Lab · Knowledge graphs

Who's who in research?
From messy records to a graph you can question.

The same researcher appears on papers as “Priya Raman”, “P. Raman” and “Raman, Priya”; the same university as “Northbridge University”, “Univ. of Northbridge” and “NBU”. Until you resolve who's who, you can't answer simple questions — who leads a field, or which institutions work together. Move the sliders, then ask the graph.

  1. 01

    Messy records

    221 author mentions across 77 papers, written 85 different ways.

  2. 02

    Who's who

    Names, institutions, ORCIDs and shared co-authors decide who is the same person.

  3. 03

    The graph

    Researchers, papers, institutions and topics, connected.

  4. 04

    Ask the graph

    Questions answered from the connections — with the query behind each answer.

  5. 05

    AI on the graph

    Claude turns plain-English questions into graph queries, then answers only from the results.

221

Author mentions

24

People found (24 real)

100%

Precision: no wrong merges

100%

Recall: no missed matches

How much each signal counts

45
25
20
10

Decision thresholds

75%
60%

Try this

Lower “same person” to 65% — two different J. Chens, at different institutions, become one.

Set Institution to 0 — the same mistake, because nothing else tells them apart.

Raise it to 95% — real people split into duplicates.

An ORCID on both mentions always decides, either way.

Step 01

Messy records

Every author mention, as printed on the paper. Search for a surname to see its variants.

PaperName as printedAffiliation as printedResolved institutionORCID
p001 · 2019Jun ChenNorthbridge UniversityNorthbridge University—
p001 · 2019Priya RamanNorthbridge UniversityNorthbridge University0000-0001-0001-0001
p002 · 2019Jun ChenNorthbridge UniversityNorthbridge University—
p002 · 2019S. O'BrienAurora Institute of FinanceAurora Institute of Finance—
p002 · 2019N. HaddadNorthbridge UniversityNorthbridge University—
p003 · 2019Kenji WatanabeMeridian Pay LabMeridian Payments Lab—
p003 · 2019O. BrandtRhine Applied AI LabRhine Applied AI Lab0000-0001-0005-0005
p003 · 2019S. RomanoMeridian Pay LabMeridian Payments Lab—
p004 · 2019H. WeberRhine Applied AI LabRhine Applied AI Lab—
p004 · 2019Brandt, OliverRhine Applied AI LabRhine Applied AI Lab—
p004 · 2019Lars LindqvistNordic Centre for Trustworthy AINordic Centre for Trustworthy AI—
p005 · 2019Haddad, NadiaNorthbridge Univ.Northbridge University—

Step 02

Who's who

Mentions are compared when their family names look alike. Names, institutions, shared co-authors and topic add up to a score; your thresholds decide. Below: the closest calls, then everyone found.

  • O. Brandt (Rhine Applied AI Lab, 2021)Oliver Brandt (Meridian Payments Lab, 2024)

    69%Steward review
    Name
    94%
    Institution
    0%
    Shared co-authors
    100%
    Same topic
    —

  • Brandt, Oliver (Rhine Applied AI Lab, 2022)Brandt, Oliver (Meridian Payments Lab, 2023)

    69%Steward review
    Name
    100%
    Institution
    0%
    Shared co-authors
    —
    Same topic
    100%

  • Brandt, Oliver (Rhine Applied AI Lab, 2022)Oliver Brandt (Meridian Payments Lab, 2024)

    69%Steward review
    Name
    100%
    Institution
    0%
    Shared co-authors
    —
    Same topic
    100%

  • Adeyemi, Grace (Aurora Institute of Finance, 2023)Grace Adeyemi (Thames Institute of Technology, 2024)

    69%Steward review
    Name
    100%
    Institution
    0%
    Shared co-authors
    —
    Same topic
    100%

  • Adeyemi, Grace (Aurora Institute of Finance, 2023)Grace Adeyemi (Thames Institute of Technology, 2025)

