Development & Technical Support Web

HOME What's New FAQ Bugs Feedback Download Discussion Search

Frequently Asked Questions

This page contains answers to common questions handled by our support staff, along with some tips and tricks that we have found useful and presented here as questions.

Note: In these answers we will follow a few shorthand conventions for describing user-interface procedures. Key combinations will be presented like this: Ctrl+Alt+Delete, which means that you should press and hold down the Control key, the Alt key, and the Delete key at the same time. Menu selections will be presented like this: File->Open, which means that you should open the File menu, and then make the Open selection.

  1. How do I load two tables, and link them using a common field?
  2. Is it possible to load part of a table?
  3. Is it possible to export a view of a table?
  4. How can I search for a particular record in the whole table?
  5. When I run the fact engine on my table, I get some trivial facts, How can I avoid that?
  6. I have a table and I would like to discover the facts between two columns X and Y, How can I do that?
  7. I have a table and I would like to discover the facts between a group of columns versus another group of columns. How can I do that?
  8. What is meant by higher level facts?
  9. I have numeric columns that are continuously valued. How can I generate facts from them?
  10. How can I eliminate the rows that have a particular field value from the view?
  11. Can I compute other statistics than the average, maximum, and minimum in the data table?

1. How do I load two tables, and link them using a common field?

You can open any number of tables using the Open Wizard, then you need to define primary and foreign keys for each table. The Open Wizard will help you do that also.

For example, suppose that you have Customer and Order tables. The common field in the two tables is CustomerNumber. CustomerNumber should be the primary key in Customer Table and should be defined as a foreign key in the Order table.

To define primary and foreign keys, use Table->Link, then follow instructions.

Back to Top

2. Is it possible to load part of a table?

Yes. The Open Wizard allows you to define criteria, select columns, and define a scope.

For example, you can read using a criterion like year = 97, you can also read a random sample of your data say 20%, and you can select which columns you would like to read.

To read part of your table, use Table->Open Wizard, then follow the instructions in the Open Wizard.

Back to Top

3. Is is possible to export a view of a table?

Yes. You can export any view that you create on your data.

To export a view, use View->Export.

Back to Top

4. How can I search for a particular record in the whole table?

You can search for a particular record by moving to the desired column and pressing the find button . The current cell will be positioned on the first row that has required value. You can find next and previous.

Back to Top

5. When I run the fact engine on my table, I get some trivial facts. How can I avoid that?

You can get some trivial facts when some of your columns contain obvious dependencies. For example, if you have a State, City, ZIP, and Country columns in your table, the engine will report facts like
All City = NYC has State = NY
You can avoid these facts by 2 ways:

  1. 1. Remove all these columns from the view using the setup button . The engine only works on columns in the view.
  2. 2. If you need to find facts involving these columns but not among them, for example facts between Profit and Sate and between Profit and City but not between State and City, then you need to specify Dependent Column Groups that contain State and City. The engine will not report any facts among each group.
    To specify dependent columns, press the fact engine button , then press Dependency button and follow instructions.
Back to Top

6. I have a table and I would like to discover the facts between two columns X and Y. How can I do that?

Press the fact engine button , then select column X from the "If Condition" list and Column Y from the "Then Conclusion" list.

Back to Top

7. I have a table and I would like to discover the facts between a group of columns versus another group of columns. How can I do that?

If you have a group of columns say ProductType, ProductSize, and ProductColor and another group like Profit, QtySold, and StockLevel, and you would like to find any correlation between the two groups but not among each group. In other words, you are interested in facts between ProductType and Profit but not between ProductType and ProductColor . Follow these 2 steps:

  1. Create a view with only columns in the two groups, then remove all other columns. Use the setup button .to deselect all other columns.
  2. Press the fact engine button , then use the Dependency button to define the two groups, this way the engine will not report facts among each group.
Back to Top

8. What is meant by "Higher Level Facts"?

A fact that contain one condition is a Level 1 fact while a fact that 2 conditions is a Level 2 fact. For example

Level 1 fact: All Products where Color = Black have Profit = High

Level 2 fact: All Products where Color = Black and Size = Large have Profit = High..

Back to Top

9. I have numeric columns that are continuously valued. How can I generate facts from them?

Yes. However, you need to partition these column into ranges or levels first. To do that we create Range Columns using the virtual column button then follow instructions.

Back to Top

10. How can I eliminate rows that have a particular field value from the view?

Yes, You can create a Filter Column using this button on that value that you want to eliminate from the view. Move to any row that contain false and press the filter button .

Back to Top

11. Can I compute other statistics than the average, maximum, and minimum in the data table?

Yes. Press the right mouse button while in the Totals pane and press Select Calculations.

Back to Top

 


SuperQuery and Knowledge Agent are trademark and servicemark of AZMY Thinkware Inc. All other products mentioned are registered trademarks or trademarks of their respective companies.

Questions or problems regarding this web site should be directed to mail@azmy.com.
Copyright © 1997 AZMY Thinkware Inc. All rights reserved.
Last modified: Saturday March 06, 2004.