 |
|
|
 |
What's New?
VAI has just released the latest update to VisuaLinks 3.0. The 3.0.4 release is now available for download from our Support Site.
This interim release includes a number of updates intended to address issues identified by our user community. Updates in this release include:
- Changes to the VBase structure to address and accommodate Oracle9i-specific issues such as LONG fields and caching.
- Changes to the Model Wizard including:
- When changing data source settings, all related settings in all portions of the model are automatically updated.
- The Replace Characters settings are now retained consistently when models are loaded and accessed.
- The Transaction Isolation Level now accepts a blank value to accommodate drivers that do not support it.
- Changes to Reports including:
- The ability to generate Reports with annotated images in the View.
- The proper presentation of association Lookup values within reports.
- The retention of additional user settings. For example, the last used Database Query mode (Basic or Advanced) and the Status Bar color selection are now retained according to user selections.
- The addition of a sample Hit List (./commonresources/HitList.txt) that can be imported using the List Management function under the Edit menu.
- The addition of a "null" option in the View Filter allowing users to filter data in the View on null values.
- Fixes to the Merge mode enabling users to use the UnMerge setting to restore manually merged objects to their original settings.
- Changes to the Edit User Label option to keep the edit box open when the cursor is moved to a different screen position.
- Updates to the Quick Data Load function that now allows import of text files with null values.
This release also addresses issues in the licensing process. It is possible that you may
need to request a new license when upgrading
to this release. Should you require a new license, simply send your licensing requirements
(Host ID, Number of users, etc.) to license@visualanalytics.com.
|
|
|
 |
 |
...that you can change label size for better resolution on output?
In the Bottom Toolbar (at the bottom of the View area), there is an icon that looks like a luggage tag followed by a Font Type and Size selector. These entries help format the look and feel of the objects contained in the View.
When there are a lot of objects present in the display, it can sometimes be difficult to see the actual labels. This is especially true when printing hardcopy. To achieve a better output, more emphasis can be placed on the labels. By selecting the Label Styles icon (e.g., the luggage tag) you can easily set the font styles and sizes.
For example, the following diagram shows a simple money laundering example with the default font size and style assigned by VisuaLinks. Although there are only 14 objects in this display and their labels are visible, they tend to be a bit hard to read. (For clarity, association labels have been turned off.)
Using the Label Styles to set the font type to Bold and the fort size to 12, a dramatic difference in the presentation and readability of the display is realized as is shown in the diagram below.
These changes are best made right before printing and may require some additional spacing between the objects because the larger labels may tend to overlap.
|
|
|
 |
The HTTP Search feature has been presented in several different scenarios in past LinkLetter issues and other materials
released by VAI over the past year. This feature has proven helpful for validating a wide range of information contained
in active investigations while providing additional content for final presentation.
For example, given the following sample data consisting of ORG, PHONE, and ADDRESS, additional information can be acquired
from different sources and incorporated directly into the diagram.
Right-clicking on any object within an active View displays a menu containing an entry called HTTP Search.
Selecting this option displays a window with several choices.
The top selection box contains the names of several popular Internet search engines. By default, we have included
Alta Vista, Google, and Yahoo. These entries are set in the property resources and can be
adjusted to include other search engines or proprietary systems.
The Current Search Selection table contains all the attributes for the selected object. Simply double-click on any
value in the left side of the table to move it to the right side. This sets the value as "selected" and indicates
that it will be submitted as the search.
Additionally, the Add to Current Selection field allows you to manually enter additional data as part of the search.
For this example, we selected the PHONE object and selected the actual phone number value to search on Google.
Since Google has a built-in phone book, the results show us that the subscriber of this phone number is indeed VAI.
If a result is not returned, it means that the phone number is not listed or is not valid.
Using this search feature is an excellent way to help validate data or find additional leads (when the number
is registered to a different person). Additionally, because this number was found, it also provides us an option
to map the data (MapQuest or Yahoo! Maps).
