Support

In Lyza, a dataset is a collection of columns, much like a database table. Each column must be one of three data types - Text, Number, or Date. Within a given column, every value must be the same data type. So, for a given column, that means it is not possible for row 1 to be Date, for row 2 to be Number, and for row 3 to be Text. The entire column must be the same type.

Lyza handles Dates in an intuitive, context-sensitive manner. In some situations, it is natural to treat a date as a number, while in others it is natural to treat it as a text string. Note: In Lyza, a date is a number on a timeline. To be precise, it is the number of milliseconds since midnight on January 1, 1970. However, it is possible to convert this number to a meaningful text string - a readable, calendar-based date/time label. Throughout Lyza, when we refer to a date value, we are talking about a number on a timeline. This number is used to perform date calculations, such as adding 5 days to a date, or subtracting two dates. By default, Lyza does not display dates as numbers; it displays them as formatted text. So, when we refer to a formatted date, we are talking about the readable, meaningful text string to which a date value has been converted.

If any records have been dropped during the import process, the data connection icon will glow yellow. This is to alert you that records have been dropped, and to encourage you to investigate. It is helpful to underscore the reasons why a record might have been dropped. 1. A value in a particular cell does not match the data type of the column. This is the most common cause for dropping a record. 2. A record does not contain the right number of columns.

A filter is a logical condition that must be satisfied for records to pass from the input table to the output table. For example, "Cost < 10" allows records having a Cost value less than 10 to flow to the output, and excludes records with a Cost greater than 10.

Lyza supports an extensive library of functions - Math, Text, Financial, Date, Logical, etc., that allow you to calculate new values with sophisticated formulas. All formula computations and derivations are conducted from within a Transform step.

Lyza creates a ghost step within a workbook to act as a proxy for a step from earlier generation. Ghost steps are transparent. Lyza creates ghosts to avoid drawing connector lines that span more than one generation in a workbook and to keep your workbooks tidy.

Workgroup software subscription that enables the sharing and monitoring of data for Lyza and Lyza Lite desktop analytics.

A dataset created in one Lyza workbook that is shared with another Lyza or Lyza Lite desktop user. Lyza Datasets can also be shared between workbooks by a single user.

Lyza Lite is desktop data integration, analysis and reporting software with data connections to Excel, Access and text/flat files.

Lyza Reference Tables allow you to create and manually maintain datasets called reference tables.

A merge key is a set of columns that contain matching values from two datasets. The columns that comprise the merge key must exist in both input tables. Therefore, by matching values in key columns, you can create a single table with attributes from each input table.

Merge joins two inputs on a common key. The result is an output that has columns from both inputs.

Number is the data type that Lyza uses for all numeric constructs, including decimals, integers, negative numbers, currency values, or percentages. For example, 3.14, 10, -1.4, 110%, and $19.99 are valid numbers in Lyza.

Row Multiplication occurs when multiple matching key values are found in both merge inputs. When this happens, Lyza will generate one output record for each matched pair (Cartesian product). The likely outcome of row multiplication is that the output dataset will contain more rows than either input, potentially a lot more.

A snapshot is an archived read-only report that can be shared with others.

Records in the output table may be sequenced according their numeric, chronological, or alphabetic value. An ascending sort puts the small values at the top of the list and displays them in increasing order; a descending sort does the opposite.

Text is the most generic data type since all data can be interpreted as a string of characters. Text includes the lower-case characters "a-z", the upper-case characters "A-Z", the digits "0-9", plus all printable, symbolic keyboard characters like "!@#$%^&*()" to list a few. For example: "Lyza", "1998", "@Work", and "3/7/2007" are text strings.

Have you ever looked at a chart or report and wondered, Where did that number come from? A feature in Lyza - called Tracing - helps you answer that question. When you trace a column, Lyza presents you with a map of the data’s lineage - the filters, formulas, manipulations and sources that document how the data came to be.

Transform is used to filter, sort, and compute new columns using formulas.