Skip to main content
Version: 4.3

Learn about input and output data

The Adverse Media Monitoring AI Agent supports a variety of ways for seamless integration into each financial institution’s ecosystem, such as client lifecycle management (CLM) systems, news sources, and case management archives, including through RPA and API connections.

Search input

The Adverse Media Monitoring search is triggered by specifying a name via the Ad Hoc search in Workspace or a batch upload of names set up and scheduled via Control Tower. To conduct a search, a name is required.

To configure a batch process, WorkFusion can be integrated with the bank’s internal systems to mimic how analysts are currently retrieving names that need to be monitored for adverse media events.

Examples include:

  • Navigating alert queues in a transaction monitoring system and conducting searches on a non-customer in a potentially unusual transaction.
  • Identifying customers who meet certain criteria within a CLM system. For example, you have a customer rated as high-risk, and yet the most recent Know-Your-Customer refresh is more than one year ago.
  • New applications created within a customer onboarding system or business line product extension system (such as a new lending application).
  • A CSV file sent to a shared email inbox or uploaded to a shared folder.

You can also trigger a search via REST API that passes all required inputs as parameters and then receives the results from WorkFusion.

CSV input

The data input format for the Adverse Media Monitoring File Ingestion Business Process is a CSV file with simple fields and one or more rows with entity names to search.

Below is an example of the CSV file with possible content. You can also download the template when configuring the External Source as the AMM input.

Input data schema

Only the search_request field is obligatory. All other fields are optional and are used to increase the relevancy of search results.

FieldDescriptionTypeRequiredExample
search_requestEntity name. It is a person or a company subject to Adverse Media Monitoring.StringYES
  • Danske Bank
  • John Doe
genderEntity gender.StringNO
  • MALE
  • FEMALE
entity_typeEntity type. Currently, two possible values are supported (non-case-sensitive):
  • Company
  • Individual
StringNO
  • INDIVIDUAL
  • COMPANY
residence_operating_countryTwo-letter country code. Supported format: ISO 3166-1 alpha-2. See the full list here.StringNOUS (United States) / CA (Canada) / AU (Australia)
subdivisionTwo-letter subdivision code according to ISO 3166-2.StringNO
  • US-FL (Florida state of the USA)
  • US-CA (California state of the USA)
  • CA-QC (Quebec province of Canada)
  • CA-YT (Yukon territory of Canada)
citizenship_incorporation_countryTwo-letter country code. Supported format: ISO 3166-1 alpha-2. See the full list here.StringNOUS (United States) / CA (Canada) / AU (Australia)
cityFull city’s name. Used only for Thomson Reuters CLEAR Investigations.StringNONew York
year_of_birth_incorporationYear of birth for a single screened person. Not applicable when entity_type is Company.StringNO1980
date_of_birth_incorporationDate of birth for a single screened person. Not applicable when entity_type is Company. List of accepted date formats:
  • MM/dd/yyyy
  • yyyy/MM/dd
  • MM-dd-yyyy
  • yyyy-MM-dd
  • dd MMM yyyy
  • dd MMMM yyyy
The date format is selected in the course of configuration.
IntegerNO
  • 02/25/1980
  • 1980/02/25
  • 02-25-1980
  • 1980-02-25
  • 25 Feb 1980
  • 25 February 1980
additional_informationAdditional information about the screened entity.StringNOJohn Doe has a brother Edward Doe
add_to_ongoing_monitorThe field is used only when Thomson Reuters is configured as the data provider. If true, the corresponding entity will be automatically added to the ongoing monitoring scope.BooleanNOtrue/ false
_sys_mt_hit_assigned_user_groupsThe name of the user group assigned to manage the manual review task in Workspace.StringNOReviewers
article_dataThe field is used only when the External Source is selected as the input data provider. It represents stringified JSON with an array of article data (links and additional information). For the full schema, see the Article data item fields table.StringNO[{\"link\":\"https://www.politico.com/news/2024/08/27/nancy-pelosi-jan-6-fury-00176529\",\"additional_information\":\"article level information\"}, {\"link\":\"https://www.bbc.com/news/articles/czdpq4en1vvo\"}]
entity_idA unique ID of the screened entity used to reapply decisions on the content previously reviewed by an analyst. Free text, in any format.StringNOE1234567
batch_idA unique ID to group multiple search entities into combined reports. To be sent together with the batch_size field.StringNOUnique free text in any format. See some examples below:
  • BI123
  • Batch01
  • Group01
batch_sizeThe number of investigations in the batch. To be sent together with the batch_id field.IntegerNO15
external_provider_nameThe name of the provider used for collecting all articles.StringNOGOOGLE
external_idA unique ID to identify the search request. Unique free text, in any format.StringNOR1234567
cost_center_codeCode of the Cost Center team for billing transactions.StringNOR1234567
cost_center_nameName of the Cost Center team for billing transactions.StringNOR1234567
sla_dateService Level Agreement (SLA) date for filtering in Manual Tasks.StringNOThe date format is selected in the course of configuration.
business_lineIdentifies the business line for better prioritization.StringNOFinance
screening_serviceCustom screening service name.StringNOAdvNews
investigation_dateSpecific investigation date.StringNOThe date format is selected in the course of configuration.
initiatorThe name of the request initiator.StringNOJohn Doe
sla_durationSLA duration in a text format.StringNO5 hours
related_entity_idRelated entity ID.StringNOREI123

