Semantisation of smart contracts

With the development of smart contracts, for their integration into the real business, it will be necessary to solve a number of problems of accounting for goods, which are now handled manually. For this, it is necessary to create a mechanism for the semantic construction and implementation of smart contracts for ordinary users without the skill of programming.

While in the block is testing smart contracts with 1-2-3 names of goods sold, the problem of re-sorting is not obvious. However, in real business, when invoices contain hundreds or even thousands of items (and this is an assortment of only a small store), inevitably there are unmanageable differences in the names of the same goods, their characteristics, both consumer (color) and technical . This leads to incorrect accounting of goods in warehouses and points of sale - to re-sorting. In the existing business model, this problem is solved by involving an average level of logistics managers - storekeepers, commodity experts. However, to fully automate the work of a smart contract, you need other solutions, otherwise you will have to check every position in a smart contract manually.

To do this, it is necessary to create a mechanism for additional semantisation of the smart contract, so that accounting systems have the ability to extract the necessary information not only from the title and description of the product in the human language, but also from additional semantic fields. It is possible to implement this on the basis of the RDF-resource definition framework, a project that grew out of the Semantic Web and OWL-ontology web language. The most elaborate is the RDF classifier , developed and maintained by Google, Microsoft, Yahoo and Yandex. Now the classifier has 589 types, 860 properties, and 114 numbered values.

The mechanism for constructing and implementing smart contracts is implemented in the form of the RDF W3C (resource definition framework) designer in accordance with, by adding a UI that allows you to describe entities and account persons in accordance with the standard in order to uniquely interpret their properties and relationships, Which will allow the implementation of "machine-readable" contracts.

Technically, in order to implement a system of semantic smart contracts, it is necessary to make a comparison between the nomenclature and the RDF node in the classifier of goods (services, other entities) and their properties in the ERP system. When creating a smart contract - write the values ​​of these properties in the format Then, upon reading, the smart contract destination can not only see the name of the product, but also a standardized description of its properties, which will automatically classify the product (service) and determine the correctness of the smart contract fulfillment at the semantic level.

We also want to do this in the DARFChain project: