Periodic Reporting for period 2 - Lynx (Building the Legal Knowledge Graph for Smart Compliance Services in Multilingual Europe)
Periodo di rendicontazione: 2019-06-01 al 2021-03-31
National, regional and local administrations in the EU publish legislation in different formats and languages. This double fragmentation of law and languages hampers the development of a unified market with fluid commercial exchanges in Europe. Whereas the legal department of large companies can easily cope with these heterogeneous compliance problems, SMEs have to face important barriers to engage in trade abroad.
At the same time, language technologies and information retrieval techniques have achieved the necessary maturity to support automated compliance services. This project is important to demonstrate that this approach is viable, and that the use of standard technologies, open data resources, and common data models enable industry-level quality solutions for compliance.
The first objective of this project is the creation of the Lynx multilingual Legal Knowledge Graph, a knowledge base related to compliance that integrates data and documents from multiple jurisdictions, in various languages, as well as open standards and sectorial best practice guidelines. The second objective is the creation and exploitation of a number of interoperable compliance-related software services whose joint use as building blocks can help in complex scenarios. These services are to be used in the ingestion and integration of documents related to compliance, its annotation and analysis, and finally its ulterior discovery by information retrieval services. The third objective is to demonstrate the composition of these services in three pilots that responds to the industry needs by three different business cases: (i) contract analysis across jurisdictions; (ii) compliance with legislation and industry standards in the geothermal energy sector; (iii) access to labour law information in cross-lingual, cross-jurisdictional scenarios.
At the same time, language technologies and information retrieval techniques have achieved the necessary maturity to support automated compliance services. This project is important to demonstrate that this approach is viable, and that the use of standard technologies, open data resources, and common data models enable industry-level quality solutions for compliance.
The first objective of this project is the creation of the Lynx multilingual Legal Knowledge Graph, a knowledge base related to compliance that integrates data and documents from multiple jurisdictions, in various languages, as well as open standards and sectorial best practice guidelines. The second objective is the creation and exploitation of a number of interoperable compliance-related software services whose joint use as building blocks can help in complex scenarios. These services are to be used in the ingestion and integration of documents related to compliance, its annotation and analysis, and finally its ulterior discovery by information retrieval services. The third objective is to demonstrate the composition of these services in three pilots that responds to the industry needs by three different business cases: (i) contract analysis across jurisdictions; (ii) compliance with legislation and industry standards in the geothermal energy sector; (iii) access to labour law information in cross-lingual, cross-jurisdictional scenarios.
The main achievements of the Lynx project can be summarised as follows:
1) A multilingual legal knowledge graph (LKG) that integrates data and documents from multiple jurisdictions, in various languages, as well as open standards and sectorial best practice guidelines. The original documents are transformed into a homogeneous format according to the data model defined by the Legal Knowledge Graph Ontology. For this purpose, more than 30 harvesters and a so-called curation workflow have been set in place to “ingest” documents, that is, to collect documents, enrich them with annotations (named entities, temporal expressions, entity relations, etc.), and make them ready for consumption.
2) A CKAN-based Lynx Data Portal to document datasets of interest for the regulatory domains, published according to the FAIR data principles proposed by the EC. Multilingual terminologies and lexical data resources have been created and/or made available to Lynx business cases in Semantic Web formats to easy their integration and reuse.
3) The implementation of the Lynx Services Platform (LynxSP), an online platform that provides the usage of (a) single and standalone or (b) piped/orchestrated Lynx services (in total 16 services in place) that can be ordered/used as Software as a Service (Cloud Services). Such services allow a unified treatment of Lynx data and documents, and cover the following functionalities: Word Sense Induction and Disambiguation (WSID), Dictionary Access (DA), Terminology Extraction (TermEx), Name Entity Recognition (NER), Temporal Expression Extraction (TimEx), Geographical Entity Recognition (Geo), Relation Extraction (RelEx), Entity Extraction (EntEx), Cross-lingual search (Sear), Question Answering (QADoc), Semantic Similarity (SeSim), Terminology Query (TermQ), Summarization (Summ), and Neural Machine Translation (Trans). Using standard practices, these services are well documented (OpenAPI), secured (OAuth), containerized (Docker) and managed (OpenShift).
4) Three pilot applications:“Pilot 1: Contract Analysis”, whose goal is automated contract analysis and contract archiving; “Pilot 2: Geothermal Energy”, mainly centered on the efficient access and comparison of regulations and industry practices in the energy sector in Europe; and “Pilot 3: Labour Law”, focused on an efficient access to cross-lingual and cross-jurisdictional data.
5) A website with a summary of project objectives and links to publications and project results has been created and is mantained. Also, a Zenodo community has been created were all deliverables, scientific publications and datasets are openly available.
6) Organisation of events for academia and industry in order to to disseminate project results (TeReCom workshop series, JURIX Industry Event). A series of Webinars has been organised and videos have been made available in a YouTube channel.
7) An important and ongoing commercialization effort. Project results have been pitched to companies outside the consortium and to clients of Lynx partners to perform initial market validations. Business and pricing models have been designed with a focus on a Business to Business model. A sustainability plan for Lynx Services has been developed to support the exploitation plans through the European Language Grid platform.
