November/December Release Notes¶
Aunalytics is excited to announce the last release of 2021 to our clients. This release will provide clients with model and site enhancement information along with any fixes to existing functionality we have included.
Daybreak¶
Natural Language Answers Autocomplete¶
This month we are releasing an exciting new feature to Daybreak users: autocomplete for Natural Language Answers™. Similar to the autocomplete behavior of search engines like Google, users will now see a list of their ten most recent searches when they place the cursor in the natural language search box, and when they begin typing, Daybreak will make suggestions based on what the user enters.
This feature will make it easier for users to ask questions both because it will suggest commonly searched after questions and offer autocompletion for field names that a user might not remember (for example, "age of account" versus account age").
Enhanced Information for Field Definitions¶
Daybreak users working with the structured Query Wizard and SQL editor have access to a field definitions tab (formerly called "Data dictionary") that lists the names of fields available in the datamart. This feature has helped users correctly identify the names of fields, but will now also show the data type (integer, floating point, string, etc.), and a brief description of the data point. These field descriptions or definitions help explain what a data point is and in some cases, how it is determined. For example, "CustomerAge" is "Integer representing the customer's age in years, dynamically generated each day from the customer's date of birth." This additional information will help Daybreak users understand their data marts so that they can create richer and more useful queries from them.
Workspace Descriptions¶
As we prepare to support increased numbers of workspaces for Daybreak in 2022, we've made a minor enhancement to the Daybreak app to help guide users to the workspaces they want to use. Now, when a user hovers the mouse cursor over the icon for a workspace, there will be a pop-up showing that workspaces description. While at present most Daybreak users only access one or two workspaces, our forthcoming release of scoped datamart views will increase the number of workspaces users can access and these descriptions will help them to better understand what view of the datamart each workspace offers.
Natural Language Answers Model Retrain¶
This month we have updated the Natural Language Answers model to support four new tables: Collateral, Branch, Lending, and Cards. While at present, not all Daybreak subscribers have chosen to implement all the tables in the data model, those who have subscribed to these four tables or a subset of them will have NLA support for all their fields, while subscribers without them will still be able to use NLA for the subset of fields that they have implemented.
We're excited about the large number of fields supported in this update and the new kinds of questions our users will now be able to ask!
| Table | Field | Example Question |
|---|---|---|
Collateral |
AccountValue |
Customers with 0 collateral account value |
Collateral |
AccountValue |
Customers with collateral account value 0 |
Collateral |
ApplicationCode |
Collateral with an application code of 70 |
Collateral |
AppraisalDate |
collateral appraisal date last month |
Collateral |
AppraisalValue |
collateral appraisal value above 5000 |
Collateral |
IsPledged |
Customers with pledged collateral |
Collateral |
IsPrimary |
Customers with no primary collateral |
Collateral |
MSACode |
Collateral with MSA Code 0 |
Collateral |
Make |
Collateral make is CHEVROLET |
Collateral |
Model |
Collateral Model is FUSION |
Collateral |
OriginalAppraisalDate |
Collateral with Original Appraisal Date this year |
Collateral |
OriginalAppraisalValue |
Customers with Original Appraisal Value above 10000 |
Collateral |
RecordingDate |
Collateral with Recording Date last week |
Collateral |
RelatedAccountType |
Collateral with Related Account Type 125 |
Collateral |
TotalValue |
Collateral with Total Value below 20000 |
Collateral |
TypeCode |
collateral type code is A1 |
Collateral |
VIN |
Collateral with VIN |
Collateral |
Year |
Collateral with Year 2016 |
Branch |
N/A | I want a list of branches |
Branch |
Branch_CloseDate |
branches closed date is last week |
Branch |
Branch_OpenDate |
branch opened date is last month |
Branch |
Branch_Name |
show me Topeka branch |
Branch |
WasAcquired |
branches are acquired |
Branch |
WasAcquired |
acquired branches |
Lending |
BoardingDate |
Loans with boarding date last month |
Lending |
BookBalance |
Loans with Book Balance over 1000 |
Lending |
ApplicationCode |
Loans with Federal Call Code 300 |
Lending |
NonAccrualDate |
Loans with Non Accrual Date last week |
Account |
OfficerID |
Loans with officer id 697 |
Lending |
OriginalPrincipalAmount |
Loans with Original Principal Amount above 500 |
Lending |
OriginalTerm |
Loans with Original Term below 120 |
Lending |
PaymentFrequencyCode |
Loans with Payment Frequency Code 4 |
Cards |
ATMDailyWithdrawalLimit |
Customer with ATM daily withdrawal limit 5000 |
Cards |
ATMTransactionWithdrawalLimit |
Customer with ATM Transaction withdrawal limit above 5000.00 |
Cards |
BIN |
Customer with bin number 551336 |
Cards |
StatusChangeDate |
Customers changed cards status after 2019 |
Cards |
Cards_Status |
cards status is active |
Cards |
CloseReason |
Cards with close reason 0 |
Cards |
LastTransactionDate |
cards expired last week |
Cards |
IsCorporateCard |
Customers with Corporate Cards |
Cards |
IsPrimaryAccount |
show me primary cards |
Cards |
IsSavingsCard |
Customers with savings cards |
Cards |
LastTransactionDate |
Customers with cards last transaction this week |
Cards |
LastUsageAmount |
Cards with last usage amount below 100 |
Cards |
OriginalFirstUsageDate |
Customers with original first usage after 2000 |
Cards |
OriginalCardIssueDate |
Customers with Original Card Issue Date this week |
Lending |
IsInterestRateVariable |
How many loans with variable interest rate |
Lending |
OriginationDate |
loans with an origination date between April 1st, 2020 and March 31st, 2021 |
Lending |
LoanType |
Loans with loan types 135 |
Lending |
CurrentLoanStatus |
delinquent loans |
Lending |
CurrentLoanStatus |
chargeoff loans |
Lending |
ProductCode |
number of credit card loans |
Lending |
ProductCode |
list of agricultural loans |
Cards |
IsDebitCard |
How many members have debit cards |
Cards |
IsDebitCard |
How many debit cards do we have |
Cards |
IsCreditCard |
Number of credit cards |
Cards |
ProductType |
Cards are ATM type |
Cards |
ProductType |
I want a list of cards with product type ATM card |
Aunsight™ Golden Record¶
Log4j Vulnerability Patch¶
This month, we have reviewed our platform infrastructure to address an evolving security threat in one of our software dependencies: Apache Log4j, a Java logging utility. A recently discovered security vulnerability with this library can allow a bad-actor to execute arbitrary code injected into log messages generated by an attacker-controlled LDAP server. After a thorough review of our infrastructure, it was determined that the Log4j vulnerability only affected one of our underlying systems, and a patch has been applied to secure our platform from this ongoing security threat. As reports of successful attacks continue to appear in the news, users can rest assured that AuGR has been hardened to prevent an attack against the platform.