SmartCare. SPSS Workshop. Rick Durham - North American Advanced Analytics Channel Team IBM Corporation. Date: 5/28/2014

Size: px
Start display at page:

Download "SmartCare. SPSS Workshop. Rick Durham - North American Advanced Analytics Channel Team IBM Corporation. Date: 5/28/2014"

Transcription

1 SPSS Workshop Key Presenter Rick Durham - North American Advanced Analytics Channel Team Date: 5/28/2014

2 Agenda What is Predictive Analytics? What is the architecture of the IBM/SPSS technology stack? What is the basic methodology of Predictive Analytics (CRISP- DM)? What is SPSS Modeler and how is it used to build Healthcare Models? Readmissions Complications Cost Predictions Q/A Wrap-up 2

3 What is predictive analytics? Predictive Analytics helps connect data to effective action by drawing reliable conclusions about current conditions and future events Gareth Herschel, Research Director, Gartner Group 3

4 IBM SPSS Predictive Analytics Enhances other IBM Technology Predictive Customer Analytics Acquire Grow Retain Predictive Operational Analytics Manage Maintain Maximize Predictive Threat & Fraud Analytics Monitor Detect Control Data Collection Social Media Analytics Statistics Modeler Decision Management Collaboration and Deployment Services IBM Research Etc 4

5 IBM SPSS Modeler High-performance data mining and text analytics workbench Utilizes structured and unstructured data Creates predictive analytics for data driven decision making Enables superior outcomes and positive ROI 5

6 IBM SPSS Modeler Easy-to-use, interactive interface without the need for programming Automated modeling and data preparation capabilities Access ALL data structured and unstructured from disparate sources Natural Language Processing (NLP) to extract concepts and sentiments in text Entity Analytics ensures the quality of the data and results in more accurate models Leverage existing investment in Cognos, Netezza, InfoSphere and System Z 6

7 7 IBM SPSS Analytic Server Delivers fast time to solution for predictive analytics of big data Visual, easy to use interface abstracts analysts & line of business users from complexities of big data systems Data-centric architecture ensures scalability & performance Enables organizations to Empower analysts to create & deploy predictive analytics over big data without technical skills or coding

8 SQL / UDF IBM SPSS Modeler Stream File Big Data Request IBM SPSS Analytic Server Relational Database Modeler Client Modeler Server Hadoop Job IBM SPSS Analytic Catalyst Analytics Analytic Catalyst Tablet Client Analytic Catalyst Browser Client IBM InfoSphere BigInsights & Other Hadoop Distributions 8 Modeler Server utilizes Analytic Server for Big Data Analysts define analysis in a familiar & accessible workbench to conduct analysis, modeling & scoring over high volumes of varied data Federation of heterogeneous data sources to use legacy & external data in model building & scoring Transformations, sampling & write-back of output to big data systems Next generation Analytic Catalyst clients utilize Analytic Server for automated analysis

9 CRISP-DM Phases 6 Phases Business Understanding Data Understanding Data Preparation Modeling Evaluation Deployment Not strictly ordered Several possible entry points into the loop Reflects iterative nature of data mining 9 9

10 IBM SPSS Modeler Modeling Techniques Technique Algorithms Usage Classification (or prediction) Auto Classifiers, Decision Trees, Logistic, SVM, Time Series, etc Used to predict group membership (ie will this employee leave?) or a number (ie how many widgets will I sell?) Segmentation Auto Clustering, K- means, etc. Anomoly detection Used to classify data points into groups that are internally homogenous and externally heterogeneous. Identify cases that are unusual Association APRIORI, Carma, Sequence Used to find events that occur together or in a sequence (ie market basket). 10

11 Predictive Analytics Building Healthcare Models Using SPSS Modeler 11 11

12 12 12