Research Graph

Analysts use the graph to discover emerging protocols and perform deep due diligence by mapping relationships and activity flows. Key analyses include:

• Entity Resolution: Identifying wallets controlled by the same entity (e.g., a project treasury, team members, or VCs).
• Flow-of-Funds Analysis: Tracing capital inflows/outflows to assess protocol health and investor sentiment.
• Smart Contract Interaction Mapping: Understanding how a protocol's contracts interact with others in the DeFi ecosystem.

The graph creates detailed profiles for wallets by clustering addresses and analyzing historical behavior patterns. This enables:

• Clustering & Labeling: Grouping addresses belonging to the same user (e.g., an EOA and its associated smart contract wallets).
• Behavioral Archetypes: Classifying wallets as whales, market makers, arbitrage bots, or retail traders based on transaction patterns.
• Anomaly Detection: Spotting unusual activity, such as sudden large withdrawals or interactions with high-risk contracts.

Developers and quantitative analysts leverage the historical state of the graph to test trading and liquidity provision strategies. This involves:

• Historical State Queries: Reconstructing the exact state of liquidity pools, oracle prices, or lending positions at any past block.
• Strategy Simulation: Running 'what-if' scenarios to evaluate the performance of a strategy against historical market conditions.
• Slippage & MEV Analysis: Modeling transaction execution to estimate real-world costs and frontrunning risks.

Institutions and protocols use the graph for real-time compliance checks and systemic risk assessment. Core functions include:

• Sanctions Screening: Screening transaction counterparties against known sanctioned addresses or high-risk jurisdictions.
• Exposure Analysis: Calculating a protocol's total exposure to a specific token, counterparty, or vulnerable smart contract.
• Governance Power Mapping: Visualizing voting power concentration and delegation patterns within DAOs to assess centralization risks.

Researchers identify macro trends and emerging narratives by analyzing aggregate activity across the graph. This process includes:

• Topic Modeling: Using Natural Language Processing (NLP) on transaction calldata and event logs to categorize activity.
• Cross-Protocol Correlation: Discovering relationships between activity in different sectors (e.g., NFT minting and DeFi borrowing).
• Early Signal Detection: Identifying nascent trends, such as capital rotation into a new L2 or the early adoption of a novel token standard.

Data providers and infrastructure teams build enriched data products using the graph as a foundational layer. Common outputs are:

• Enriched APIs: Offering endpoints that return not just raw transactions, but contextualized data (e.g., 'transactions for Entity X').
• Pre-computed Metrics: Generating and serving derived metrics like Total Value Secured (TVS), wallet profitability, or protocol market share.
• Real-Time Alerting Feeds: Creating customizable data streams that trigger alerts based on on-chain graph events.

The Research Graph protocol is built on a decentralized network of nodes that index and link data. It uses a graph database structure where entities (like papers, datasets, authors, and grants) are nodes, and their relationships (citations, funding, authorship) are edges. This creates a machine-readable, interconnected web of knowledge that is verifiable and persistent on the blockchain.

At its core, the Research Graph employs a semantic data model (often using schemas like Schema.org or custom RDF ontologies) to standardize how research objects are described. This allows disparate data sources—from arXiv preprints and PubMed to on-chain grant data—to be interoperably linked, enabling complex queries across the entire research lifecycle.

A key component is the use of Decentralized Identifiers (DIDs) to provide persistent, verifiable identities for all entities in the graph. This means:

• A researcher has a DID, not just an institutional email.
• A research paper or dataset is minted as a non-fungible token (NFT) or linked to a Content Identifier (CID), anchored by its DID.
• This creates a trustless attribution system resistant to link rot and centralized platform failure.

The network is sustained by a native utility token that incentivizes participation. Curators earn tokens for adding high-quality data and verifying links. Indexers operate nodes to serve graph queries. Consumers (like apps or researchers) may spend tokens to access premium query services or datasets, creating a circular economy for open knowledge.

The structured data enables powerful applications:

• Discovery Engines: Find related work across publications, code, and data.
• Reputation Systems: Track a researcher's contributions and impact via on-chain verifiable credentials.
• Funding Transparency: Map the flow of grants from funders to institutions to published outcomes.
• Interdisciplinary Research: Uncover hidden connections between fields by traversing the graph.

The Research Graph intersects with several key Web3 and data concepts:

• The Graph Protocol: While The Graph indexes blockchain state, the Research Graph indexes scholarly knowledge.
• Decentralized Science (DeSci): The graph is foundational infrastructure for the DeSci movement.
• Verifiable Credentials (VCs): Used to attest to authorship, peer review, or educational background.
• IPFS & Arweave: Often used as the persistent storage layer for the research objects referenced in the graph.

On-chain analytics is the practice of extracting, analyzing, and visualizing data from public blockchain ledgers. A Research Graph operationalizes this by structuring raw transaction data into queryable relationships. This enables analysts to track fund flows, measure protocol adoption, identify wallet clusters, and detect patterns like money laundering or voting cartels. Tools like Nansen and Dune Analytics build upon similar graph-like data models.

Entity resolution is the process of determining when multiple records or data points refer to the same real-world entity. In a Research Graph, this is critical for clustering related wallet addresses under a single identity (e.g., a DAO treasury, a CEX hot wallet, or a user's multi-sig). Techniques involve analyzing transaction patterns, common funding sources, and interaction graphs to de-anonymize and label blockchain actors accurately.

Graph traversal is an algorithm for visiting nodes in a graph by following edges. In blockchain analysis, it is used to trace the flow of assets or influence through a network. A Research Graph enables powerful traversals, such as:

• Pathfinding: Discovering the shortest transaction path between two wallets.
• Community Detection: Identifying tightly-knit groups of addresses (e.g., arbitrage bots).
• Influence Analysis: Mapping governance power distribution via delegate relationships.