    69%Steward review
    Name
    100%
    Institution
    0%
    Shared co-authors
    —
    Same topic
    100%

  • J. Chen (Northbridge University, 2025)J. Chen (Calder Bank Research, 2026)

    68%Steward review
    Name
    94%
    Institution
    0%
    Shared co-authors
    80%
    Same topic
    100%

  • Aisha Okafor

    Thames Institute of Technology (2023)

    8 mentions · 8 papers

    A. OkaforAisha OkaforOkafor, A.

  • Daniel Okafor

    Calder Bank Research (2021)

    5 mentions · 5 papers

    Daniel OkaforD. Okafor

  • Emma Clarke

    Calder Bank Research (2021)

    5 mentions · 5 papers

    E. ClarkeClarke, EmmaEmma Clarke

  • Erik Johansson

    Nordic Centre for Trustworthy AI (2019)

    9 mentions · 9 papers

    Erik JohanssonJohansson, ErikE. Johansson

  • Fatima Malik

    Thames Institute of Technology (2021)

    6 mentions · 6 papers

    Malik, FatimaFatima MalikF MalikF. MalikMalik, F.

  • Grace Adeyemi

    Aurora Institute of Finance (2020) → Thames Institute of Technology (2024)

    9 mentions · 9 papers · ORCID 0000-0001-0016-0016

    Grace AdeyemiAdeyemi, G.Adeyemi, Grace

  • Hannah Weber

    Rhine Applied AI Lab (2019)

    8 mentions · 8 papers

    H. WeberHannah WeberWeber, H.Weber, Hannah

  • Ingrid Nilsson

    Nordic Centre for Trustworthy AI (2022)

    8 mentions · 8 papers

    Ingrid NilssonNilsson, IngridI. Nilsson

  • Jia Chen

    Calder Bank Research (2021)

    3 mentions · 3 papers

    Jia ChenJ ChenJ. Chen

  • Jun Chen

    Northbridge University (2019)

    16 mentions · 16 papers

    Jun ChenJ ChenChen, J.J. Chen

  • Kenji Watanabe

    Meridian Payments Lab (2019)

    10 mentions · 10 papers

    Kenji WatanabeK. WatanabeWatanabe, Kenji

  • Lars Lindqvist

    Nordic Centre for Trustworthy AI (2019)

    12 mentions · 12 papers · ORCID 0000-0001-0007-0007

    Lars LindqvistLindqvist, LarsLindqvist, L.L. LindqvistL Lindqvist

  • Lukas Fischer

    Rhine Applied AI Lab (2020)

    6 mentions · 6 papers

    Fischer, L.Fischer, LukasLukas FischerL. Fischer

  • Maria Santos

    Lisbon School of Data Science (2020)

    9 mentions · 9 papers · ORCID 0000-0001-0004-0004

    Santos, MariaM. SantosM SantosMaria Santos

  • Mei Lin

    Northbridge University (2019)

    12 mentions · 12 papers · ORCID 0000-0001-0012-0012

    M. LinLin, M.Lin, MeiMei LinM Lin

  • Nadia Haddad

    Northbridge University (2019)

    13 mentions · 13 papers

    N. HaddadHaddad, NadiaNadia HaddadN HaddadHaddad, N.

  • Oliver Brandt

    Rhine Applied AI Lab (2019) → Meridian Payments Lab (2023)

    16 mentions · 16 papers · ORCID 0000-0001-0005-0005

    O. BrandtBrandt, OliverOliver BrandtBrandt, O.

  • Pieter de Vries

    Meridian Payments Lab (2019)

    9 mentions · 9 papers · ORCID 0000-0001-0021-0021

    P. de Vriesde Vries, PieterPieter de Vries

  • Priya Raman

    Northbridge University (2019)

    11 mentions · 11 papers · ORCID 0000-0001-0001-0001

    Priya RamanRaman, PriyaP. RamanP Raman

  • Ravi Iyer

    Thames Institute of Technology (2022)