Article data item fields

FieldDescriptionTypeRequiredExample
linkLink to the website from which an article is to be downloaded.StringYES if content and content_link are null.https://example.com/path/to/article
contentArticle content. If both link and content are provided, content is overwritten with the content from link, provided the latter is successfully downloaded.StringYES if link and content_link are null.A large text that represents the article content in HTML or a plain text format.
content_linkLink to the article content. If both link and content_link are provided, content_link is overwritten with the content from the link, provided the latter is successfully downloaded. If both content and content_link are provided, content is ignored.StringYES if link and content are null.https://s3.web-server.com/bucket/path/article-content.txt
titleArticle title.StringNOUnlocking the Secrets of Everyday Life: How Small Habits Shape Big Success
summaryArticle summary.StringNOIn a world driven by goals and aspirations, the importance of daily habits often goes unnoticed. Yet, research suggests that small, consistent actions can lead to transformative results.
additional_informationPlain text or part of HTML containing additional information that can be helpful during the article review.StringNO
  • Industry: Technology and Software Development
  • Founded: 2015
  • Headquarters: San Francisco, California, USA
external_provider_nameThe name of the provider used for collecting the article.StringNOUltimate Article Searcher
investigation_statusIf an article is initially reviewed with Evelyn NSS, you can pass the NSS decision and it will be considered by Evan AMM.StringNO
  • NEEDS_INVESTIGATION
  • FALSE_POSITIVE
  • TRUE_POSITIVE
external_review_commentIf an article is initially reviewed with Evelyn NSS, you can pass the NSS comment and it will be considered by Evan AMM.StringNOThis article was reviewed by Evelyn NSS as False Positive.

Quality control input schema

FieldDescriptionTypeRequiredExample
investigation_uuidUnique identifier of a previous investigationStringYES123e4567-e89b-12d3-a456-426614174000

Input data validation

Where applicable, validations are configured as part of the validation settings on the Input step.

Core fields and validations

The table below summarizes all core, date, location, article, and numeric field validations.

FieldValidation requirement and rulesError message
search_requestRequired
  • Non-blank
  • Max length 256
  • The search_request field is required but was not provided or was blank.
  • The search_request exceeds the maximum length of 256 characters. Provided length: <len>.
entity_idOptional
  • Max length 256
The entity_id exceeds the maximum length of 256 characters. Provided length:<len>.
entity_typeOptional
  • Allowed values are individual,company. The values are case-insensitive.
Invalid value for entity_type. Accepted: individual or company. Provided:<value>.
genderOptional
  • Allowed values are male,female. The values are case-insensitive.
The gender field must contain either Male or Female. Gender:<value>.
date_of_birth_incorporationLocalDate
  • Parsed using the platform date parser (optional preferred format).
  • Allowed year range is >= 1900 (if entity_type is individual), not in the future.
  • The <field> has an incorrect format. Provided: <value>.
  • <field> is out of range. Year cannot be before 1900/future. Provided: <year>.
sla_dateLocalDateTime
  • Must use the ISO 8601 format: yyyy-MM-dd'T'HH:mm:ss.
  • The <field> has an incorrect format. Use ISO 8601 datetime (yyyy-MM-dd'T'HH:mm:ss).
investigation_dateLocalDateTime
  • Must use the ISO 8601 format: yyyy-MM-dd'T'HH:mm:ss.
  • The <field> has an incorrect format. Use ISO 8601 datetime (yyyy-MM-dd'T'HH:mm:ss).
year_of_birth_incorporation
  • Must be integer (if present).
  • Year range is >= 1900 if entity_type is individual, less than the current year (for all).
  • The <field> must be a number. Provided: <value>.
  • Mismatch between <field1> and <field2>: <v1> vs <v2>.
residence_operating_country
  • Must be a valid ISO-2 code from the Country repository.
  • Controlled by the RESIDENCE_OPERATING_COUNTRY_VALIDATION flag.
  • Invalid residence/operating country. Please ensure it is valid.
subdivision
  • Applies if SUBDIVISION_VALIDATION is enabled.
  • Requires a valid country.
  • Valid if it matches the ISO-3166 code, country-specific alpha code, or unique subdivision name (in BP_LANGUAGE).
  • Subdivision requires a valid residence country. Provided: [<countryField>: <countryCode>, subdivision: <value>].
  • Invalid subdivision/state: <value>. Please ensure it is valid.
  • Multiple matches for subdivision <value>. Use ISO-2 code.
city
  • Applies if CITY_VALIDATION is enabled.
  • If city is present, residence_operating_country must also be present.
  • The residence or operating country is required to perform a search by city. Provided city:<city>.
article_data
  • Each element must contain at least one non-blank value: content, link, contentLink.
  • article_data contains element that does not contain link or content!
batch_size
  • When provided as strings, the values must be valid integers.
  • The <field> field must be a number. Provided: <value>.

