KnowledgeSTUDIO HIGH-PERFORMANCE PREDICTIVE ANALYTICS USING ADVANCED MODELING TECHNIQUES



Similar documents
KnowledgeSEEKER POWERFUL SEGMENTATION, STRATEGY DESIGN AND VISUALIZATION SOFTWARE

KnowledgeSEEKER Marketing Edition

How to Optimize Your Data Mining Environment

Predictive Analytics Software Suite

WHITEPAPER. How to Credit Score with Predictive Analytics

Product recommendations and promotions (couponing and discounts) Cross-sell and Upsell strategies

WHITEPAPER. Creating and Deploying Predictive Strategies that Drive Customer Value in Marketing, Sales and Risk

Make Better Decisions Through Predictive Intelligence

KNIME UGM 2014 Partner Session

Some vendors have a big presence in a particular industry; some are geared toward data scientists, others toward business users.

WebFOCUS RStat. RStat. Predict the Future and Make Effective Decisions Today. WebFOCUS RStat

SAS. Predictive Analytics. Overview. Turning Your Data into Timely Insight for Better, Faster Decision Making. Challenges SOLUTION OVERVIEW

2015 Workshops for Professors

Pentaho Data Mining Last Modified on January 22, 2007

An Introduction to SAS Enterprise Miner and SAS Forecast Server. André de Waal, Ph.D. Analytical Consultant

Data Mining from A to Z: Better Insights, New Opportunities WHITE PAPER

Better planning and forecasting with IBM Predictive Analytics

Angoss Predictive Analytics Software Suite

Turning Data into Actionable Insights: Predictive Analytics with MATLAB WHITE PAPER

WHITE PAPER. Harnessing the Power of Advanced Analytics How an appliance approach simplifies the use of advanced analytics

Predictive Analytics

IBM SPSS Modeler 15 In-Database Mining Guide

SEIZE THE DATA SEIZE THE DATA. 2015

SAS. Predictive Analytics Suite. Overview. Derive useful insights to make evidence-based decisions. Challenges SOLUTION OVERVIEW

How To Make A Credit Risk Model For A Bank Account

IBM SPSS Modeler Premium

Business Intelligence in Microsoft Dynamics AX 2012

Data Science and Business Analytics Certificate Data Science and Business Intelligence Certificate

FundGUARD. On-Demand Sales and Marketing Optimization for Mutual Funds and Wealth Management

INSIGHT NAV. White Paper

Hurwitz ValuePoint: Predixion

Tax Fraud in Increasing

Demonstration of SAP Predictive Analysis 1.0, consumption from SAP BI clients and best practices

A fast, powerful data mining workbench designed for small to midsize organizations

White Paper. Redefine Your Analytics Journey With Self-Service Data Discovery and Interactive Predictive Analytics

Building and Deploying Customer Behavior Models

IBM SPSS Modeler 14.2 In-Database Mining Guide

Harnessing the power of advanced analytics with IBM Netezza

Data Mining + Business Intelligence. Integration, Design and Implementation

SAP Solution Brief SAP HANA. Transform Your Future with Better Business Insight Using Predictive Analytics

Business Intelligence and intuitive reporting in one comprehensive solution

CART 6.0 Feature Matrix

Using Adaptive Random Trees (ART) for optimal scorecard segmentation

IBM SPSS Direct Marketing

High-Performance Analytics

BUSINESS INTELLIGENCE

Azure Machine Learning, SQL Data Mining and R

Improve Model Accuracy with Unstructured Data

STATISTICA. Financial Institutions. Case Study: Credit Scoring. and

The Predictive Data Mining Revolution in Scorecards:

Easily Identify the Right Customers

Practical Data Science with Azure Machine Learning, SQL Data Mining, and R

Data Mining. Dr. Saed Sayad. University of Toronto

Data Mining with MicroStrategy

Overview, Goals, & Introductions

Empowering the Masses with Analytics

IBM SPSS Modeler Professional

Discovering, Not Finding. Practical Data Mining for Practitioners: Level II. Advanced Data Mining for Researchers : Level III

Oracle Advanced Analytics 12c & SQLDEV/Oracle Data Miner 4.0 New Features

The QlikView Business Discovery platform

An In-Depth Look at In-Memory Predictive Analytics for Developers

In-Database Analytics

Solve Your Toughest Challenges with Data Mining

IBM SPSS Modeler Professional

Databricks. A Primer

Today, the world s leading insurers

Get to Know the IBM SPSS Product Portfolio

Beyond Traditional Management Reporting IBM Corporation

Making confident decisions with the full spectrum of analysis capabilities

SAP Predictive Analytics: An Overview and Roadmap. Charles Gadalla, SESSION CODE: 603

Confidently Anticipate and Drive Better Business Outcomes

Application of SAS! Enterprise Miner in Credit Risk Analytics. Presented by Minakshi Srivastava, VP, Bank of America

