Stay on Top of TM1 10.1 with New Training Courses
With the new release of IBM Cognos TM1 10.1, it’s important to stay on top by ensuring your skills are up to date. See below for the latest TM1 training courses which are conveniently offered both online and in-class.
P8151 – IBM Cognos TM1: Interact with Data (1 day – $700/student)
- Online – Jan. 4
P8157 – IBM Cognos TM1: Analyze and Share Data (3 days – $2,100/student)
- Ottawa – Feb. 25
- Online – Jan. 21
P8155 – IBM Cognos TM1: Administer the Technical Environment (2 days – $1,400/student)
- Markham – Dec. 3
- Ottawa – Feb. 28
- Online – Jan. 24
P8152 – IBM Cognos TM1: Design and Develop Models (5 days – $3,500/student)
- Markham – Dec. 17, Mar. 11
- Ottawa – April 22
- Online – Dec. 3, Jan. 7, Feb. 18
Interested in registering for one of these courses? Have questions about upcoming training changes with IBM? Feel free to contact me directly at 905-474-5141 or csingh@newcomp.com.
Cheers,
Upcoming IBM Predictive Analytics Courses in Toronto and Ottawa
Want to learn more about Predictive Analytics and Forecasting?
We are currently filling spots for IBM SPSS courses in both the Toronto and Ottawa areas.
TORONTO (3600 Steeles Avenue East, Markham, Ontario)
Introduction to IBM SPSS Modeler and Data Mining (2 days – $1400/student) October 15
Introduction to IBM SPSS Modeler and Data Mining is a two day instructor-led classroom basic course that provides an overview of data mining and the fundamentals of using IBM SPSS Modeler. The course structure follows the stages of a typical data mining project, from reading data, to data exploration, data transformation, modeling, and effective interpretation of results. The course provides training in the basics of how to read, explore, and manipulate data with IBM SPSS Modeler, and then create and use successful models. (Course Code = 0A003ce)
Predictive Modeling with IBM SPSS Modeler (3 days – $2100/student) October 29
This course demonstrates how to develop models to predict categorical and continuous outcomes, using such techniques as neural networks, decision trees, logistic regression, support vector machines, and Bayesian network models. Use of the binary classifier and numeric predictor nodes to automate model selection is included. Feature selection and detection of outliers are discussed. Expert options for each modeling node are reviewed in detail and advice is provided on when and how to use each model. You will also learn how to combine two or more models to improve prediction. (Course Code = 0F032ce)
OTTAWA (65 Auriga Drive, Ottawa, Ontario)
Introduction to IBM SPSS Statistics (2 days – $1400/student) October 17
Introduction to IBM SPSS Statistics is a two day instructor-led classroom course that guides students through the fundamentals of using IBM SPSS Statistics for typical data analysis process. Students will learn the basics of reading data, data definition, data modification, and data analysis and presentation of analytical results. Students will also see how easy it is to get data into IBM SPSS Statistics so that they can focus on analyzing the information. In addition to the fundamentals, students will learn shortcuts that will help them save time. This course uses the IBM SPSS Statistics Base features. (Course Code = 0G500ce)
Introduction to IBM SPSS Modeler and Data Mining (2 days – $1400/student) November 01
Introduction to IBM SPSS Modeler and Data Mining is a two day instructor-led classroom basic course that provides an overview of data mining and the fundamentals of using IBM SPSS Modeler. The course structure follows the stages of a typical data mining project, from reading data, to data exploration, data transformation, modeling, and effective interpretation of results. The course provides training in the basics of how to read, explore, and manipulate data with IBM SPSS Modeler, and then create and use successful models. (Course Code = 0A003ce)
Contact us for registration information.
Crystal Singh | Senior Technical Consultant
Office: (905) 474-5141 | Email: csingh@newcomp.com
10 Ways IBM SPSS Destroys the Competition
1) Amalgamating various data sources
This is a very common practice in data mining. You will often receive data files from various formats (i.e. excel, flat file, database, BI) that you will have to merge to form one master dataset. This is especially true when merging in outside variables to build a data model. The Merge Node in IBM SPSS Modeler allows you to merge various datasets together by simply dragging the Merge node from the Record Ops tab and connecting it to all datasets in question.
