Skip to content

ReconGraph: Probabilistic Consistency Inference over Heterogeneous Temporal Graphs for Enterprise Financial Reconciliation

Chandrakanth Thadkapally

Manuscript complete. Not submitted; no venue or submission date set.

§01
Abstract

Enterprise reconciliation is not a record-pair problem. A single economic event can produce an order, an invoice, a shipment, a receipt, a payment, a settlement, a commission, an adjustment and a ledger posting, written by different systems on different clocks. ReconGraph represents these artefacts as a heterogeneous temporal graph and infers a globally consistent reconciliation subgraph over it, rather than accepting or rejecting candidate pairs independently. Relation- and time-specific message passing combines attribute, amount, temporal, neighbourhood and business-rule evidence; a factorised consistency layer penalises violations of quantity conservation, temporal ordering, debit–credit balance and link cardinality. On RECONBENCH — a synthetic benchmark constructed from the public IBM AMLSim / AML-Data ecosystem — ReconGraph reaches 0.949 link F1, 0.944 graph-consistency accuracy and 0.892 anomaly recall, while reducing the false-reconciliation rate to 0.033, a 32.7% reduction against the strongest graph-matching baseline. Ablation, calibration, robustness and scaling studies isolate the contribution of the accounting constraints. The result is a reconciliation decision that is auditable as a subgraph with residuals attached, rather than an isolated similarity score.

Keywordsenterprise reconciliation · heterogeneous temporal graph · probabilistic inference · accounting constraints · entity matching · anomaly detection · calibration

§02
Contributions
  1. 01Poses reconciliation as inference of a globally consistent subgraph over a heterogeneous temporal graph, so that one-to-many, many-to-one and partial allocations are ordinary cases rather than exceptions to a bijection.
  2. 02Combines relation-aware and time-aware message passing over typed financial records with a learned local edge probability, so neighbourhood evidence can override a locally attractive but globally impossible link.
  3. 03Adds a factorised consistency layer that scores quantity-conservation, temporal-order, debit–credit balance and cardinality residuals, making the reason a subgraph was rejected inspectable per constraint family.
  4. 04Releases RECONBENCH, a synthetic order-to-cash and procure-to-pay benchmark derived from public AML simulation data, with ground-truth allocation edges and controlled corruption applied only after ground truth is frozen.
  5. 05Evaluates on link quality, false reconciliation, graph consistency, anomaly recall, calibration, runtime and review workload, rather than on pairwise accuracy alone.
§03
Method
  • Financial artefacts become typed nodes — invoices, shipments, receipts, payments, settlements, adjustments, ledger postings — and candidate correspondences become typed edges (fulfils, settles, allocates, reverses, posts-to), each carrying event time and ingestion time.
  • Candidate generation runs three cascaded lanes: exact normalised identifiers for precision, phonetic and edit-distance keys for recovery, and amount–time windows for the identifier-free case. Every retained edge records why it was blocked in, so candidate-generation misses can be audited separately from classifier errors.
  • A logistic local model scores each candidate edge from attribute similarity, normalised amount agreement, temporal evidence, graph context and business-rule features, with missing fields masked so absence is not read as agreement.
  • Two layers of relation- and time-specific attention propagate neighbourhood evidence without smoothing local reconciliation chains away.
  • A factorised consistency layer converts accounting structure into residuals over candidate subgraphs; posterior edge beliefs are temperature-calibrated so that a confidence threshold means something operationally.
§04
Results
All figures are from a controlled synthetic benchmark. No production or employer data appears anywhere in this work.
Table 01Reconciliation, consistency, anomaly and calibration performance
MethodPrec.RecallF1FRRGCAAnom.ECE
Exact rules.962.681.798.038.742.611.118
Fuzzy probabilistic.931.814.869.069.801.702.083
Neural pairwise.944.873.907.056.834.748.061
Graph matching.951.895.922.049.897.821.048
ReconGraph.967.931.949.033.944.892.026

FRR — false-reconciliation rate; GCA — graph-consistency accuracy; ECE — expected calibration error. Lower is better for FRR and ECE.

Table 02Performance by relationship topology, against the strongest baseline
TopologyMethodPrec.RecallF1GCA
One-to-oneGraph matching.962.929.945.928
One-to-oneReconGraph.973.951.962.966
One-to-manyGraph matching.941.872.905.881
One-to-manyReconGraph.963.929.946.936
Many-to-oneGraph matching.927.841.882.854
Many-to-oneReconGraph.951.887.918.910
Partial paymentGraph matching.938.858.896.867
Partial paymentReconGraph.958.914.936.925

The gap widens as the topology moves away from one-to-one — which is the population pairwise matching handles worst.

§05
What the numbers say
  • Pairwise systems improve F1 from rules to neural matching, but their graph-consistency accuracy stays below their F1 throughout: they get more links right without getting the explanation right. Global inference closes that gap.
  • Under 50% joint corruption of identifiers, fields, timestamps and amounts, F1 holds at 0.864, because weight moves off damaged identifiers and onto amount conservation, event ordering and neighbourhood structure.
  • Ablation separates the two mechanisms: removing graph context costs the most F1 (0.949 → 0.918), while removing accounting constraints costs the most consistency (GCA 0.944 → 0.872). They are not redundant.
  • Temperature calibration cuts ECE by 63.4%. At an auto-accept threshold of 0.92 the simulated auto-reconciliation rate is 71.6%, which is the number an operations team actually plans against.
  • Scaling from 0.1M to 5M records is near-linear: peak GPU memory 1.1 → 16.8 GB, inference 8 → 226 seconds, because blocking keeps connected components independent and average degree near-constant.
§06
Limitations
Stated by the author, not extracted by a reviewer.
  • RECONBENCH is synthetic. It is built to be reproducible and ethically shareable, not to be a substitute for a live enterprise trial, and no claim here transfers to production without one.
  • Candidate blocking bounds recall. A true allocation that never became a candidate cannot be recovered by any downstream stage, and the blocking-reason audit exists precisely because that failure is otherwise invisible.
  • The constraint set is the one this work formalised. Foreign exchange, tax treatment and late adjustment practice vary by jurisdiction and organisation, and an over-weighted constraint suppresses genuine exceptions rather than surfacing them.