 |
|
|
 |
What's New?
As you may already know, VisuaLinks is 100% Java. This means that it runs under all Microsoft Platforms,
UNIX, Linux, and a host of different machines. We have recently had the pleasure of testing our
VisuaLinks 4.0 on an iMac G5 - Mac OS X Version 10.3.5. All VisuaLinks features and functions are
fully supported because the same Java code runs across all platforms. The following screen capture was
taken on the iMac.
|
|
|
 |
 |
...that all of the reference data for all unique data records are maintained for each object?
When viewing the detailed results for any object, there are a fixed number of merged values that are displayed by default in VisuaLinks. A merged value represents the different instances for the attribute defined; for example, a cargo manifest may contain many different types of commodities or goods that are being shipped within a container. Since a manifest contains many items, there needs to be a reliable method by which to represent the range of products being shipped. This many-to-one relationship is handled within VisuaLinks through the user of the merged objects. By default, the number of values supported by a merged object is artificially limited to five (5), however, it can be extended to support any number of values. To change the basic settings to show at least 100 different merged values, the administrator should set the following properties:
In ServerResources.properties:
##Max Objects merge count
com.vai.vl.merge.objects.max.count=500
com.vai.vl.merge.attributes.flag=true
com.vai.vl.merge.node.attribute.max.count=100
com.vai.vl.merge.attributes.label.max.count=10
Also, in ./serverresources/ClientResources.properties:
##Max Objects merge count
com.vai.vl.merge.objects.max.count=500
com.vai.vl.merge.attributes.flag=true
com.vai.vl.merge.node.attribute.max.count=100
com.vai.vl.merge.attributes.label.max.count=10
After making the appropriate changes, any subsequent representation of data within any VisuaLinks drill down will show at least 100 unique values.
|
|
|
 |
Maintaining the integrity and representation of data is very important for generating accurate analysis. VisuaLinks is outfitted with the ability to show all of the unique value representations of data for every object generated through the use of Merged Nodes. Many other systems do not provide this level of support and their results may prove to be inconsistent, inaccurate, or misleading. VisuaLinks is the only system that properly supports multi-valued attributes for its object representations by encoding each distinct record value into a merged-attribute representation.
For example, the diagram below depicts a single financial transaction (SAR=Suspicious Transaction Report) with four (4) distinct SUBJECTS plus a single account number. This type of situation is common in all sorts of databases including financial transactions, travel events, telephone tolls, medical/health diagnostics, and other sources where many-to-one or many-to-many relationships are required.
As with all objects represented in VisuaLinks, a detail (drill-down) can be performed on the SAR object to view all of its supporting information. The following table shows the underlying data for the SAR object shown above. Careful attention to the data shows there are "MERGED_" attributes where the different values for each of the SUBJECTS are displayed.
As can be seen in the details window, the MERGED_ attributes contain multiple values as they appeared in the database from the records that were used to construct the SAR object. For example, there are four (4) different Social Security Numbers (SSNEIN) provided on this SAR - one for each of the SUBJECTS. Additionally, there are different STREET names, multiple VIOLATION-AMOUNTS, and several other merged values. Thus, VisuaLinks is able to keep track of all the different values and present them to the user for review.
The details window shows a condensed/collapsed version of each of the individual values derived from the database. However, it may not be obvious which values belong together (e.g., which ADDRESS goes with which LAST NAME). To see the explicit representations of the data used to construct the objects, the DETAILS PANEL provides an explicit mapping for each value encountered. As shown below, the SAR object (shown as a DCN) has a special column identified with a *MERGED label. As can be seen, the value of this column states "MERGED… 3 More…" which indicates there are a total of 4 unique value representations contained in this single object. Obviously, there is one record for each unique SUBJECT represented on this SAR.
The values for the primary object (e.g., the first record encountered in the data set) are presented as the default value in the details panel. The 2nd, 3rd, and 4th objects appear in a separate pop-up window showing their respective data values. Each row represents a different data record and the columns provide the unique values for each record. Scrolling through the attributes shows how each record differs from one another.
Using this approach, VisuaLinks can maintain an endless number of values supplied for any object representation thereby providing accurate representation and comprehensive analytical foundations unmatched by any other system in the marketplace.
