Efficient Densest Flow Queries in Transaction Flow Networks (Complete Version)

TL;DR

An approximate flow-peeling algorithm is introduced to optimize the performance of CONAN, enhancing its efficiency in processing large transaction networks and showcasing CONAN's applications in fraud detection on transaction flow networks from its industry partner, Grab, and on non-fungible tokens (NFTs).

Abstract

Transaction flow networks are crucial in detecting illicit activities such as wash trading, credit card fraud, cashback arbitrage fraud, and money laundering. \revise{Our collaborator, Grab, a leader in digital payments in Southeast Asia, faces increasingly sophisticated fraud patterns in its transaction flow networks. In industry settings such as Grab's fraud detection pipeline, identifying fraudulent activities heavily relies on detecting dense flows within transaction networks. Motivated by this practical foundation,} we propose the \emph{\(S\)-\(T\) densest flow} (\SDMF{}) query. Given a transaction flow network \( G \), a source set \( \Src \), a sink set \( \Dst \), and a size threshold \( k \), the query outputs subsets \( \Src' \subseteq \Src \) and \( \Dst' \subseteq \Dst \) such that the maximum flow from \( \Src' \) to \( \Dst' \) is densest, with \(|\Src' \cup \Dst'| \geq k\). Recognizing the NP-hardness of the \SDMF{} query, we develop an efficient divide-and-conquer algorithm, CONAN. \revise{Driven by industry needs for scalable and efficient solutions}, we introduce an approximate flow-peeling algorithm to optimize the performance of CONAN, enhancing its efficiency in processing large transaction networks. \revise{Our approach has been integrated into Grab's fraud detection scenario, resulting in significant improvements in identifying fraudulent activities.} Experiments show that CONAN outperforms baseline methods by up to three orders of magnitude in runtime and more effectively identifies the densest flows. We showcase CONAN's applications in fraud detection on transaction flow networks from our industry partner, Grab, and on non-fungible tokens (NFTs).

Figures (22)

• Figure 1: Densest Flow Query on Transaction Flow Networks and Example Applications in Grab
• Figure 2: A transaction flow network
• Figure 3: Difference between Maximum Flow and Densest Flow
• Figure 4: A two-stage solution overview of $\mathsf{Conan}$
• Figure 5: (a)-(c) classic algorithms cannot return maximum temporal flow, and (d) Residual network of maximum temporal flow.

Theorems & Definitions (53)

• Definition 2.1: Flow
• Definition 2.2: $\mathsf{TFN}$
• Example 2.1
• Lemma 2.1
• proof
• Definition 4.1: Regret-disabling vertex ($\mathsf{RDV}$)
• Example 4.1
• Example 4.2
• Theorem 4.1
• Example 4.3