=Paper=
{{Paper
|id=Vol-2451/paper-25
|storemode=property
|title=The Hubs and Authorities Transaction Network Analysis using the SANSA framework
|pdfUrl=https://ceur-ws.org/Vol-2451/paper-25.pdf
|volume=Vol-2451
|authors=Danning Sui,Gezim Sejdiu,Damien Graux,Jens Lehmann
|dblpUrl=https://dblp.org/rec/conf/i-semantics/SuiSG019
}}
==The Hubs and Authorities Transaction Network Analysis using the SANSA framework==
https://ceur-ws.org/Vol-2451/paper-25.pdf
The Hubs and Authorities Transaction Network
Analysis using the SANSA framework
Danning Sui1 , Gezim Sejdiu2 , Damien Graux3 , and Jens Lehmann2,3
1
Alethio
danning.sui@consensys.net
2
Smart Data Analytics, University of Bonn, Germany
sejdiu@cs.uni-bonn.de, jens.lehmann@cs.uni-bonn.de
3
Fraunhofer Institute for Intelligent Analysis and Information Systems, Germany
damien.graux@iais.fraunhofer.de, jens.lehmann@iais.fraunhofer.de
Abstract. With the recent trend on blockchain, many users want to
know more about the important players of the chain. In this study, we
investigate and analyze the Ethereum blockchain network in order to
identify the major entities across the transaction network. By leveraging
the rich data available through Alethio’s platform in the form of RDF
triples we learn about the Hubs and Authorities of the Ethereum trans-
action network. Alethio uses SANSA for efficient reading and processing
of such large-scale RDF data (transactions on Ethereum blockchain) in
order to perform analytics e.g. finding top accounts, or typical behavior
patterns of exchanges’ deposit wallets and more.
1 Introduction
With the hype on blockchain technologies and in particular in the Ethereum
blockchain [4], many participants wanted to know more about the most impactful
players across the blockchains transaction network. In parallel, as the number
of statements, actions and transactions in the network are increasing quickly,
many “Big Data” challenges arise. First, transactions are raw data and one
cannot take advantage of them for further analysis. To do so, Alethio designed
EthOn (The Ethereum Ontology) [3] which models such raw data as triples using
the RDF (Resource Description Framework)1 standard. This ontology describes
all Ethereum terms including blocks, transactions, contract messages, event logs
etc., as well as their relationships. Afterword, performing querying and analysis
on such large-scale RDF datasets is computing intensive. To overcome these
challenges, we have explored the potential of the SANSA [2] framework. SANSA
is an open-source2 framework for distributed processing and analysis of large-
scale RDF data. With SANSA on Spark, RDF triples are loaded into Spark
distributed and resilient data structured, namely the data frames, for further
analysis.
1
https://www.w3.org/TR/rdf-primer/
2
https://github.com/SANSA-Stack
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License
Attribution 4.0 International (CC BY 4.0).
SANSA Engine
Top Accounts, Hubs & Authorities, Wallet
Data visualization using the Exchange behavior
Databricks notebooks or SANSA
notebooks
Querying Connected
Connected Components PageRank
(SPARQL) Components
EthOn RDF
triples Data partition
Hubs &
Authorities
Amazon S3
entities
buckets Data ingestion
Fig. 1. Hubs and Authorities analysis workflow.
In this paper, we perform an analysis (using well-known graph processing
algorithms) of the value transaction network graph with the main focus on the
Hubs and Authorities behaviors. “Authorities” are accounts who pay out to a
large crowd of addresses, with high volume; while “Hubs” are entities who receive
extensive Ether (ETH) flow into their accounts. In this study, we do not differ-
entiate these two roles but rank them all together as the biggest players/entities.
2 Finding big Ethereum players with SANSA
The Ethereum network graph contains nodes of external accounts which have
had a transaction on the Ethereum blockchain. The connection (edges) between
such nodes on the network indicate the transaction relationship between them;
when a node (an external account) sends ETH to another, a transaction record is
written, and an edge between them is added in the network with the direction of
the ETH flow. When we encounter multiple edges between same pairs of nodes,
we summarize the edges as a single one3 . The edge weight is the total transaction
value in Ether. As an example, if address A sends x ETH to address B in total,
there will be an edge of weight x from node A to node B. In this study, self-loops
i.e. transactions from an address to itself are omitted.
SANSA framework has been used for efficient reading and querying of RDF
datasets using SPARQL as depicted on Figure 1. First, the data need to be
loaded on an efficient storage that SANSA can read from. For that purpose, we
3
This optimization is also convenient practically as it is easier not to have duplicated
edges in a graph.
Fig. 2. PageRank Score Distribution of Fig. 3. Category Distribution of Top-50
Top-50 Accounts. Accounts.
use Amazon S3 buckets containing the whole RDF Ethereum network transac-
tions. Afterword, SANSA data representation layer loads the data in a form of
Resilient Distributed Datasets (RDD) [5] of triples. During this process, SANSA
performs a data partition for fast processing and then aggregate and filter the
data using the its query layer [1]. Further, we applied two classic graph analysis
algorithms via Apache GraphX: Connected Components and Page Rank. Con-
nected Components algorithm enables us to find the largest cluster of connected
nodes, regardless of transaction direction. Within this largest cluster, we can
derive the page rank score of all nodes. Top-ranked entities and their relation
are visualized.