    4 mentions · 4 papers

    Ravi IyerIyer, RaviR Iyer

  • Sean O'Brien

    Aurora Institute of Finance (2019)

    14 mentions · 14 papers

    S. O'BrienO'Brien, SeanSean O'BrienSean OBrien

  • Siobhan Murphy

    Aurora Institute of Finance (2021)

    11 mentions · 11 papers

    S. MurphySiobhan MurphyMurphy, Siobhan

  • Sofia Romano

    Meridian Payments Lab (2019)

    10 mentions · 10 papers

    S. RomanoSofia RomanoRomano, Sofia

  • Tomás Oliveira

    Lisbon School of Data Science (2019)

    7 mentions · 7 papers

    Tomas OliveiraTomás OliveiraT. OliveiraT Oliveira

Step 03

The graph

Each circle is a person, coloured by institution and sized by papers; each line is co-authorship, thicker for more joint papers. Click a person to explore their work.

  • Northbridge University
  • Thames Institute of Technology
  • Calder Bank Research
  • Lisbon School of Data Science
  • Rhine Applied AI Lab
  • Meridian Payments Lab
  • Aurora Institute of Finance
  • Nordic Centre for Trustworthy AI

Step 04

Ask the graph

Each answer is computed live from the graph you just built — so a wrong merge upstream changes the answer here. The Cypher shows how a graph database such as Neo4j would ask the same question.

Answer

  1. Priya RamanNorthbridge University · 5 papers14
  2. Jun ChenNorthbridge University · 4 papers10
  3. Mei LinNorthbridge University · 3 papers5
  4. Pieter de VriesMeridian Payments Lab · 2 papers5
  5. Nadia HaddadNorthbridge University · 2 papers3

The Cypher behind it

MATCH (a:Author)-[:WROTE]->(p:Paper)-[:ABOUT]->(:Topic {id: 'gnn'})
OPTIONAL MATCH (p)<-[:CITES]-(c:Paper)
RETURN a.name, count(DISTINCT p) AS papers, count(c) AS citations
ORDER BY citations DESC, papers DESC LIMIT 5

Step 05

AI on the graph

GraphRAG: Claude turns a plain-English question into a read-only Cypher query, a guard checks it, Neo4j runs it, and Claude answers only from the rows that come back. These are real answers, recorded from one run against an evaluation set — so this page needs no API keys.

15/15

Answers with the expected facts

15/15

Queries that ran first time or after one self-correction

3/3

Out-of-scope or write requests refused

Claude’s answer

Correct

Based on the query results, Priya Raman is the most cited researcher on graph neural networks for fraud, with 11 citations to their papers on this topic.

The Cypher Claude wrote

MATCH (a:Author)-[:WROTE]->(p:Paper)-[:ABOUT]->(:Topic {id: 'gnn'}) OPTIONAL MATCH (p)<-[:CITES]-(c:Paper) RETURN a.name AS author, count(DISTINCT c) AS citations ORDER BY citations DESC LIMIT 1

Recorded 2026-10-07 · Anthropic claude-haiku-4-5-20251001 on Amazon Bedrock · Neo4j AuraDB · temperature 0

Under the hood

The same entity resolution in Python, the graph in Neo4j, GraphRAG with Claude on Amazon Bedrock and its evaluation — plus an MCP server, so Claude Desktop or Claude Code can explore the graph as read-only tools.

How I built it →Code on GitHub ↗

A fictional dataset on AI in payments and fraud detection: 77 papers and 221 author mentions, all invented. It shows the concepts — entity resolution, knowledge graphs and graph queries — not any employer's system.

Contact

Let's talk.

I'm based in London, UK. Whether it's a leadership role, an advisory engagement or a transformation that needs shaping — my inbox is open.

srini.vankee@gmail.com
LinkedIn
↑