Data from The GDELT Project · Reviewed and interpreted by RSRC
Search the research collection
Find reporting by topic, place or language. Open a record to see why it was included and what the available evidence supports.
The chart uses every search filter above and all matching records, across every results page. The percentage denominator is all included URLs in the same dates, before theme, language, country or text filters.
Distinct included URLs matching these filters in each observation month. Select a point to explore its records. On small screens, scroll the chart horizontally.
A URL counts once per month and can appear in several months or themes. Themes describe the current record assessment; months show URL observations, not when events or each claim first appeared. Warnings, allegations and reported experiences are all included when coded. Partial months, collection gaps and changes in sources affect both measures. A higher share does not establish a higher risk to victims.
| Observation month | Matching URLs | All included URLs | Share |
|---|---|---|---|
| Jan 2025 | 352 | 591 | 59.6% |
| Feb 2025 | 489 | 1,629 | 30.0% |
| Mar 2025 | 209 | 561 | 37.3% |
| Apr 2025 | 162 | 390 | 41.5% |
| May 2025 | 232 | 398 | 58.3% |
| Jun 2025 | 86 | 209 | 41.1% |
| Jul 2025 | 272 | 522 | 52.1% |
| Aug 2025 | 253 | 447 | 56.6% |
| Sep 2025 | 273 | 613 | 44.5% |
| Oct 2025 | 251 | 483 | 52.0% |
| Nov 2025 | 174 | 392 | 44.4% |
| Dec 2025 | 200 | 333 | 60.1% |
| Jan 2026 | 222 | 530 | 41.9% |
| Feb 2026 | 447 | 1,106 | 40.4% |
| Mar 2026 | 175 | 360 | 48.6% |
| Apr 2026 | 239 | 491 | 48.7% |
| May 2026 | 171 | 785 | 21.8% |
| Jun 2026 | 214 | 842 | 25.4% |
| Jul 2026 | 469 | 1,143 | 41.0% |
| Aug 2026 | 319 | 640 | 49.8% |
| Sep 2026 | 373 | 638 | 58.5% |
| Oct 2026 | 17 | 30 | 56.7% |
Each record represents a source web page, not a verified incident. Different URLs can describe the same event. About these records.
Headline reports a romance victim losing GBP72,000 while the opening reports broader106million annual loss or FCA missed chances. Retain the headline’s individual amount and regulatory context without assuming72,000is the entire aggregate or names/bank misconduct. Last financial year is not a precise incident date.
Research & trendsFCA reportedly found missed bank opportunities to prevent romance scams with aggregate GBP106million annual losses. Retain warning and source aggregate financial context; unanchored last year is not a calendar year, missed opportunity is not a court-proved failure or implemented legislative change.
Prevention & supportUK romance-fraud losses reportedly exceeded GBP100million, roughly NTD4.1billion, and cases increased nearly10percent last year, prompting regulator warnings to banks. Retain attributed aggregate trend and warning; currencies are source conversions, last-year precision is unanchored and incidence is not independently verified.
Research & trendsTakanezawa66-year-old man reported JPY5.6million loss after a dating-site claimed Japanese woman, SNS intimacy and an online-shop owner opportunity; Sakura police investigate as romance fraud. Retain online/platform, investment opportunity, loss/report/inquiry without promised return, exact year or false nationality from a mere claimed persona.
Reported caseHeadline reports one Brit losing GBP428,000; the opening says love/friendship fraud cost Brits over100million last year. Keep individual versus aggregate scope and source claims distinct, without independent prevalence verification or a period derived from observation. A lonely descriptor is not a proved causal vulnerability, diagnosis or named case bridge. The headline explicitly calls its individual victim lonely; retain only that qualified source descriptor, not a diagnosis, causal risk claim or condition of every British victim.
Research & trendsHeadline reports a romance victim losing GBP72,000 while the opening reports broader106million annual loss or FCA missed chances. Retain the headline’s individual amount and regulatory context without assuming72,000is the entire aggregate or names/bank misconduct. Last financial year is not a precise incident date.
Research & trendsRoundup’s substantive FCA paragraph reports missed red flags and requests stronger bank monitoring, with one428,000loss case, another403payments/72,000case and wanted crypto transfer to a partner in Iraq. Retain general warning and multiple-source examples without summing them or claiming all are independent victims. Intended crypto is planned, Iraq is partner claim and403payments alone is not403explicit demands.
Prevention & supportTaiwan/US authorities reportedly broke a cross-border romance-fraud group with1.7hundred-million alleged loss; Liao was stopped at Taoyuan airport early in the year and a50million-value watch seized. Bail request and two-month continued custody are not conviction/sentence. Preserve the source’s currency and partial context; do not borrow passports, account roles, US victim residence or exact calendar year from a fuller variant.
Reported casePolice reportedly arrested two Utrecht women in a dating-fraud inquiry with at least26victims and total loss exceeding EUR550,000. Retain aggregate loss/arrest/inquiry, no exact date, defendant ages or false-name/dating-platform details borrowed from fuller variants.
Reported caseTaiwan prosecutors describe Liao36, Taiwan/US romantic fraud, stolen passports/bank accounts, small-profit bait, large repeated demands and US-national targets, yielding USD5,328,855/NTD170.52million attributed aggregate proceeds. Reported airport arrest,50million watch seizure and pending criminal charges remain allegations; bail denied and two-month custody extension fromOctober21 are not conviction or sentence. Passport-owner death by unrelated sedation is not romance-caused homicide. Nationality does not prove victim residence. Stolen passports for bank accounts are document fraud/account facilitation, not evidence of a romantic profile identity or online communication medium. Small initial investment profits are not an explicit future-return promise. False-friendship/sweet-word relationship interpretation remains qualified; custody extension is not a sentence and the passport owner’s unrelated death is not romance homicide. Translate the provisional document-fraud shorthand Oo to the existing codebook builder alias Doc with the same document_identity_fraud subtype. The strict alias preflight caught the shorthand before saving/building; original draft and failed helper are preserved, with no codebook change.