IBM's Fraud and Abuse, Analytics and Management Solution

SAP Predictive Analysis: Strategy, Value Proposition

Predictive Analytics Powered by SAP HANA. Cary Bourgeois Principal Solution Advisor Platform and Analytics

Assessing campaign management technology

Solve your toughest challenges with data mining

APPROACHABLE ANALYTICS MAKING SENSE OF DATA

BIRT Performance Scorecard Root Cause Analysis and Data Visualization The Path to Higher Performance

Databricks. A Primer

Data Analysis with Various Oracle Business Intelligence and Analytic Tools

The Power of Predictive Analytics

Microsoft Services Exceed your business with Microsoft SharePoint Server 2010

Solve your toughest challenges with data mining

TDWI Best Practice BI & DW Predictive Analytics & Data Mining

Three steps to put Predictive Analytics to Work

III JORNADAS DE DATA MINING

Three Open Blueprints For Big Data Success

Transcription:

HIGH-PERFORMANCE PREDICTIVE ANALYTICS USING ADVANCED MODELING TECHNIQUES Translating data into business value requires the right data mining and modeling techniques which uncover important patterns within a dataset, identify good predictors, and produce accurate and stable predictions. Advanced predictive modeling can help companies understand customers, predict their behaviour, drive strategic decision making and ultimately improve business performance. In order to get started with predictive analysis via modeling techniques, users must first identify the business problem. Some of the most common business applications in which predictive modeling techniques play a crucial role are risk, fraud detection, marketing, and sales. In these, business problems can range from creating the best cross-sell or upsell offer that resonates with a particular audience to identifying fraudulent activity and being able to assess which customers are most likely to default. So how do businesses apply the right models and build effective strategies to solve a perceived business problem? Advanced Modeling Application Designed to Address Business Challenges Predictive analytics applications are complex and have a long learning curve Many analytics applications require statistical programming skills Expensive training and consulting is often required with predictive analytics software Many predictive applications lack scorecard building capabilities Decision Trees used in most predictive analytics applications lack interactivity and rich graphical interface Analytics software products on the market are expensive and do not provide value for money spent Traditional BI and reporting tools cannot provide predictive insights to improve business decisions Most predictive models are not easy to explain to business stakeholders Lack of integration and model deployment options for data exchange and exporting models to other environments Lack of streamlined way to use model predictions for building strategies in a single application results in lower productivity 1

Widely known for its industry-leading, patented Decision Trees, Strategy Trees, and its workflow and wizard-driven graphical user interface, KnowledgeSTUDIO is an advanced data mining and predictive analytics workbench for high-performance business analysts and quantitative analysts. The application provides a robust set of capabilities for model development and deployment for a wide range of applications and use cases. With advanced predictive modeling capabilities, KnowledgeSTUDIO offers comprehensive and advanced predictive analytics for all phases of the data mining cycle with focus on model development and deployment cycle. The high-performance visual environment provides users with an intuitive workflow and wizard-driven graphical user interface which helps users build effective models without having to code. Quantitative analysts have access to more complex functionality with configurable settings for fine tuning advanced model parameters in order to accommodate models of any complexity. KnowledgeSTUDIO Features No programming skills required for building advanced models Just use the GUI! Advanced modeling techniques: linear and logistic regression, scorecards, neural networks, cluster analysis,market basket analysis, and factor analysis Best-in-Class Decision Trees for segmentation analysis and modeling Ensemble Trees built with Bagging, Boosting, and Random Forest algorithms Unique Strategy Trees for prescriptive analytics Weight of Evidence Optimizer automatically optimizes bins for WOE variables with the click of a button 50% time savings! Segment Viewer for the identification of candidate predictive attributes Angoss Strategy Trees are great... much better than having to learn code Data Preparation and Profiling for easy extraction, manipulation and transformation of data for modeling Easy-to-Use GUI with a powerful automated workflow canvas 2

KnowledgeSTUDIO is unique in its universality and efficiency. It is used by customers across a multitude of industries and departments such as credit risk, fraud, marketing, sales, and CRM analytics. The easy-to-use application with an automated canvas for building, displaying, refreshing and reusing analytic workflows drastically increases data mining efficiency by more than 40%. KnowledgeSTUDIO Differentiators Features Continued Model Analyzer featuring versatile model evaluation and comparison metrics, and charts Model Deployment via scoring and code generation for easy integration with other platforms Advanced visual analytics available via Tableau integration The easy-to-use graphical user interface with automated workflow is rated to be one of the best in its category by industry analysts and users. Point-and-click, workflow and wizard-driven interface is fast and easy to learn. The powerful automated canvas for building, displaying, refreshing, and reusing analytic workflow enables the creation of a workflow in minutes with a few simple drag-and-drop steps eliminating the need to write code thereby increasing efficiency. Segment Viewer can be used as a qualitative assessment tool to identify candidate predictor variables for use in modeling. It displays the distribution charts of the independent variables segmented by the categories of the dependent (target) variable. This allows for easy identification of variables with markedly different distributions across the target categories. Scorecard Building Highlights Workflow automation of the entire scorecard building process Scorecards are built on the basis of logistic regression models Automated generation of Weight of Evidence (WOE) variables using decision tree interface and advanced WOE binning optimization algorithms Reject Inference methods for improving the accuracy of application scorecards, including: proportional, hard cut-off, parceling, and fuzzy augmentation 3