1) A multilingual legal knowledge graph (LKG) that integrates data and documents from multiple jurisdictions, in various languages, as well as open standards and sectorial best practice guidelines. The original documents are transformed into a homogeneous format according to the data model defined by the Legal Knowledge Graph Ontology. For this purpose, more than 30 harvesters and a so-called curation workflow have been set in place to “ingest” documents, that is, to collect documents, enrich them with annotations (named entities, temporal expressions, entity relations, etc.), and make them ready for consumption.
2) A CKAN-based Lynx Data Portal to document datasets of interest for the regulatory domains, published according to the FAIR data principles proposed by the EC. Multilingual terminologies and lexical data resources have been created and/or made available to Lynx business cases in Semantic Web formats to easy their integration and reuse.
3) The implementation of the Lynx Services Platform (LynxSP), an online platform that provides the usage of (a) single and standalone or (b) piped/orchestrated Lynx services (in total 16 services in place) that can be ordered/used as Software as a Service (Cloud Services). Such services allow a unified treatment of Lynx data and documents, and cover the following functionalities: Word Sense Induction and Disambiguation (WSID), Dictionary Access (DA), Terminology Extraction (TermEx), Name Entity Recognition (NER), Temporal Expression Extraction (TimEx), Geographical Entity Recognition (Geo), Relation Extraction (RelEx), Entity Extraction (EntEx), Cross-lingual search (Sear), Question Answering (QADoc), Semantic Similarity (SeSim), Terminology Query (TermQ), Summarization (Summ), and Neural Machine Translation (Trans). Using standard practices, these services are well documented (OpenAPI), secured (OAuth), containerized (Docker) and managed (OpenShift).
4) Three pilot applications:“Pilot 1: Contract Analysis”, whose goal is automated contract analysis and contract archiving; “Pilot 2: Geothermal Energy”, mainly centered on the efficient access and comparison of regulations and industry practices in the energy sector in Europe; and “Pilot 3: Labour Law”, focused on an efficient access to cross-lingual and cross-jurisdictional data.
5) A website with a summary of project objectives and links to publications and project results has been created and is mantained. Also, a Zenodo community has been created were all deliverables, scientific publications and datasets are openly available.
6) Organisation of events for academia and industry in order to to disseminate project results (TeReCom workshop series, JURIX Industry Event). A series of Webinars has been organised and videos have been made available in a YouTube channel.
7) An important and ongoing commercialization effort. Project results have been pitched to companies outside the consortium and to clients of Lynx partners to perform initial market validations. Business and pricing models have been designed with a focus on a Business to Business model. A sustainability plan for Lynx Services has been developed to support the exploitation plans through the European Language Grid platform.
The results of the Lynx project come to contribute to the realisation of the “European Area of Justice”, as defined in “The EU Justice Agenda for 2020 - Strengthening Trust, Mobility and Growth within the Union”. With the creation of the multilingual Legal Knowledge Graph (LKG), Lynx provides more efficient access to aggregated and complementary legal and regulatory information from different sources and jurisdictions, building bridges between Member States’ legal systems. This is mainly enabled by Semantic Web standards and technologies, which guarantee homogeneous access to open legal data.
In parallel, the Lynx Services Platform (LynxSP), which consists of a set of smart legal and regulatory services ready to be used by third companies, lays the foundations for mass-customised assurance to SMEs, large companies, and citizens in general. Reusability and scalability of the software services are granted by the adoption of standard practices with respect to encapsulation, security, documentation, and management. This may contribute to the economic growth in Europe, since it will help in building the necessary trust that businesses and consumers need in a truly single market. The impact of the solutions based on Lynx is expected to reduce the time devoted to find the appropriate data and the costs incurred by companies when managing compliance.
The three pilot solutions build on top of the LKG and the LynxSP are a good demonstrator of how Lynx solutions can help companies in reducing costs and efforts when monitoring legislation and regulations, or solving questions related to compliance. Lynx pilots have been validated in the market and have attracted the attention of companies in the legaltech market sector.
Citizens can also benefit from Lynx solutions, since they get open access to aggregated European legislation, regulations, and open standards, as well as mass-customised information, including highlighting and tagging. The solutions provided by Lynx for multilingual and cross-domain data integration in the field of legal information, standard good practices and industry norms, will translate into a broad range of applications and solutions that will bring a clear benefit to European citizens.
In parallel, the Lynx Services Platform (LynxSP), which consists of a set of smart legal and regulatory services ready to be used by third companies, lays the foundations for mass-customised assurance to SMEs, large companies, and citizens in general. Reusability and scalability of the software services are granted by the adoption of standard practices with respect to encapsulation, security, documentation, and management. This may contribute to the economic growth in Europe, since it will help in building the necessary trust that businesses and consumers need in a truly single market. The impact of the solutions based on Lynx is expected to reduce the time devoted to find the appropriate data and the costs incurred by companies when managing compliance.
The three pilot solutions build on top of the LKG and the LynxSP are a good demonstrator of how Lynx solutions can help companies in reducing costs and efforts when monitoring legislation and regulations, or solving questions related to compliance. Lynx pilots have been validated in the market and have attracted the attention of companies in the legaltech market sector.
Citizens can also benefit from Lynx solutions, since they get open access to aggregated European legislation, regulations, and open standards, as well as mass-customised information, including highlighting and tagging. The solutions provided by Lynx for multilingual and cross-domain data integration in the field of legal information, standard good practices and industry norms, will translate into a broad range of applications and solutions that will bring a clear benefit to European citizens.