Predictive Data Management™
PDM is a fully integrated Data Quality Solution which includes: Technical Metadata, Data Profiling, User Statistics, Data Cleansing of all forms/content of data, Flexible Use of Regular Expressions, CASS Certified Address Correction, Fuzzy Logic, and Standardization Tables for Data Validation/Correction and Identification of Duplications of Data.
Features & Functions
Technical Metadata - Stores the following information: Table size, number of columns, column size, attribute type, sql type, key information, ordinal position and more
Data Profiling - Includes the following column information: Minimum, maximum and average values, duplicate count, null count, blank count, zero count, max bytes used, unique count, percent unique, unique domain patterns, has special characters, preceding and trailing spaces and more
User Statistics - Provides the user with the ability to: Create profiling statistics via SQL statements that get appended on the PDM Profile Table at the same time and one pass of the source data. User Stats ad necessary extensibility to the PDM Profiling Results.
Domain / Pattern analysis - Analyzes the following: Column value, column value length, frequency of occurrences, discrete percent, distribution percent, SoundX, pattern of the data and more
Relationship Finding – Exact Match & Fuzzy Match Identifies and counts all the times that each table’s column values match its own table columns and other table columns’ values, accomplished by both Exact Matches and Fuzzy Matching
Master De-Duping – Provides the ability to: Analyze one column or multiple columns concatenated and identifies duplicates via an Exact Match or a Fuzzy Match – then displays the number of duplicates and the detail records from the source system of the duplicate source records. The De-Duper concatenated column process first creates a Standardize From-to lookup table then concatenates the source data and Fuzzy matches against the from or to value in the De-Dup Standardize table.
Source Data Validation – Validates that the source data is properly formatted, such as a SSN, EIN, customer number, data, e-mail address or equal to specific or a range of values and creates an output table identifying the values as Valid or Invalid. This is accomplished with standard delivered regular expressions in our Regular Expressions engine which also allows users to create their own regular expressions for validation and Find/Replace Cleansing.
Fuzzy Matching – After the creation of a Standardize From-To Look-Up table, Fuzzy Matching takes the source column data and matches row at a time and matches the data in the from-to-look-up table values via an advanced Fuzzy Algorithm based on requested Score of Confidence of the Match and the count of how many Fuzzy Matches were found and the Value that it Matched against.
Double Metaphone Data Value Sound Like Analysis – Creates a Primary and Secondary 4 byte key for each columns data value. This value is either the same or close in value when the data values sound like each other. This process is good for analysis of Duplicates or Misspells.
Standardize Table Creation for Process Maps – Creates a Standardize table name in the PDM Standardize menu. Assigns a column to that table name and by running Standardize in a Data Map Process it will populate that table with all the Unique Data Values for use in the Process Map for Cleansing and Fuzzy Lookups and Matching
Standardize Table Creation for Master De-Duping – Creates a Standardize table name in the PDM De-Duping Standardize menu. After Creating a De-Duping Map, right click on Column Name and click on De-Duping Standardize. This action creates a Standardize From-To Look-Up table to be used in the Exact Match and Fuzzy Match of De-Duping Concatenated Processes.
Cleansing – Standard Deliverable
Trim White Space – Removes Preceding and Tailing Spaces from the Input Data Source
Find & Replace – Select Drop Down box to choose the Group containing the Regular Expressions to be used in the Validation of the source data and the action to be applied to that data when it meets the validation criteria
Validate – Validates that the source data does or does not meet the criteria of the validation regular expression selected in the process map
Standardize Lookups and Replace – Allows the selection of the Standardize From-To Look-up table to be used to find the input value from the source data and then converts that value to the user created and verified value in the Convert To value in the From-To Look-up table that was selected.
Name Matching and Sounds Like Using Double Metaphone – Creates a Primary and Secondary 4 byte key for each columns data value. This value is either the same or close in value when the data values sound like each other. This process is good for analysis of Duplicates or Misspells.
Fuzzy Matching Algorithm – After the creation of a Standardize From-To Look-Up table takes the Source column data and matches via an advanced Fuzzy Algorithm based on requested Score of Confidence of the Match and the count of how many Fuzzy Matches were found and the Value that it Matched against.
Regular Expressions – PDM delivers an interface that allows users to add their own Regular Expression Groups and within that create Regular Expression for the Find or Validate and the Replace to modify the source data that meets the criteria of the Find Regular Expression.Regular Expressions are used in both the Find & Replace Cyclone Operation and the Validate Cyclone Operation. Regular expression can simply change values from one value to another and can even parse source data for later processing.
CASS Certified Address Correction
Unlike traditional data management solutions that rely on confirming suspected problems
after the fact, causing down-time, interrupted customer service and missed schedules,
Predictive Data Management™ discovers unknown potential problems before the fact
and prevents data irregularities from ever taking place.
PDM is your go-to Data Governance solution!
Given that a high level of data quality is mandatory to maintain corporate data
standards, PDM is the ideal solution for Data Governance standards required for
your enterprise information. With PDM, you can monitor, identify and correct the d
ata necessary to meet compliance regulations.
No more garbage-in, garbage-out projects!
PDM is the most economically-priced Technical Metadata, Data Profiling,
Data Cleansing, Data Quality, Address Correction Suite ever to be offered both as a product and an ASP application!