Configuration flags

FlagTypeDefaultDescription
RESIDENCE_OPERATING_COUNTRY_VALIDATIONBooleantrueValidates ISO-2 country codes for residence and operating country fields.
CITIZENSHIP_INCORPORATION_COUNTRY_VALIDATIONBooleantrueValidates citizenship country codes.
SUBDIVISION_VALIDATIONBooleantrueEnables the subdivision and state validation.
CITY_VALIDATIONBooleantrueRequires a country when a city is specified.
BP_LANGUAGEString-Used to resolve subdivision names in a given language.
PREFERRED_DATE_FORMATString-Used in parsing and suggested in date-related errors.

Error message context

All validation errors are wrapped with a context generated from externalId and entityId to help locate the failed record.

Search output

Once an entity search is completed, the results are displayed in a risk-prioritized manner, and you can review the results.

Once you complete the review, the AI Agent generates a PDF report audit trail of the identified news, including the confidence score per article, the final decision, and other data.

By default, the audit trail report is stored inside the WorkFusion platform. It is possible to configure this report to be pushed directly into your internal applications, for example:

  • Save to a case management or transaction monitoring system
  • Store in a shared folder
  • Use a custom connector to integrate it into an in-house application

Output data schema objects

Investigation data schema

FieldDescriptionTypeExample
uuidA unique identifier for the investigationString123e4567-e89b-12d3-a456-426614174000
runUuidA unique identifier for the execution runString123e4567-e89b-12d3-a456-426614174001
batchIdA unique ID to group multiple entities into a combined reportStringB1234567
entityIdA unique ID for identifying entityStringE1234567
externalIdA unique ID to identify the search requestStringR1234567
tenantIdA unique identifier for the tenantStringtenant-123
variationIdVariation identifierInteger1
versionVersion of the AMM AI AgentString4.0.2
statusCurrent status of the investigationStringSTARTED / IN_PROGRESS / READY_FOR_REVIEW / REVIEWED / COMPLETED / FAILED / MT_EXPIRED
inputStartTimeTimestamp when the input processing startedDateTimeS2024-03-18T10:15:30Z
inputEndTimeTimestamp when the input processing endedDateTime2024-03-18T10:16:30Z
processingStartTimeTimestamp when the main processing startedDateTime2024-03-18T10:16:31Z
processingEndTimeTimestamp when the main processing endedDateTime2024-03-18T10:20:30Z
reviewStartTimeTimestamp when the review startedDateTime2024-03-18T10:20:31Z
reviewEndTimeTimestamp when the review endedDateTime2024-03-18T10:25:30Z
outputStartTimeTimestamp when the output processing startedDateTime2024-03-18T10:25:31Z
outputEndTimeTimestamp when the output processing endedDateTime2024-03-18T10:26:30Z
originalSearchRequestOriginal search entity name as providedStringJohn Doe
searchRequestProcessed search entity nameStringJohn Doe
genderEntity's genderStringMALE / FEMALE
entityTypeType of the search entityStringCOMPANY / INDIVIDUAL
cityFull city nameStringNew York
residenceOperatingCountryTwo-letter country code in the ISO 3166-1 alpha-2 formatStringUS
citizenshipIncorporationCountryTwo-letter country code in the ISO 3166-1 alpha-2 formatStringCA
subdivisionISO 3166-2 code of the subdivisionStringUS-FL
yearOfBirthIncorporationYear of birth for an individual or year of incorporation for a companyInteger1980
dateOfBirthIncorporationDate of birth for individual or incorporation for a companyDate1980-05-15
htmlReportLinkLink to the HTML reportStringhttps://example.com/reports/123.html
pdfReportLinkLink to the PDF reportStringhttps://example.com/reports/123.pdf
investigationStatusThe resulting status of the investigationStringFALSE_POSITIVE / TRUE_POSITIVE / NEEDS_INVESTIGATION / NO_RESULTS_FOUND
restExecutionIndicates if the execution is via REST APIBooleantrue / false
assigneeUser assigned to the investigationStringjohn.smith@example.com
assignedUserGroupsThe name of the user group to which a manual review task is to be assignedStringreviewers
additionalInformationAdditional information of the screened entityStringJohn Doe has brother Edward Doe
addToOngoingMonitorIndicates if an entity should be added to ongoing monitoringBooleantrue / false
costCenterCodeCode of the Cost Center team for billing transactionsStringCC12345
costCenterNameName of the Cost Center team for billing transactionsStringRisk Department
batchSizeThe number of investigations in the batchInteger15
providersArray of providers used in the investigationArray[{"provider": "GOOGLE", "active": true}]
articleCountThe number of articles foundInteger25
manualTaskManual Task detailsObject{"status": "NEW", "assignee": "john.smith@example.com"}
investigationDateInvestigation date from the input or BP start timeDateTime2022-02-23T21:00:25Z
initiatorThe name of the request initiatorStringJohn Doe
sla_durationSLA duration in a text formatString5 hours
related_entity_idRelated entity IDStringREI123