Reported caseDutchNews headline reports two women arrested for dating fraud yielding EUR550,000. Retain attributed aggregate financial loss and arrest, without names, Utrecht location,26/41ages, false persona or contact-platform detail borrowed from fuller Dutch sources.
Reported caseIna police classify a50s man’s JPY4million crypto loss as SNS investment/romance fraud; partial source gives a purported female contact. Preserve explicit mixed police classification and financial/crypto/SNS/investment context, without asserting an individual romantic bond from the combined label, false persona or an incident year.
Related mentionHückelhoven man reportedly lost a substantial sum to an internet love-scam acquaintance and police warn readers. Retain financial/online and advice context. US acquaintance claim in the headline is a persona detail, not actual operation country, victim nationality, exact loss or age56borrowed from another source.
Reported casePrahova61-year-old woman reportedly lost EUR170,000in a virtual love-scam relationship with a Nigerian man said to reside in Ilfov and claim illness. Retain money/online relationship/medical-pretext attribution; no actual sickness, Nigerian operation location, victim health injury, conviction or repeated demands inferred. Alleged offender residence is not automatically an operation-country assertion. The virtual-love fraud framing supports only qualified false-courtship interpretation. Claimed illness is not proved physical harm or a false nationality; offender residence is not automatically an operation location.
Reported caseTwo Utrecht women ages26/41were reportedly arrested in a dating-fraud inquiry, with at least26people filing reports between June2022and October of this year and total loss overEUR550,000. Retain arrest, complaint period, minimum victim count and aggregate loss. Unanchored October is not assigned2025; ages26/41are not victim counts and offences may predate reports. June2022is the explicit start of the complaint-report period. October of this year remains unanchored, not filled from observation; the period is not necessarily when all offences occurred.
Reported caseA32-year-old Hyderabad resident reportedly lost Rs650,000in online dating/friendship fraud. Retain attributed loss and online-dating context, not paid-service details, false love/persona, police action, psychological injury or the victim’s actual currency beyond sourceRs.
Reported caseDutch source says two Utrecht women arrested for dating fraud exceeding EUR550,000, approaching targets through a dating platform under false names and building bonds to obtain money or goods. Retain alleged financial loss, false persona/trust, platform and arrest. Goods requested are not proof of completed property loss; no conviction or exact dates. Utrecht defendant origin is not every victim’s residence.
Reported caseDutch headline reports overhalf-million euros dating-fraud proceeds, with Utrecht women26/41under arrest/inquiry. Retain alleged financial/arrest/investigation context; exact550,000,26victims and false names/dating medium occur only in other variants.
Reported caseUS/UK action addresses cyber-scam networks luring workers through deceptive jobs or fake romantic relationships, forcing extraction of billions and laundering. Retain mixed attributed coercive recruitment/false courtship/network/financial context, not every worker voluntary or all global billions romance-only. No independent proof of sanctions, coercion or seizure beyond saved attribution. Coercion concerns recruited/controlled workers. Financial totals refer to a range of deceptions, not every worker a voluntary offender or every extracted dollar romance-only.
Related mentionKuroishi police say an Aomori woman in her30s lost JPY610,000 in SNS-romance fraud; September6dating-app contact claimed to be Korean Namuru. Retain financial loss, platform and claimed persona without assumed fake nationality, operation in Korea or a year inferred from the14th/September.
Reported caseUK/US reportedly seized GBP11.3billion Bitcoin from Cambodian Chen Zhi and froze London property, linked to romance fraud and GBP92million British2024loss. Retain attributed mixed asset/organization/financial context; asset value is not per-victim loss or compensation and UK2024aggregate is distinct from106million2024/25reporting. No actual victim crypto-payment medium inferred. The saved source reports a group action and a separate2024British aggregate; it does not explicitly classify mixed fraud types. Keep those different scopes and the92million versus106million fiscal-period comparison separate.
Related mentionA false lover reportedly took money from a64-year-old Bochum woman and extorted her with private photos. Retain financial loss, fabricated romance and private-image extortion; no nude/sexual content, exact sum, image publication, suicide or diagnosed distress inferred.
Reported caseEschach66-year-old reportedly transferred money to a purported online acquaintance in love fraud and police offer protection advice. Retain loss/online/warning and explicit age/place, not a precise sum, actual US persona or year borrowed from another item.
Reported caseOkinawa police reportedly arrested an Uruma45-year-old worker suspected of SNS romantic-affection deception and JPY100,000from a Saga woman in her50s. Retain allegations/loss/fabricated relationship/SNS/arrest; no disclosed confession, conviction or full date from14th. A claimed first local detection is not no earlier fraud ever.
Reported caseLSE convicted of laundering/receiving crime proceeds linked to an elderly Maltese woman reportedly sending EUR38,650to accounts after a man posed as an American doctor in Yemen. Retain named occupational role, victim loss/account facilitation/conviction and false persona. Malta nationality does not alone prove residence, Yemen/US are persona claims, and laundering participant is not automatically a coercion victim. Victim payments to multiple accounts do not establish that she supplied an account to facilitate other people’s money. The convicted educator’s laundering is reported; separate her facilitation from the elderly woman’s38,650loss and claimed doctor/Yemen persona.
Reported case