2) Data Cleaning
Prior to building an accurate predictive model, you need to ensure your dataset is ‘clean.’ This can mean anything from assessing the degree of missing data to understanding where the outliers are and addressing them. The Data Audit node in IBM SPSS Modeler does just that, an audit of your data. This ensures you have a solid dataset to work with. You simply drag a Data Audit Node from the Output tab and attach it to your dataset and click on ‘Run’ to receive a full report on your data set.
3) Automated Data Preparation
When building an accurate predictive model, because we want to look at how a combination of various significant variables or factors work together, sometimes we need to transform some variables to increase accuracy in the data model. This is known as ‘Data Preparation’ and can consume a large amount of time. How do we know which variables to transform and why? The Automated Data Preparation node looks at your Target Variable (what you want to predict) against all the input variables and transforms the necessary variables to increase accuracy. It also keeps your dataset organized by creating new fields containing the transformations, rather than replacing the existing fields. You do so by dragging an Auto Data Prep node from the Field Ops tab and connect it to your dataset.
4) Variable conversions
One of the most tedious tasks in a data mining project is variable creation and conversions. This can mean anything from transforming a numeric variable to a string or vice versa or creating a new variable out of existing variables. For instance, if we have an age variable and we want to create groups out of it. These tedious yet sometimes necessary tasks is made easy by using the Derive node. Drag a Derive Node from the Field Ops tab and attach it to your dataset to activate the dialog box below as an example.
5) Building graphs
Graphs are very useful in data mining and data modeling when looking for relationships between certain variables. Graphs are also a very interactive way to present your data findings to your team in a presentation setting. One of the challenges in constructing a graph is determining the best type of graph that will suit said variables. Whether it be a plot, histogram, pie chart, etc. The Graphboard node in IBM SPSS Modeler, looks at the structure of variables selected and chooses the most appropriate graphs that would best display the data in the most logical manner. You simply drag a Graphboard node from the Graphs tab and attach it to your dataset. Select the variables you wish to see and a list of applicable graphs appear for you to choose.
6) Automated data modeling
Building an accurate data model can be a very daunting task, especially because it can require a degree of statistical knowledge. This is why IBM SPSS Modeler has built-in automatic modeling tools based on what you want to predict. Attaching the Automatic Modeling node to your dataset, it looks at the structure of your target variable plus the factors and outputs the models that best represent your data. To do so, drag an Automatic Modeling Node from the Modeling tab and attach it to your dataset, click run and a list of applicable models is displayed.
7) Automatic variable selection
When given a dataset, a common task is determining which variables are significant to what you want to predict. Sometimes a dataset will have an enormous amount of variables and it’s difficult to know which variables to focus on and which to ignore. The Feature Selection node in IBM SPSS Modeler looks at your variables and tells you which ones are important, which ones are not and allows you to filter out only the important variables. You do so by dragging a Feature Selection Node from the Modeling tab and attach it to your dataset. Click on Run and you get a list of important and unimportant variables.
8) Determining the accuracy of your data model
A data model is only useful if it is accurate and hence actionable. IBM SPSS Modeler has a built-in Data Model Analysis node that compares the actual values with the predicted values and tells you how correct your model is. Drag an Analysis Node from the Output tab and attach it to your data model and click Run to view results of your data model.
9) Deployment to IBM Cognos BI
Deploying the results from a data model is easy because of the many export features and the ease of use. For example, we can build a Cognos Report off our results for easy consumption by dragging an IBM Cognos BI Export node to the data model to export the results.
10) Text Analytics
We can easily leverage survey data and analyze it quickly by extracting key words and phrases about what is being said about your organization. The IBM SPSS Text Analytics package automatically groups them into “negative,” “positive” and “neutral.” It also groups like words together. This allows for easy and accurate sentiment analysis.
For more information regarding IBM SPSS, please contact us at 905-474-4087 x5141 or csingh@newcomp.com.
