{"paper_id":"2507.01980","title":"Detecting Fraud in Financial Networks: A Semi-Supervised GNN Approach with Granger-Causal Explanations","authors":["Linh Nguyen","Marcel Boersma","Erman Acar"],"paper_date":"2025-06-25","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.01980v1","page_url":"https://thequant.space/flowcharts/detecting-fraud-in-financial-networks-a-semi-supervised-gnn/","scores":{"empirical_rigor":7,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes SAGE-FIN, a semi-supervised graph neural network with Granger-causal explanations for fraud detection in financial interaction networks.","quote":"This article proposes SAGE-FIN, a semi-supervised graph neural network (GNN) based approach with Granger causal explanations for Financial Interaction Networks.","summary":"Fraudulent activity in the financial industry costs billions annually. Detecting fraud, therefore, is an essential yet technically challenging task that requires carefully analyzing large volumes of data. While machine learning (ML) approaches seem like a viable solution, applying them successfully…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Labels of financial transactions are expensive to obtain for model training.","type":"data","stated":"explicit","quote":"non-fraud) are often expensive to obtain for model training"},{"assumption":"The model can be trained on weakly labeled or unlabeled data points.","type":"data","stated":"explicit","quote":"SAGE-FIN learns to flag fraudulent items based on weakly labeled (or unlabelled) data points"},{"assumption":"No prior assumption on the network structure is required.","type":"model","stated":"explicit","quote":"without any prior assumption on the network structure"},{"assumption":"The financial transaction data can be represented as a bipartite graph of wallets and transactions.","type":"data","stated":"implied","quote":"The dataset is presented in a bipartite format, where monetary flows from wallets to specific transactions are presented as edges"},{"assumption":"The Elliptic++ dataset contains no missing values.","type":"data","stated":"explicit","quote":"Elliptic++ dataset is set up with high quality, ensuring no missing values within its comprehensive set of features"},{"assumption":"Granger causality provides a valid causal explanation for flagged fraud items.","type":"statistical","stated":"implied","quote":"transparent causal insights without relying on a specific model structure"}]},"datasets":{"status":"available","items":[{"name":"Elliptic++","provider":null,"kind":"crypto-onchain","access":"public","period":null,"universe":"Bitcoin transaction and wallet address network","quote":"The real-world Elliptic++ Bitcoin data is utilized. The dataset is presented in a bipartite format, where monetary flows from wallets to specific transactions are presented as edges"}],"known":[{"id":"kaggle","name":"Kaggle","provider":"Kaggle","kind":"other","access":"public","quote":"The labels of the fraudulent/non-fraudulent for nodes are provided by the author [15] based on the original Elliptic dataset published on Kaggle.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"SAGE-FIN significantly outperformed the Logistic Regression model but did not surpass Random Forest and XGBoost, which had higher precision.","kind":"performance","quote":"Across several model choices, the model significantly outperformed the Logistic Regression model but did not surpass RF and XGB, primarily due to the latter model’s noticeably higher precision"},{"finding":"The SAGE-FIN model demonstrated commendable precision, recall, and F1‑score metrics for anomaly detection on the Elliptic++ dataset.","kind":"performance","quote":"The SAGE-FIN model has demonstrated commendable performance in terms of precision, recall, and F1-score metrics when applied…"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":22,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:34:39Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}