Watchlists: OFAC · UN · EU · HMT · PEP
SCREENING ACTIVE
--:--:-- IST
Sanctions & PEP Screening Automation Engine
Kishore U. · github.com/ukishore33 · linkedin.com/in/kishore-techie  |  rapidfuzz · Precision 97.1% · Recall 100% · F1 98.6%
Total Screened
86
Names in queue
HIGH Alerts
33
Score ≥ 90
MEDIUM Alerts
2
Score 75–89
For Review
4
Score 60–74
Clear
47
Score < 60
PEP Matched
9
Politically exposed
Sanctions Hit
26
OFAC/UN/EU/HMT
Alert Level Distribution
Match Score Distribution
≥ 90 HIGH
33
75–89 MED
2
60–74 REV
4
< 60 CLEAR
47
By Match Type
Watchlist Breakdown
OFAC-SDN14
PEP · Tier 1 Senior6
EU-Consolidated5
UN-Consolidated4
HMT-UK3
PEP · Other Tiers3
Screening Performance Metrics
Precision (HIGH + MEDIUM alerts) 97.14%
Recall (zero missed hits) 100.00%
F1 Score 98.55%
True Positives 34
False Positives (false alerts) 1
False Negatives (missed hits) 0 ✓
Watchlist entries (names + aliases) 55
Algorithm rapidfuzz · token_sort_ratio
34
TRUE POSITIVE
1
FALSE POSITIVE
0
FALSE NEGATIVE
51
TRUE NEGATIVE
Fuzzy Match Technique
ALGORITHM
rapidfuzz.fuzz.token_sort_ratio
Sorts tokens alphabetically before comparison — handles word-order variations and middle name reordering
EXAMPLE · FUZZY HIT DETECTED
Query: "Ali Hasan Al-Farsi"
Matched: "Ali Hassan Al-Farsi"
Score: 96 / 100 → OFAC-SDN · Terrorism Financing
EXAMPLE · PEP MATCHED
Query: "Priya Nanda Kumar"
Matched: "Priya Nandakumar"
Score: 92 / 100 → PEP Tier 2 · State Chief Secretary
THRESHOLD LOGIC
≥ 90 → HIGH alert · Immediate escalation
75–89 → MEDIUM · Analyst review required
60–74 → REVIEW · Enhanced due diligence
< 60 → CLEAR · No match detected
Alert Queue · Screening Results 39 alerts
Filter:
Cust. IDCustomer NameAlert Matched NameScoreType Watchlist / TierWatchlist IDReason / RoleMatch Country