Article schema

FieldDescriptionTypeExample
uuidA unique identifier for the articleString123e4567-e89b-12d3-a456-426614174002
newsProviderProvider of the news articleStringGOOGLE / FACTIVA_HEADLINES / LEXIS_NEXIS_LP_MEDIA / THOMSON_REUTERS_CLEAR / WORLD_CHECK_ONE / EXTERNAL_SOURCE
duplicateArticleUUIDUUID of the duplicate article if foundString123e4567-e89b-12d3-a456-426614174003
titleArticle titleStringExample News Article
statusProcessing status of the articleStringPROCESSED / UNPROCESSABLE / EXCLUDED
publishingDateDate when the article was publishedDate2024-02-15
summaryArticle summaryStringBrief summary of article content
authorArticle authorStringJane Smith
languageLanguage of the articleStringEN / ES / UNKNOWN
wordCountNumber of words in the articleInteger850
sourceSource of the articleStringThe New York Times
originalArticleLinkOriginal link to the articleStringhttps://example.com/article/12345
downloadedArticleLinkLink to the downloaded articleStringhttps://storage.example.com/12345.html
parsedArticleLinkLink to the parsed articleStringhttps://storage.example.com/12345.html
taggedArticleLinkLink to the tagged articleStringhttps://storage.example.com/12345.html
investigationStatusStatus of the investigation for the articleStringFALSE_POSITIVE / TRUE_POSITIVE / NEEDS_INVESTIGATION
adjudicationReasonReason for adjudicationStringNON_ADVERSE / NON_FOCAL / YOB_MISMATCH / ENTITY_IN_ARTICLE_IS_OLDER_THAN_SCREENED_ENTITY / ENTITY_WAS_NOT_FOUND / DUPLICATE / ARTICLE_PARSING / UNSUPPORTED_LANGUAGE / UNSUPPORTED_FORMAT / BEHIND_PAYWALL / HUGE_ARTICLE / POOR_ARTICLE / EMPTY_ARTICLE / DOWNLOAD_ARTICLE_EXCEPTION / WEBSITE_IGNORE / ARTICLE_ON_SANCTIONS_SITE / NSS_DECISION / LLM_RESPONSE_FALSE_POSITIVE / LLM_RESPONSE_NEEDS_INVESTIGATION
mlScoreMachine learning scoreNumber0.87
taggedTitleTagged version of the titleStringExample tagged title
taggedSummaryTagged version of the summaryStringExample tagged summary
taggedHighRiskCountriesTagged high-risk countriesStringUS / UK /RU
taggedKeywordsTagged keywordsStringfraud / money laundering
taggedPoliticallyExposedPersonsTagged politically exposed personsStringJohn Smith / Jane Doe
taggedEntitiesTagged entitiesStringABC Corp / XYZ Ltd
errorCodeError code if processing failedStringARTICLE_DOWNLOAD / ARTICLE_CONTENT_EMPTY / UNSUPPORTED_LANGUAGE
historyIdHistory identifierStringhistqwdwqdund234dwqdwqd21und2345hqwe12d2345
additionalInformationAdditional information about the articleStringContains references to subsidiaries
reviewCommentComment from the reviewStringNot relevant to the search entity
extractedYearOfBirthIncorporationExtracted year of birth or incorporationInteger1981
providerSearchUrlURL used for the provider searchStringhttps://search.example.com?q=John+Doe
articleReviewArticle review detailsObject{"workerFullName": "John Reviewer", "investigationStatus": "FALSE_POSITIVE"}
previousReviewPrevious review detailsObject{"workerFullName": "Jane Reviewer", "investigationStatus": "NEEDS_INVESTIGATION"}
contentParsedByTool used to parse the contentStringPOSTLIGHT / TRAFILATURA
authorParsedByTool used to parse the authorStringPOSTLIGHT / TRAFILATURA
publishingDateParsedByTool used to parse the publishing dateStringPOSTLIGHT / TRAFILATURA
titleParsedByTool used to parse the titleStringPOSTLIGHT / TRAFILATURA
externalProviderNameName of the external providerStringGOOGLE
downloadModeMode used for downloadingStringSTANDARD / SCRAPE_DO_STANDARD / SCRAPE_DO_BROWSER_RENDERING
translatedTitleTranslated version of the article titleStringSample translated title
translatedSummaryTranslated version of the article summaryStringSample translated summary
translatedArticleLinkLink to the translated articleStringhttps://storage.example.com/12345.html
originalLanguageArticle's original detected languageStringen, es, fr, de, fi
investigationDateInvestigation date from the input or Business Process start timeDateTime2022-02-23T21:00:25Z
initiatorThe name of the request initiatorStringJohn Doe
sla_durationSLA duration in a text formatString5 hours