3 Results
3.1 Datasets
The Ethereum dataset in the format of RDF contains more than 17B triples. For
the sake of the experiment, we limited the dataset to 10,000 blocks which contain
around 38M triples, including both value transactions and contract messages.
3.2 Top Accounts Analysis
The PageRank algorithm was run over the largest connected component of
185,741 nodes (accounts) and 250,637 edges (aggregated transaction relations).
Figure 2 plots the top 50 account’s distribution. Based on the findings, we can
see that these accounts are grouped on two different types: mining pool wallets,
and (mostly centralized) exchange wallets.
Figure 3 shows that 58% of the addresses are controlled by exchanges, while
another 12% with convincing tags related to the mining pools. The exchange
and mining pool wallets can be found in the top position of our ranking, under-
lining the effectiveness of PageRank: Addresses related to mining pools allocate
extensive amounts of payouts to their subscribed miners, resulting in large out-
degrees, as well as high accumulated transaction value. We can see that the main
Fig. 4. Transaction Network of Top Hubs and Authorities.
wallets are centralized exchanges which distribute (and receive) large volumes
of the transaction to (and from) their deposit wallets, token contracts, etc.
Our PageRank implementation successfully detects the most influential ac-
counts across the network, corresponding to the Hubs and Authorities, connect-
ing various transactors and carrying heavy flow weights.
Focusing on those known accounts (with labels from Etherscan4 ), we present
(see Figure 4) the network overview of top hubs and authorities with transactions
as edges surrounding them.
3.3 Typical Behavior Patterns of Exchanges’ Deposit Wallets
We investigated the associated transaction behavior of the exchange wallets.
Based on our finding, these behaviors can be grouped into three categories:
1. Frequently paying out to certain exchanges’ main wallets with a fixed, large
value – From the scatter plot, the payout amount is always around a same
value.
2. Frequently receiving funds from the same exchange main wallets, and paying
out to various token contracts – This is due to the activity which is associ-
4
https://etherscan.io/
ated with exchanges as they use external accounts as deposit addresses for
collecting tokens based on trading needs.
3. Frequently receiving funds from a group of “miner” accounts, with “proxy”
accounts in between, which clean out their received ETH within a short time
frame – Usually, these addresses receive funds from miner accounts, which
again get paid reasonable amounts by known mining pools, which we assume
are mining rewards (usually around 0.11-0.12 ETH).
Despite pointing out the three typical behaviors above, they are not nec-
essarily mutually exclusive. There are addresses which share more than one of
the deducted patterns. These behavior patterns explored here are based on the
labels we have gathered, and this may be different for other use cases.
4 Conclusion
SANSA provides a scalable solution for reading and querying large scale RDF
data, providing compatibility with machine learning libraries on Spark including
GraphX as a graph processing library. With conventional graph analysis tools,
we successfully identified Hubs and Authorities in the Ethereum transaction
network and discovered that they are mainly related to exchange wallet and
mining pool activities.
This pipeline also provides a possibility to filter out top accounts, which are
likely to be exchanges’ deposit wallets. Furthermore, with the filtered top rank
accounts, the “mixing” patterns of exchanges’ deposit wallets become recogniz-
able. This can be a promising tool for detecting previously unknown exchange
wallets and lead to a deeper understanding of their behavior patterns for future
analyses.
References
1. Ermilov, I., Lehmann, J., Sejdiu, G., Bühmann, L., Westphal, P., Stadler, C., Bin,
S., Chakraborty, N., Petzka, H., Saleem, M., Ngonga, A.C.N., Jabeen, H.: The Tale
of Sansa Spark. In: 16th International Semantic Web Conference, Poster & Demos
(2017)
2. Lehmann, J., Sejdiu, G., Bühmann, L., Westphal, P., Stadler, C., Ermilov, I., Bin,
S., Chakraborty, N., Saleem, M., Ngonga Ngomo, A.C., Jabeen, H.: Distributed
semantic analytics using the SANSA stack. In: ISWC Resources Track (2017)
3. Pfeffer, J., Beregszazi, A., Detrio, C., Junge, H., Chow, J., Oancea, M., Pietrzak,
M., Khatchadourian, S., Bertolo, S.: Ethon - An Ethereum ontology (2016)
4. Wood, G.: Ethereum: A secure decentralised generalised transaction ledger.
Ethereum project yellow paper 151, 1–32 (2014)
5. Zaharia, M., Chowdhury, M., Das, T., Dave, A., Ma, J., McCauley, M., Franklin,
M.J., Shenker, S., Stoica, I.: Resilient distributed datasets: A fault-tolerant ab-
straction for in-memory cluster computing. In: Proceedings of the 9th USENIX
conference on Networked Systems Design and Implementation. pp. 2–2. USENIX
Association (2012)