Upcoming Framework Manager Course in Markham
Wanting to learn more about Framework Manager?
We are currently trying to fill a spot in the upcoming Framework Manager course. The course will be held in Markham and will start July 30, 2012. If interested please find the contact details below as we are needing to fill the remaining spot by the end of the week.
B5152 – Cognos Framework Manager: Design Metadata Models (5 days – $3500/student)
IBM Cognos Framework Manager: Design Metadata Models (V10.1) is a five-day, instructor-led course that provides participants with introductory to advanced knowledge of metadata modeling concepts, and how to model metadata for predictable reporting and analysis results using Framework Manager. Participants will learn the full scope of the metadata modeling process, from initial project creation, to publishing of metadata to the Web, enabling end users to easily author reports and analyze data.
Rebecca Swart | HR & Training Coordinator
Office: (905) 474-5259 | Mobile: (416) 660-0603
Newcomp Training and Education
It’s time to maximize your IT investments with IBM training courses.
Newcomp is now able to sell all IBM training courses, including but not limited to: Cognos, SPSS, TM1, Rational, WebSphere and InfoSphere. It is our hope to provide each of our diverse clients with the skill sets that they require to use their technology to its fullest potential. Our courses are available In-Class, Instructor Led Online (ILO), Self-Paced Training and Custom Onsite.
For more information or to book a training course, please contact:
Rebecca Swart | HR & Training Coordinator
Office: (905) 474-5259 Email: rswart@newcomp.com
February e-NEWScomp
This year Newcomp will be issuing monthly e-newsletters to keep everyone informed with all that is happening. The idea is to open up dialogue for two-way communication with the hopes that you will give us your feedback and support throughout the year. Newcomp is constantly looking for new trends in technology with regards to Business Analytics and it is our hopes that this newsletter initiative will keep everyone educated with all that we have to offer. Each newsletter will contain upcoming workshops, webinars and events and will provide links for registration forms, support renewals and other useful information. This year we will also be maintaining social media accounts including a LinkedIn profile, Twitter account, a WordPress blog and a YouTube Channel. Please stay connected and take part in our online discussions.
Here is the February 2012 Edition of e-NEWScomp.
Now Trending… IBM Netezza
On last week’s Q4 analyst call, IBM CFO Mark Loughridge announced that IBM actually ran out of their Netezza product during their fourth quarter and that the product saw a 70 per cent growth.
More about Netezza:
IBM Netezza is a market-leader in data warehousing and offers plug-and-play capabilities that optimize the data warehouse for analytics. IBM Netezza easily integrate with your preferred ETL and BI tools.
IBM Netezza is scalable and offers the ability to process extremely large amounts of data. Using Massive Parallel Processing (MPP), IBM Netezza breaks queries into smaller components for optimized parallel processing.
Happy New Year
It is that time of the year again when we are all trying to make resolutions of how we are going to better ourselves, our businesses and our lifestyles in the upcoming year. In the end all that we want is to make the new year the best it possibly can be. Newcomp would like to wish everyone a very happy and prosperous new year. 
Predicting the Holiday Shopping Rush
It’s the holiday season and everyone is out shopping for that perfect gift! We have all been there frantically looking for that one item that is SOLD OUT everywhere. So how do companies prepare for the busy holiday rush? It’s predictive analytics and its changing the way in which stores order and plan their inventory both during the holiday season and otherwise.
Check out this article that discusses how companies can use IBM’s predictive analytics tool to help determine the amount of inventory needed for their stores. The tool can forecast to a very specific level, meaning that not only will they know they should sell “x” number of that new shirt but that they will need “x” number in each size and colour. Think about how this tool can help save companies a lot of time, effort and money.
http://www.computerworld.com/s/article/357932/The_Art_Science_of_Fashion?taxonomyId=18&pageNumber=1
The article later discusses potential use for the tool. The options are endless and the possibilities extreme. As long as you can dream up what questions you are seeking answers for and find the pertinent data, predictive analytics can take away a lot of the “what if” in life.
