DV KG provides 3 categories of linkage relationships: bind relationship, transaction relationship, and shared value relationship. （Set up please refer to Section 2.3 Add Node and Relation)
The bind relationship refers to the user’s basic information, such as identification information, mobile phone number, e-mail address, bank account number, residential address, work unit, device number, IP, etc.
The transaction relationship generally refers to transactions under the user’s account or bank card such as transfer (to others or other personal accounts), consumption, cash withdrawal, monetary gifts, etc.
The shared value relationship refers to the generalized association information shared among user nodes, which could be taken as a supplement risk signal to help case review, e.g. all of users access their accounts from the same channel; or located in the same city, etc.
The share value relationship is similar to the bind relationship. The only difference is that, for bind relationships, it will expand the graph using the entity. For example, when we analyze a user node, it has drivers license number 12345 and we config it as a bind relationship, then all other users sharing the same driver license number will be automatically pulled and plotted on the graph. For a shared value relationship, it is a weaker relationship. System won’t pull the additional node sharing the value. Instead KG would just be based on the current graph, linking all nodes with the same property together.
These three types of linkage relationships correspond to different line types (solid for bindi, arrows for transaction, dotted for shared value) to facilitate intuitive visual distinction. An example is shown in the figure below. The newly imported data and attribute nodes can all be classified into these three existing relationships, and the relationship sub-categories can be added according to business needs.
The complexity of the KG is based on the number of nodes and the relationship between them. For example, the case in the figure below contains two types of relationships: bind and transaction.
Uncheck the transaction relationship to only view the bind relationships in the KG. This function is convenient for observing particular relationships of interest to the business.
Another example with the shared value relationship is shown below.
6.1 Transaction Relationship
Hover the mouse over the transaction line to display the transaction overview (m, n), where m indicates that m transactions have occurred in the direction of the arrows during the selected time period and n indicates the total transaction amount. The currency unit depends on the currency used in the transaction.
Click the transaction overview (m, n) on the transaction line, and the corresponding details about the exchange flow under the current transaction direction will pop up. If there is no custom configuration, the default fields in the transaction flow will be displayed by default.
6.1.1 Time Range of Graph Filter
In the transaction scenario, the transaction display is based on the transaction time range. The side panel contains the function for setting the time range.
The Time Range function includes the start date and the end date, with the 24-hour time adjustable to the second as shown in the figure below.
6.1.2 Transaction Amount of Graph Filter
The side panel on the left has the function of filtering the transaction amount by a single transaction amount or by the total amount for a one-way transaction. In the current graph, the upper and lower limits of the amount range can either be set by using the sliding bar or by inputting precise values manually. After setting this filter, the graph will only display the transaction relationships that meet the conditions.
An additional function displays the top transaction amounts by top (one-way) transaction lines (links) or by top transaction account (user) nodes as shown in the figure below.
As shown in the figure below, the number of top transaction links has been set to 10, and only the top 10 one-way transaction lines are displayed in the KG.
Following the previous examples, there are 6 total transaction account nodes. However, after setting the top transaction users to 3, only the first 3 accounts are displayed as shown in the figure below.
6.2 Create KG Linkage Template
The KG has functions for template customization and template switching. If the template is not specially configured, the default KG template will be displayed.
In the case of business needs or personal preferences, the required display templates can be customized through the template library, or an entirely new template can be created.
The template library is shown in the figure below. It contains all the existing graph templates. You can edit, update, or delete any existing template; however, the default template cannot be deleted. In the template selection drop-down menu, directly select the existing template you want to use.
If the existing template can no longer meet the display requirements of the KG, a new template can be created as shown in the figure below.
The steps of creating a new template are as follows. First, select the type of linkage relationship. Next, select the corresponding attribute node under the selected linkage relationship. After the linkage relationship and the corresponding attribute node have been configured, name the new template. Click Cancel to cancel the template creation or click Save to save the new template into the existing template library and drop-down menu. Templates in the existing template library can be called directly.
In the example above, the template “newmb1” is being created. In the example below, the template “newmb1” has been saved and created successfully.