Also, see a related article in our April, 2004 Newsletter.
|
|
 |
Wire Transfers for Alien Smuggling
The amount of data reported by financial institutions is increasing every year as more filing requirements and
compliancy laws are being enacted to help combat financial crimes and money laundering activities, especially
those involving terrorist financing. The focus this month returns to the SAR-MSB (Suspicious Activity Report -
Money Service Business) where a high-value SUBJECT is identified in making false representations regarding her
identify when transferring money (via wire transfers).
As with many proactive criminal investigations, the exact nature of the pattern is not entirely known
beforehand and often, only becomes clear once we see the entire picture. In this particular case, the
indicators used to identify a well-qualified starting point were derived using the Network Miner service
in VisuaLinks. The system was configured to show all networks using the linkage between
SUBJECTS and ID NUMBERS where the total network size exceeded 10 nodes. As is presented in the
following table, the top entry in this list shows a single ID NUMBER being utilized by 12 different SUBJECTS.
Although there are other, larger networks generated by the Network Miner, it was decided to review only
heavily utilized ID NUMBERS to expose intentional misrepresentation.
The first step in this process was to select the targeted network and regenerate it in the VisuaLinks display. The diagram shown below represents the results returned. For security purposes, any specific details have been hidden, removed, or changed in this example - but the structural integrity of the diagram has not been altered in any way. The 12 SUBJECTS connected to this single ID NUMBER are all variations of the same name. The last name (not shown) clearly represents the same individual; however, the number of different variations of how her name was spelled resulted in this specific network configuration.
At this point, it is not entirely clear if the name variations are intentional or due to the way in which this name can be interpreted (it is Spanish in origin). One thing that is positively known from this diagram is that there will be a minimum of 12 SAR-MSB transactions used to support these connections, and perhaps more. Thus, we have a potential target based on a large filing frequency.
From here, the network is expanded one level using the Database Walk feature provided in VisuaLinks. The next diagram shows the individual SAR-MSBs associated with each SUBJECT depicting a total of $115,450 dollars transacted over a 4 month period of time. Close observation also indicates that there were two transactions that occurred on 08/19/2003. Due to the fact that there are exactly 12 SAR-MSBs, it starts to become evident that this individual may be trying to misrepresent herself to avoid detection because it would be highly unlikely for the same financial institution to misrepresent her name wrong 12 different times.
The network is expanded a third level to bring in the corresponding ADDRESSES. As shown in the next diagram, there are some common ADDRESSES that are shown [the common ADDRESSES connect together two or more SUBJECTS]. Each ADDRESS icon in the display represents a completely different location. In this situation, the SUBJECT has listed multiple addresses throughout the Phoenix, Arizona region. Again, we must question why there are eight distinct and unique addresses listed for this person in such a short period of time.
The next diagram introduces the four Social Security Numbers (SSNs) for the SUBJECT. The largest connected SSN has been used in eight of the transactions. A close examination of the other three SSNs indicates they are entirely different numbers and do not reflect any type of transposition of numbers, which is often seen when multiple SSNs are connected to the same SUBJECT.
Finally, the PHONE numbers are presented showing nine unique numbers. Using the HTTP Search feature supported by VisuaLinks, each number is checked against the Google White Pages and none come back with any additional information. Thus, these PHONE numbers are most likely made-up or non-existent. Based on the overall structure and representation of this network, it can be concluded that intentional misrepresentation is occurring for this SUBJECT.
In fact, the NARRATIVES for many of these SAR-MSBS explicitly state this SUBJECT is using multiple ADDRESSES or SSNS or PHONES. Several of the NARRATIVES indicate she comes in "several times a day" to conduct transactions and each time provides a different ADDRESS, PHONE, or SSN. All of the transactions are in-bound which means she is receiving money sent by other people. Closer review of the NARRATIVES reveals the senders are located in different parts of the United States and the amounts are between $1,500 and $3,500. This most likely represents some type of enterprise involved with illegal alien smuggling - because they are one-time payments being made by multiple people across the country.
For additional VAI material and coverage of the SAR-MSB data and various smuggling examples, please review these previous newsletter entries:
Issue 8 - August 2004 - Suspicious SAR-MSB Filing Data
Issue 10 - October 2003 - Compliance Analysis
Issue 5 - May 2003 - Detecting Financial Crimes
|
|
|
 |
|
