  long,  ...  )  ;   define  table  BlacklistedUserTable(userID  string,7me  long,requestCount  long);     from  RequestStream[tear==‘BRONZE’]#window.7me(1  min)   select  userID,  requestTime  as  7me,  count(correla7onID)  as  requestCount   group  by  userID   having  up  requestCount  >  5   insert  into  BlacklistedUserTable  ;   9  
  • 10. Batch  Analy+cs  with  Spark  SQL     create temporary table product_data using carbonanalytics options (schema …) create temporary table products using carbonanalytics options (schema …) insert into products select product_name from product_data group by … 10  
  • 12. Smart  Home   •  DEBS  (Distributed  Event  Based  Systems)  is  a  premier  academic   conference,  which  post  yearly  event  processing  challenge  ( hRp://www.cse.iitb.ac.in/debs2014/?page_id=42)     •  Smart  Home  electricity  data:  2000  sensors,  40  houses,  4  Billion  events   •  We  posted  fastest  single  node  solu7on  measured  (400K  events/sec)   and  close  to  one  million  distributed  throughput.     •  WSO2  CEP  based  solu7on  is  one  of  the  four  finalists  (with  Dresden   University  of  Technology,  Fraunhofer  Ins7tute,  and  Imperial  College   London)   •  Only  generic  solu7on  to  become  a  finalist   12  
  • 13. Healthcare  Data  Monitoring   •  Allows  to  search/visualize/analyze  healthcare  records  (HL7)    across  20  hospitals  in   Italy   •  Used  in  combina7on  with  WSO2  ESB   •  Custom  toolbox  tailored  to  customer’s  requirement  (  to  replace  exis7ng  system)         •      13  
  • 14. Cloud  IDE  Analy+cs   •  Custom  solu7on  created  in  partnership  with  Codenvy  to  bring  analy7cs  to  Codenvy   management  team  and  its  customers   •  Developed  in  less  than  a  month,  with  a  custom  plug-­‐in  to  MongoDB.   •  Deployed  in  the  codenvy.com  plamorm.   14  
  • 15. Addi+onal  Customers  Use  Cases     •  Cisco  (BAM  +  CEP)  -­‐  OEM,  Healthcare,  Parking  Monitoring  (see  Solu7on  paRerns  based   approach  to  rapidly  create  IoE  solu7ons  across  industries,     •  hRp://us14.wso2con.com/videos/#Coumara-­‐Radja   •  Used  by  a  Large  Scale  IoT  System  Provider  for  use  cases  including  Vehicle  tracking,    Smart   City,  Building  Monitoring  (CEP)   •  See  “Internet  of  Big  Things:  The  Story  of  Pacific  Controls,  hRp://us14.wso2con.com/videos/#Sajaad-­‐Chaudry”     •  Transac7on  Monitoring  in  a  Large  Bank  (CEP)   •  Knowledge  Mining  and  tracking  Prospec7ve  Customers  through  Natural  Language  data   sources  (CEP)   •  CEP  Embedded  in  edge  Devices     •  See  WSO2Con  2013  -­‐  Keynote:Emerging  Founda7ons  of  Next-­‐Genera7on  Business  Systems   hRps://www.youtube.com/watch?v=7CyG3JKUxWw   •  ThroRling  and  Anomaly  Detec7on  by  Group  of  Telecom  Companies     15  
  • 16. WSO2  Machine  Learner    (Technical  Overview)     1
  • 18. Overview   18   o  Open source Machine Learning (ML) tool o  Scalable way to perform machine learning o  Visually explore uploaded data sets o  Support for various machine learning algorithms o  Metrics to evaluate and compare built ML models. o  Ability to export ML models o  Extensions for real-time predictions o  REST API to expose all features i.e. ML jobs are scriptable
  • 19. Func+onality   19   o  Manage and explore your data o  Analyze the data using machine learning algorithms o  Build machine learning models o  Compare and manage generated machine learning models o  Predict using the built models
  • 20. Manage  Data  set   20   o  Supported data sources o CSV/TSV files from local file systems. o Files from HDFS. o Tables from WSO2 Data Analytics Server o  Supports data set versioning. o Version data collected overtime from the same data set o  Generate models from the different versions. o  Manage datasets based on projects ,users.
  • 21. Pre-­‐process  &  Explore  Data   21   o  Find key details from feature set o  Scatter plots to understand relationship between feature set o Supported graphs: o Scatter plots, Parallel sets,Trellis charts, Cluster diagram, Histogram o  Missing value handling with mean imputation and discard
  • 22. Analysis  with  ML  Algorithm   22   o  Supports deep learning o  Supports supervised and unsupervised learning. o  Includes algorithms for numerical prediction, classification and clustering. o  Supports anomaly detection algorithm. o  Supports recommendation with Collaborative Filtering Recommendation Algorithm
  • 23. Analysis  with  ML  Algorithm   23   o  Includes algorithms for numerical prediction, classification and clustering. Numerical prediction Linear Regression, Ridge Regression, Lasso Regression Classification Logistic Regression, Naive Bayes, Decision Tree, Random Forest and Support Vector Machines Clustering K-Means
  • 24. Model  Evalua+on  &  Comparison   24   o  Evaluate generated models based on metrics o Accuracy o Area under ROC curve o Confusion Matrix o Predicted vs. Actual graphs o Feature importance o  Compare models generated from different analysis. o  Set fractions for training data
  • 25. Integra+on  of  ML  Models   25   o  Models can be used via main transaction flow (WSO2 ESB) or data analysis flow (WSO2 CEP) o  Supports PMML for interoperability.
  • 26. Deployment  Op+ons   26   o  Stand alone mode o  With external Spark Cluster o  With WSO2 DAS as external Spark Cluster
  • 27. Run  Yourself  or  let  WSO2  Run  it  for  you   27   Self-Hosted •  Your operations team maintains the deployment with production support from WSO2 WSO2 Managed Cloud •  WSO2 Operations team runs the deployment in a dedicated environment in AWS datacenter of your choice •  Includes monitoring, backups, patches, updates •  Financially backed SLA on uptime and response time
  • 28. Thank  You!   Download  WSO2  Machine  Learner  at:     h]p://wso2.com/products/machine-­‐learner/