Showing posts with label SPSS. Show all posts
Showing posts with label SPSS. Show all posts

Sunday, March 22, 2015

The 2015 Rexer Analytics Data Miner Survey has launched!

The 2015 Rexer Analytics Data Miner Survey has launched. The founder of Rexer Analytics, Karl Rexer, is regarded by Deep-data-mining.com as one of 10 Most Influential People in Data Analytics.

Analytic Professionals — Share your views: Participate in the 2015 Data Miner Survey

Data Analysts, Predictive Modelers, Data Scientists, Data Miners, and all other types of analytic professionals, students, and academics: Please participate in the 2015 Rexer Analytics Data Miner Survey.


Survey Link: www.rexeranalytics.com/Data-Miner-Survey-2015-Intro.html
Access Code: CL72L4 — It is OK to share this Access Code with others: It can be used by multiple people.
Survey results will be unveiled at the Fall-2015 Boston Predictive Analytics World event.

Rexer Analytics has been conducting the Data Miner Survey since 2007. Each survey explores the analytic behaviors, views and preferences of data miners and analytic professionals. Over 1200 people from around the globe participated in the 2013 survey. Summary reports (40 page PDFs) from previous surveys are available FREE to everyone who requests them by emailing DataMinerSurvey@RexerAnalytics.com. Also, highlights of earlier Data Miner Surveys are available at www.rexeranalytics.com/Data-Miner-Survey-Results-2013.html, including best practices shared by respondents on analytic success measurement, overcoming data mining challenges, and other topics. The FREE Summary Report for this 2015 Data Miner Survey will be available to everyone Fall-2015.

Please tell other data analysis professionals about the survey.

Rexer Analytics is a consulting firm focused on providing data mining and analytic CRM solutions. Recent solutions include customer loyalty analyses, customer segmentation, predictive modeling to predict customer attrition and to target direct marketing, fraud detection, sales forecasting, market basket analyses, and complex survey research. More information is available at www.RexerAnalytics.com or by calling +1 617-233-8185.

Saturday, September 01, 2012

Data Mining Components

We identify four components or layers in a data mining engagement as shown in the figure below. Two abstract layers, business problem and data mining algorithms, are on the top. And two physical layers, data mining tools and data management, are at the bottom.

Business problems that we want to solve is one of the most important abstract layers. It could be predicting fraud (bank card, check, medical claims), new customer life time value at point of sales, online ads click rate, credit worthiness, customer segmentation. We can address the above business problems using various predictive models such as logistic regressions, neural nets, support vector machines, K-means clustering.

Data mining tools layer contains commercial or open source software such as SAS, Splus, R, Weca, SPSS, Statsoft, Oracle Data Mining. Common data mining algorithms can be found in almost all of the software mentioned above. Data management/storage layer are relational databases such as Oracle, SQL server, MySQL, or simply files such as SAS files or text files.

We can predict if a new customer will pay his car loan using logistic regression model implemented in SAS and store the data in Oracle. Or we can solve the same problem using decision tree models implemented in R and store data in SQL server. It is important to realize that items within each layer are sometimes exchangeable. We can solve any business problems with varieties of data mining algorithms implemented by commercial or open source tools and store data in any databases.  Thus it is a misconcept that neural nets are best in predicting credit card fraud. An experienced data miner can build a decision tree model to predict credit card fraud that performs equally well as a neural net does. We can select the combination of data mining models, tools and databases that suit our needs.
Four Components or Layers in Data Mining