Detection and Analysis of Ethereum Energy Smart Contracts
As blockchain technology advances, so has the deployment of smart contracts on blockchain platforms, making it exceedingly challenging for users to explicitly identify application services. Unlike traditional contracts, smart contracts are not written in a natural language, making it difficult to de...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/10/6027 |
| _version_ | 1797601268633436160 |
|---|---|
| author | Bahareh Lashkari Petr Musilek |
| author_facet | Bahareh Lashkari Petr Musilek |
| author_sort | Bahareh Lashkari |
| collection | DOAJ |
| description | As blockchain technology advances, so has the deployment of smart contracts on blockchain platforms, making it exceedingly challenging for users to explicitly identify application services. Unlike traditional contracts, smart contracts are not written in a natural language, making it difficult to determine their provenance. Automatic classification of smart contracts offers blockchain users keyword-based contract queries and a streamlined effective management of smart contracts. In addition, the advancement in smart contracts is accompanied by security challenges, which are generally caused by domain-specific security breaches in smart contract implementation. The development of secure and reliable smart contracts can be extremely challenging due to domain-specific vulnerabilities and constraints associated with various business logics. Accordingly, contract classification based on the application domain and the transaction context offers greater insight into the syntactic and semantic properties of that class. However, despite initial attempts at classifying Ethereum smart contracts, there has been no research on the identification of smart contracts deployed in transactive energy systems for energy exchange purposes. In this article, in response to the widely recognized prospects of blockchain-enabled smart contracts towards an economical and transparent energy sector, we propose a methodology for the detection and analysis of energy smart contracts. First, smart contracts are parsed by transforming code elements into vectors that encapsulate the semantic and syntactic characteristics of each term. This generates a corpus of annotated text as a balanced, representative collection of terms in energy contracts. The use of a domain corpus builder as an embedding layer to annotate energy smart contracts in conjunction with machine learning models results in a classification accuracy of 98.34%. Subsequently, a source code analysis scheme is applied to identified energy contracts to uncover patterns in code segment distribution, predominant adoption of certain functions, and recurring contracts across the Ethereum network. |
| first_indexed | 2024-03-11T03:58:46Z |
| format | Article |
| id | doaj.art-88b7706d76dc4413991956d8cc5939f7 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T03:58:46Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-88b7706d76dc4413991956d8cc5939f72023-11-18T00:19:31ZengMDPI AGApplied Sciences2076-34172023-05-011310602710.3390/app13106027Detection and Analysis of Ethereum Energy Smart ContractsBahareh Lashkari0Petr Musilek1Electrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 1H9, CanadaElectrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 1H9, CanadaAs blockchain technology advances, so has the deployment of smart contracts on blockchain platforms, making it exceedingly challenging for users to explicitly identify application services. Unlike traditional contracts, smart contracts are not written in a natural language, making it difficult to determine their provenance. Automatic classification of smart contracts offers blockchain users keyword-based contract queries and a streamlined effective management of smart contracts. In addition, the advancement in smart contracts is accompanied by security challenges, which are generally caused by domain-specific security breaches in smart contract implementation. The development of secure and reliable smart contracts can be extremely challenging due to domain-specific vulnerabilities and constraints associated with various business logics. Accordingly, contract classification based on the application domain and the transaction context offers greater insight into the syntactic and semantic properties of that class. However, despite initial attempts at classifying Ethereum smart contracts, there has been no research on the identification of smart contracts deployed in transactive energy systems for energy exchange purposes. In this article, in response to the widely recognized prospects of blockchain-enabled smart contracts towards an economical and transparent energy sector, we propose a methodology for the detection and analysis of energy smart contracts. First, smart contracts are parsed by transforming code elements into vectors that encapsulate the semantic and syntactic characteristics of each term. This generates a corpus of annotated text as a balanced, representative collection of terms in energy contracts. The use of a domain corpus builder as an embedding layer to annotate energy smart contracts in conjunction with machine learning models results in a classification accuracy of 98.34%. Subsequently, a source code analysis scheme is applied to identified energy contracts to uncover patterns in code segment distribution, predominant adoption of certain functions, and recurring contracts across the Ethereum network.https://www.mdpi.com/2076-3417/13/10/6027distributed ledgerEthereumsmart contracttransactive energy frameworkmachine learningenergy trading |
| spellingShingle | Bahareh Lashkari Petr Musilek Detection and Analysis of Ethereum Energy Smart Contracts Applied Sciences distributed ledger Ethereum smart contract transactive energy framework machine learning energy trading |
| title | Detection and Analysis of Ethereum Energy Smart Contracts |
| title_full | Detection and Analysis of Ethereum Energy Smart Contracts |
| title_fullStr | Detection and Analysis of Ethereum Energy Smart Contracts |
| title_full_unstemmed | Detection and Analysis of Ethereum Energy Smart Contracts |
| title_short | Detection and Analysis of Ethereum Energy Smart Contracts |
| title_sort | detection and analysis of ethereum energy smart contracts |
| topic | distributed ledger Ethereum smart contract transactive energy framework machine learning energy trading |
| url | https://www.mdpi.com/2076-3417/13/10/6027 |
| work_keys_str_mv | AT baharehlashkari detectionandanalysisofethereumenergysmartcontracts AT petrmusilek detectionandanalysisofethereumenergysmartcontracts |