Article review schema

FieldDescriptionTypeExample
workerUUIDUUID of the worker who performed the reviewString123e4567-e89b-12d3-a456-426614174004
workerFullNameFull name of the workerStringJohn Smith
workerEmailEmail of the workerStringjohn.smith@example.com
investigationStatusStatus determined by the investigationStringFALSE_POSITIVE / TRUE_POSITIVE / NEEDS_INVESTIGATION
commentReview commentStringNot relevant to the entity being searched
reviewTimeTimestamp of the reviewDateTime2024-03-18T14:30:45Z
shorthandDecisionList of decisionsStringDecision 1 / Decision N

Manual Task schema

FieldDescriptionTypeExample
workerUUIDUUID of the worker assigned to the taskString123e4567-e89b-12d3-a456-426614174005
workerFullNameFull name of the workerStringJohn Smith
workerEmailEmail of the workerStringjohn.smith@example.com
statusStatus of the Manual TaskStringNEW / IN_PROGRESS / SUBMITTED / EXPIRED
investigationStatusStatus determined by the investigationStringFALSE_POSITIVE / TRUE_POSITIVE / NEEDS_INVESTIGATION
reviewCommentComment from the reviewStringNo relevant findings detected
reviewTimeTimestamp of the reviewDateTime2024-03-18T14:30:45Z
processingPeriodSecondsTime spent processing in secondsInteger300
assigneeUser assigned to the taskStringjohn.smith@example.com
assignedUserGroupsUser groups assigned to the taskStringreviewers
selectedForAutoQcIndicates if the task is selected for automatic quality controlBooleantrue / false
shorthandDecisionList of decisionsStringDecision 1 / Decision N

Provider schema

FieldDescriptionTypeExample
providerProvider nameStringGOOGLE / FACTIVA_HEADLINES / LEXIS_NEXIS_LP_MEDIA / THOMSON_REUTERS_CLEAR / WORLD_CHECK_ONE / EXTERNAL_SOURCE
externalProviderNameName of the external providerStringGOOGLE
externalArticlesDataArray of external article dataArray[{"articleLink": "https://example.com/article/123", "title": "Example Article"}]
keywordsArray of keyword dataArray[{"keywords": "fraud,laundering", "enableKeywordThesaurus": true}]
activeIndicates if the provider is activeBooleantrue / false
highRiskCountriesHigh-risk countries for the providerStringUS / UK / RU
daysSearchPeriodThe number of days to searchInteger90

Keyword data schema

FieldDescriptionTypeExample
keywordsKeywords to search forStringfraud / laundering / bribery
enableKeywordThesaurusEnables the keyword thesaurusBooleantrue / false