Synapse

How to Configure a Domain

Introduction

Domain Configuration is the process in which you create a structure for your data. By the end of this guide, you will understand how to identify scope, design a data structure and create a data structure in Synapse. You will need to read the Synapse “Key Concepts” guide as a pre-requisite to this guide.

Before diving into the step-by-step process of configuring a domain within Sapient Synapse, you must understand what a domain is and understand how to identify scope and approach to designing a data structure. The sections listed in this guide are recommended to be followed in the order as presented.

What is a domain?

A domain is an instance containing specific data information defined by the business needs. Synapse first and foremost is a data requirements tool therefore an organization can have multiple domains, each for different data-driven projects. For example, your organization can have a domain dedicated to governing glossaries and another domain dedicated to data mappings for the purpose of replacing a legacy system.

How to identify scope and architect data structure?

This process begins by analyzing the existing dataset or sample dataset because in some cases the data has yet to be collected. When analyzing the data, you need to answer five basic questions. Once the first three questions are answered, the scope will be clear as to what the business need for a domain would be. The last two questions will enable you to design a data structure in Synapse that meets your specific needs.

  1. What business needs are these data requirements going to solve?

At the high level view, this question aims to narrow down your focus to immediate needs. Data quality concerns, business process gap analysis, lack of transparency in data flow, and lack of cohesive understanding of terms are examples of business needs. In some cases you may have a combination of these needs. You may or may not be able to consolidate these needs into one domain but identifying them is the first step to identifying scope. The National Basketball Association (NBA) example strives to provide transparency into what players are coached by what coaches. In addition the NBA wants to easily find out details about a specific player or a coach.

  1. What type of information are in these data requirements?

Once a business need has been identified you need to understand the type of data that need to be collected. For example the lack of transparency in data flow may need identification of system fields and where they come from and where they are going. You may also want to know the business processes and owners related to those fields. Ultimately understanding what data you need to collect will help you establish an informal structure. Since the NBA example strives to show transparency between players and coaches, the data identifying relationship between a player and a coach needs to be collected. In addition, details about a player or coach such as age or height are also important to the NBA.

  1. Which of the data requirement concepts are related and can be classified together?

As you identified the type of data, you should group similar data points together. Business Owners, Support Team, Developers and Chief Data Officers can be grouped together as People. All processes can be grouped under Business Process. All systems can be grouped under Physical Sources. These are some of the grouping examples that you can have. Synapse has no restriction in the naming convention of these groups. For the NBA example, the classification is narrower because the focus is between players and coaches therefore those classified as players will be one group and those classified as coaches will be the other group.

  1. Who will be the primary users?

Identifying who will see and use the data collected will determine the design of the structure. Compliance, Operations, Support Team, Developers, Chief Data Officers and Data Analysts, each will have different purpose to viewing a set of data. For the NBA example, the primary users will be the NBA fans and potential NBA players.

  1. Which concepts are important for the primary users?

Knowing your target audience will help you design a structure that will be resonate with them thus encouraging them to use the data collected on a regular basis. The best practice is to interview these people. Find out what is important to them, what they want to see and what their pain points are. For the NBA example, the NBA fans and potential players need to be able to easily see who coaches who at any point in time. In addition, they want to see how old or tall a coach or a player is.

At the end of this exercise, you have identified that the NBA data structure will be Players and Coaches. Within Players and Coaches, you will need age and height as attributes. Now follow the steps in Domain Configuration to set up the defined NBA data structure.

Domain Configuration

This section is divided into seven sections and is best followed in the order that they appear. The NBA dataset is used as an example to demonstrate the step by step needed to set up a domain.

Step 1: Create Level

For the purpose of this guide, People in the NBA are classified as a concept. This high-level concept would be the Level in Synapse. As such, Level 1 will be People. The following step-by-step instructions detail the creation of Level 1(People):

  1. Navigate to Configuration → Levels
  2. Click on Add. A pop-up window will appear.
  3. Populate the following information:
    1. Level No.—Indicate the Level Number associated with the concept. The red asterisk indicates that this is a required field.
    2. Name—Indicate the Name of the Level.
    3. Alias— Indicate the Name of the concept.
    4. Background Color—Select the background color or use the default color.
    5. Foreground Color—Select the foreground color or use the default color.