Advanced predictive modeling techniques including advanced scorecard development, linear and logistic regression, decision trees, ensemble trees, neural networks, and unsupervised learning techniques such as market basket analysis, cluster analysis, and factor analysis. Scorecard building capabilities, offered in KnowledgeSTUDIO, automate the traditionally tedious, manual task of coarse classing (binning) with a flexible and intelligent Weight of Evidence Optimizer (WOE) reducing time spent on this task by up to 50%. The WOE binning editor uses the decision tree interface and specialized optimization algorithms that are unique to Angoss. Reject inference methods for scorecard development include: Proportional Assignment, Hard Cutoff, Parceling, and Fuzzy Augmentation (See Scorecard building highlights call-out box pg. 3 & 4) Best-in-Class Decision Trees outperfom the competition, allow interactive automatic and manual growth and visually display results for easy interpretation. Scorecard Building Highlights Continued Scorecard scaling is performed through an easy-to-use wizard that includes the option of incorporating Base Score into Score points Population Stability Report for measuring and monitoring the scorecard performance statistics Auto-adjustment of boundaries of Weight of Evidence bins while training the logistic regression model SAS and SQL code generation for entire workflow or any segment Scorecards can be scored on Angoss datasets or exported to SAS, SQL or PMML code Scorecard performance can be assessed using Model Analyzer Scorecards are generated in standard tabular form that can be exported to Microsoft Office applications and in the form of a SQL expression 4

Unique Strategy Trees are the first of its kind for building and deploying prescriptive strategies. Users are able to combine customer segments, scores, business rules, and calculations, and apply user-defined treatments and actions in order to support decisioning and the development of business strategies. Model Analyzer provides an extensive set of model validation and comparison features as well as model deployment tools. Model performance can be evaluated and compared using lift and cumulative lift charts, relative operating characteristic curves and other charts. This feature allows users to provide management with estimated response rates and lift for each individual model. The integration of Tableau visual analytics further enhances Angoss visualization capability by providing users with access to Tableau dashboards directly from within the Angoss workflow. Automatic SAS code generation for the entire project workflow or any of its stages. Easy integration into other analytical environments enables data import and export to/from Text, Microsoft Excel, SAS, SPSS, R, Hadoop, and databases via ODBC. Automatic code generation for Angoss models can be deployed in other rules engines and environments. Code types include SQL, SAS, SPSS, Java, PMML, XML, and structured English code. KnowledgeSTUDIO is the most robust and easy-to-use tool designed to help advanced quantitative analysts and business analysts to explore and segment data, develop, and deploy models and strategies in varied business and application areas.... Visual Analytics via Angoss native charts, graphs, and reports make every stage of the data mining process easily interpretable and presentable. KnowledgeSTUDIO is a scalable application that grows with your organizational needs and provides users with the flexibility for added functionality such as: Improve collaboration between users and 5

departments, with the integration of the languages of R and Python, all within a single environment. To enhance the user experience, the programming languages can be directly embedded in the Angoss workflows. In-Database analytics for direct data mining on data stored in a database as opposed to working on a copy of the data. Real-time scoring engine delivers intelligent, real-time scores or recommendations to operational systems during customer interactions to support timely decision making and speed-to-action. Advanced data prep functionality in the language of SAS via KnowledgeCORE integration. Users can write programs in the language of SAS, SQL, Python and R and use custom programs and pre-built functional nodes within Angoss analytic workflows. About Angoss Angoss is a global leader in delivering predictive analytics to businesses looking to improve performance across risk, marketing and sales. With a suite of big data analytics software solutions and consulting services, Angoss delivers powerful approaches that provide you with a competitive advantage by turning your information into actionable business decisions. Many of the world s leading organizations in financial services, insurance, retail and high tech rely on Angoss to grow revenue, increase sales productivity and improve marketing effectiveness while reducing risk and cost. Headquartered in Toronto, Canada, with offices in the United States, United Kingdom and Singapore, Angoss serves customers in over 30 countries worldwide. For more information, visit www.angoss.com. Model management capability via KnowledgeMANAGER for modelers looking to store, compare, monitor, and score models to ensure that only the best models are put into production. Optimization capability via InsightOPTIMIZER, enables organizations to optimize their sophisticated predictive models, subject to business constraints and rules, in order to minimize loss and maximize profits Corporate Headquarters 111 George Street, Suite 200 Toronto, Ontario M5A 2N4 Canada Tel: 416-593-1122 European Headquarters Enigma House 30b Alan Turing Road The Surrey Research Park Guildford, Surrey GU2 7AA Tel: +44 (0) 1483-661-661 6