Choose either of these map options to view the associated map for the address. Then, right-click on the map and
choose the Save Picture As option (in the pop-up menu) to save the image to your file system. Make sure that the
format is set to GIF or JPEG - and not BITMAP (BMP).
Next, in VisuaLinks, click the Annotations (drawing) icon  in the
Bottom Toolbar and choose the Picture option at the top of the pop-up menu.
Your mouse cursor will change into a  at which point you need to select the area of the View where you want to
place the picture.
After selecting the location, a window displays asking where the image file is stored - choose Local. A file dialog
box will appear and you can now select the map file that you saved off previously.
Use the resizing feature (under the  Moves mode) to set the size of the map and place it in your diagram accordingly.
This makes a great enhancement to any finished intelligence product or report that you generate. Keep in mind that instead of using the PHONE object to initiate the Google search, the ADDRESS could also be selected to create the map (using STREET, CITY, ZIP for the attributes).
Give this a try on you next case/investigation - it makes a big difference in the quality of the results you can produce.
|
|
 |
When dealing with millions upon millions of records, VisuaLinks provides a number of unique ways to "slice & dice" the data to support proactive analytics. In the context of financial crimes analysis, there are certain types of situations that will always result in further evaluation. These typically include large numbers of transactions, frequent use of accounts, or connections to multiple ID numbers, addresses, or phone numbers.
The following example is based on SAR (Suspicious Activity Report) data using the Summarize service to expose a SUBJECT using multiple Social Security Numbers (SSNs) operating out of a specific LOCATION. The interesting aspect of this Summarize configuration was that the Count By selection used the SSN/EIN and the Group By was set to the NAME/DOB of the SUBJECT. Additionally, a filter was applied to only acquire records from a certain geographic region (defined by a CITY/STATE). Furthermore, a SUM aggregation was selected for the VIOLATION AMOUNT associated with the transaction.
The result set identified quite a number of potential targets and the query results were sorted by a combination of the number of SSN/EINs and the summed VIOLATION AMOUNT. Immediately, a specific target was identified and is shown below with 11 unique SAR violations.
A quick inspection of the NARRATIVES shows that all of the SARs reflect ongoing mortgage fraud(s). Generally, the banks indicated that there were misrepresentations in employment, overstated income, or falsified assets. With historically low interest rates, rising housing prices, and record new-home construction (9.4% in August 2004), the marketplace has been wrought with various frauds - especially mortgage fraud. The next level of the diagram is shown below indicating that there are at least 1-2 additional SUBJECTs per transaction involving our primary target.
From here, the networks were expanded again to bring in any ACCOUNT information associated with the SUBJECTs. Clearly the diagram indicates these subjects are all working independent of one another because each involves a separate account. The primary subject (in the center) is the controlling entity for all of these mortgage frauds.
Finally, the ADDRESSes for these SUBJECTs are displayed, which confirms they are all based on separate transactions. Notice, for each pairing of SUBJECTs for a SAR, there is a common address shown between them. Supposedly, these addresses would be the properties involved in the mortgage fraud.
Finally, a Temporal Grid was generated using the Day-of-Week (X-axis) and Week-of-Year (Y-axis) settings to show when these SARs took place. As depicted in the diagram below, there is a heavier concentration towards the end of the summer for this activity. Keep in mind that a SAR may contain multiple, independent transactions.
The behavior of our primary subject clearly shows the few transactions conducted earlier in the year tested the waters to determine if the scheme was detected by authorities. The last transaction occurred a week prior to writing this description, which indicates the fraud is presently active and our target is subject to prosecution from law enforcement.
Money laundering takes on many different characteristics. Whether people are structuring transactions to avoid the $10,000 filing limits, kiting checks, performing terrorist financing, or committing mortgage fraud - all of these types of violations are ultimately reviewed by the government agencies (e.g., FBI, IRS, FinCEN, DEA, etc.) chartered with investigating these types of crimes.
For more information on mortgage frauds, see the following CNN news reference:
http://edition.cnn.com/2004/LAW/09/17/mortgage.fraud/
|
|
|
 |
|