Step 2: Create Level Column

There are different roles for People within the NBA. There are Players, Coaches, Owners, Agents, etc. These granularities for concepts are classified as Level Columns. For the purpose of this guide, the following step-by-step instructions demonstrate how to create two  Level Columns: Players and Coaches.

  1. Navigate to Configuration → Level Columns.
  2. Click on Add. A pop-up window will appear.
  3. Populate the following information:
    1. Default Entity Type ID—Indicate the Entity Type associated with the Level Column. At this time, leave this field blank until Step 4 is completed.
    2. Level ID—Indicate the Level name associated with the Level Column.
    3. Name—Indicate the Name of the concept.
    4. Alias—Indicate an alternative name of the concept. This is the name that will appear in the Information Mapper.
  1. Repeat steps 2 and 3 to add a Level Column called Coaches.

Step 3: Create Attributes

Each Level Column can have attributes to provide additional information about the concept. Players’ attributes can include but are not limited to age, height and wiki page. Coaches’ attributes can include age, coach type, phone number, address, etc. The following step-by-step instructions demonstrate the creation of two attributes (Age and Height) for Players. After the attributes for Players are created, repeat the steps to create two attributes (Coach Type and Age) for Coaches.

  1. Navigate to Configuration → Templates
  2. Click on Add. A pop-up window will appear.
  3. Populate the Template name:
  1. Once the Template is created, navigate to the right to the Template Attributes.
  1. Click on Add. A pop-up window will appear.
  2. Populate the following information:
    1. Key—This is the primary identifier to capture each unique attribute.
    2. Name—Indicate the name of the attribute that will appear in Information Mapper.
    3. Group Name—This is used to group related attributes under a specific category.
    4. Data Type—Indicate the type of values of the defined attribute. The options are as followed:
      1. Text—Plain text
      2. Number—Numeric value
  • Date—DD/MM/YYYY
  1. True/False—Check box for True. Leave the empty box for False
  2. Dropdown options—Free text to store possible values
  3. Link—Web link format for all websites
  4. Image Link—Web link format to images only
  1. Repeat Step 5 to create a Height attribute.

Step 4: Create Entity Type

As stated above, each Level Column can have a set of attributes grouped together through Template IDs. These Template IDs have a one-to-one relationship to Entity Types. The following step-by-step instructions demonstrate the creation of one Entity Type called Player:

  1. Navigate to Configuration → Entity Types.
  2. Click on Add. A pop-up window will appear.
  3. Populate the following information:
    1. Description—Indicate the name of the Entity Type.
    2. Template ID—This is a one-to-one relationship to Entity Type.
  1. Repeat Steps 1 through 3 to create an Entity Type for Coach.

Step 5: Link Level Column to Entity Type

Entity type represents the nature of an Entity. In this example, Entity Type called Player describes all players in the National Basketball Association represented as entities. Entities reside in Level Columns. In order to capture the nature of these players, you must link Level Column to Entity Type using the follow steps:

  1. Navigate to Configuration → Level Columns.
  2. Click on the Level Column called Player.
  3. Click on Update. A pop-up window will appear.
  4. In the drop-down menu of Default Entity Type ID, select Entity Type Player which contains a set of attributes describing the Level Column Player:
  1. Repeat Step 1 through 4 to link the Level Column Coach to Entity Type Coach.

Step 6: Create Relationship Type

Sapient Synapse enables you to write the semantic relationship type in free form. Now that the structure of Player and Coach are created along with their sets of attributes, you need to define the relationship type between these two Level Columns.

  1. Navigate to Configuration → Relationship Types.
  2. Click on the Level Column called Player.
  3. Click on Update. A pop-up window will appear.
  4. Populate the following information:
    1. Relationship Type—Indicate the two Level Columns that are going to be linked.
    2. Description—Provide a detailed explanation of the relationship type. This section is optional, but as a best practice, it should be completed for administrators to fully understand the relationship type.
    3. Entity Type From—Indicate the first Level Column of the relationship.
    4. Entity Type To—Indicate the second Level Column of the relationship.
    5. Forward Relationship—Indicate the semantic meaning from the first Level Column to the second Level Column.
    6. Backward Relationship—Indicate the semantic meaning from the second Level Column to the first Level Column.

Step 7: Start adding Meta Data!

Congratulations. You just learned how to configure a domain in Synapse.  You can now start adding metadata.

Entities Card Editor