|
Data Mining
All organizations have data in a variety of formats (spread sheets,
databases, etc.) which managers and analysts would like "mine" - reviewing
groupings of the data in order to understand past performance, problem areas,
likely trends, etc. There are many data mining products on the market, but
most fall short in one critical area - allowing end-users to easily filter and
group their data so that they quickly determine both what data is important and
how the data can best be presented. Chesapeake Data Systems has developed both a
data mining product and a data mining service to help address these needs.
The product's features include:
- A Web browser interface so the analysis can be done
anywhere in the world
- Data can be charted, reported with group totals, summarized in pivot
tables, printed on labels, and exported
- Results can be saved (and accessed with a URL) in three ways to simplify
sharing of results - a "snapshot" of the current data, a "query" which is
re-run with the most current data when requested, and a "query" which allows
the user to specify additional filter criteria when the query is re-run
- Over 60 chart formats (provided by Microsoft's Office Web Components)
- A simple "chart specification" form allows the user to easily specify
which fields to chart, which chart format to use, the chart size, labels for
the chart axes, etc.
- The ability to display all of the possible chart formats at once so that
the user can easily determine the best way to visualize the information
- The ability to display multiple charts for a selected category (e.g.,
display product sales by country and year for four employees in four charts,
one for each employee)
- A simple "report specification" form allows the user
to easily specify which fields are to be included in a report, how the
fields are to be grouped and/or summed and/or pivoted, etc.
- An upload process to create database tables from spreadsheets (or other
text files). Since a lot of the data which users want to analyze is in
spreadsheets, this allows the users to get the full advantage of both this
product and database processing without requiring the IT staff to
create a database table for them.
- A data dictionary of tables, views, fields, reporting and charting
specifications, etc.
- "Filter forms" for each table/view (which is used to select a subset of
the records for analysis).
- The user can specify how each field can be
used on the "filter form"
- e.g., a range of values, a combo box, check boxes for each possible value
for a field, not included, etc.
- The data in a table can easily be linked to a "lookup
table" to convert a "code" to the corresponding text; or the user can
specify the text directly in the table's specifications in the data
dictionary
- When installed on a customer's network, the product automatic
initializes the data dictionary with information from a SQL Server (or other
ODBC compliant) database
- And much, much more
You can view a sample data mining application for the
Northwind data at
http://www.chesdata.com/datawizard/
Call us to get a walk-through of a typical data mining
application and to explore how the product will work for your organization.
Chesapeake Data Systems will also provide a Web-based service with full
product capabilities. This will allow users to mine their data without
needing their IT staff to install a product on their network.
Note: Many of the methods used in this product are covered by US Patent
6832226 - Method of Providing Data Dictionary-Driven Web-Based Database
Applications
|