In this poster, we will present attendees how the recent stateof-the-art Semantic Web tool SANSA could be used to tackle blockchain speci c challenges. In particular, the poster will focus on the use case of CryptoKitties: a popular Ethereum-based online game where users are able to trade virtual kitty pets in a secure way. During the recent years, the interest of the research community for blockchain technologies has raised up since the Bitcoin white paper published by Nakamoto [4]. Indeed, several applications either practical or theoretical have been made, for instance with crypto-currencies or with secured exchanges of data. More speci cally, several blockchain systems have been designed, each one having their own set of properties, for example Ethereum [6] {further described in Section2{ which provides access to so-called smart-contracts. Recently, Mukhopadhyay et al. surveyed major crypto-currencies in [3]. In parallel to ongoing blockchain development, the research community has focused on the possibilities o ered by the Semantic Web such as designing e cient data management systems, dealing with large and distributed knowledge RDF graphs. As a consequence, recently, some studies have started to connect the Semantic Web world with the blockchain one, see for example the study of English et al. presented in [1]. In this poster, we will o er attendees the possibility of discovering the blockchain world, with the Semantic Web framework SANSA applied in the CryptoKitties use case.
smart contracts and deploy them on the Ethereum network, to create
applications. Moreover, anyone can also interact with them; this is done by calling
contract functions, which in turn trigger events logged to a public record. Costly
interactions requiring computational resources are provided with a fee given to
the Ethereum block miners. Via smart contracts, applications are supposed to
run exactly as de ned without any possibility of downtime, censorship, fraud or
any third-party interferences [
Alethio2 is a spin-o working with Ethereum technologies. It is an advanced
analytics platform making Ethereum more accessible and digestible for everyone.
It could be seen as a \blockchain archeology" of the Ethereum main network,
providing not only all the original transaction data and log messages, but also
analytic results with speci c metrics. Their extensive data set (currently
encompassing 36 billion rows of records) contains large-scale blockchain transaction
data modelled in RDF according to the structure of the Ethereum ontology.
EthOn (The Ethereum Ontology) [
The SANSA stack [
SemanticWeb & Ethereum: the CryptoKitties Use Case The Ethereum ecosystem generates a large amount of data, including but not limited to protocol-level data (e.g. average block time, gas prices), as well as application-level data (e.g. account interactions, smart contract deployments). To e ciently handle this volume of data, Alethio has investigated di erent tools and frameworks with one focus: the infrastructure should be resilient, loadbearing, and most importantly, scalable. And so, for that reason to overcome the variety of the di erent data sources, Alethio introduces semanti cation of Ethereum network and uses SANSA as an underlying engine for large scale distributed RDF based querying, reasoning, and machine learning on top of these RDF datasets. To show the joint e ort between SANSA and Alethio, we describe a use case on how SANSA can be used to analyze Ethereum at new scales. 1 https://www.coindesk.com/understanding-dao-hack-journalists/ 2 https://aleth.io/ (last accessed June 11th 2018) 3 https://github.com/SANSA-Stack
(a) The unique attributes of a Kitty. (b) An instance of a Kitty.
(c) History of three types of auction events. (d) The process pipeline.
(e) An illustration of a small family tree.
CryptoKitties4 is one of the rst games to be built on blockchain technology. In particular, CryptoKitties initiated and released the rst generation virtual kitties, with delicately designed icons and genes sequences. All the kitties are virtual with some biological feature settings. Shown in Figure 1b is a kitty with its speci c biological attributes displayed in Figure 1a. The attributes are stored in a sequence, succeeded from its parents' gene sequences, with possibility of mewtations. An owner can sell, breed or gift it to other user. When users sell or breed it, they will send transactions to the CryptoKitties smart contracts, which will complete the execution of either transferring ownership between users, or generating a new kitty. Based on that, game users can trade or breed kitties like traditional collectibles, while having the guarantee that the blockchain will track ownership securely. Moreover, one can breed two kitties to create a brand-new, genetically unique o spring. 4 https://www.cryptokitties.co/ (last accessed June 11th 2018) Data Challenges. Alethio has been exploring e cient means of processing large RDF data sets. SANSA empowers Alethio to read and query the data at scale as described in Figure 1d. Indeed, once the complete RDF data set is loaded, SANSA lters it to retain only the CryptoKitties triples {transactions, contract messages and log information{ before performing more speci c analyses.
Practically, the challenges tackled with SANSA can be divided into two groups: game performance and customer behaviors. The rst one focuses on time series metrics: throughout time, the event volume, number of active users and amount of spent Ether, which can jointly estimate the trend of popularity for the game. In Figure 1c, the history of CryptoKitties auctions events shows clearly that there was a peak of tra c in December after the game was launched for around one month. By this time series, we can estimate the popularity of the game throughout history. The second one requires machine learning algorithms to detect correlations between indicators (e.g. to determine whether richer owners have the tendency to collect special/rare kitties which are more expensive). and topology from a network view. In Figure 1e, we present a small subset of the kitty family tree, where incest happened during the reproduction: kitty 1057 is the secondary-degree relative (grandparent) of kitty 3200, while later it bred with kitty 3200 and gave birth to kitty 3225. 4
During this poster session, attendees will be able to discover the blockchain world and the speci cities of the Ethereum through the CryptoKitties use case. In addition, we will present how the SANSA stack {and the Semantic Web standards{ has been used to tackle some speci c problems dealing with blockchain analyses.