Nicolas Sannier

Automated extraction of product comparison matrices from informal product descriptions

We propose a procedure for extracting comparison matrices from informal product descriptions.We evaluate our proposal against numerous categories of products mined from BestBuy.Matrices exhibit numerous comparable information and can supplement or even refine technical descriptions.A user study shows that our automated approach retrieves a significant portion of correct information.Users can compute, control, edit and refine matrices in a Web environment called MatrixMiner. Domain analysts, product managers, or customers aim to capture the important features and differences among a set of related products. A case-by-case reviewing of each product description is a laborious and time-consuming task that fails to deliver a condense view of a family of product.In this article, we investigate the use of automated techniques for synthesizing a product comparison matrix (PCM) from a set of product descriptions written in natural language. We describe a tool-supported process, based on term recognition, information extraction, clustering, and similarities, capable of identifying and organizing features and values in a PCM - despite the informality and absence of structure in the textual descriptions of products.We evaluate our proposal against numerous categories of products mined from BestBuy. Our empirical results show that the synthesized PCMs exhibit numerous quantitative, comparable information that can potentially complement or even refine technical descriptions of products. The user study shows that our automatic approach is capable of extracting a significant portion of correct features and correct values. This approach has been implemented in MatrixMiner a web environment with an interactive support for automatically synthesizing PCMs from informal product descriptions. MatrixMiner also maintains traceability with the original descriptions and the technical specifications for further refinement or maintenance by users.

An automated framework for detection and resolution of cross references in legal texts

AbstractWhen identifying and elaborating compliance requirements, analysts need to follow the cross references in legal texts and consider the additional information in the cited provisions. Enabling easier navigation and handling of cross references requires automated support for the detection of the natural language expressions used in cross references, the interpretation of cross references in their context, and the linkage of cross references to the targeted provisions. In this article, we propose an approach and tool support for automated detection and resolution of cross references. The approach leverages the structure of legal texts, formalized into a schema, and a set of natural language patterns for legal cross reference expressions. These patterns were developed based on an investigation of Luxembourg’s legislation, written in French. To build confidence about their applicability beyond the context where they were observed, these patterns were validated against the Personal Health Information Protection Act (PHIPA) by the Government of Ontario, Canada, written in both French and English. We report on an empirical evaluation where we assess the accuracy and scalability of our framework over several Luxembourgish legislative texts as well as PHIPA.

A model-based framework for probabilistic simulation of legal policies

Legal policy simulation is an important decision-support tool in domains such as taxation. The primary goal of legal policy simulation is predicting how changes in the law affect measures of interest, e.g., revenue. Currently, legal policies are simulated via a combination of spreadsheets and software code. This poses a validation challenge both due to complexity reasons and due to legal experts lacking the expertise to understand software code. A further challenge is that representative data for simulation may be unavailable, thus necessitating a data generator. We develop a framework for legal policy simulation that is aimed at addressing these challenges. The framework uses models for specifying both legal policies and the probabilistic characteristics of the underlying population. We devise an automated algorithm for simulation data generation. We evaluate our framework through a case study on Luxembourgs Tax Law.

Defining and retrieving themes in nuclear regulations

Safety systems in nuclear industry must conform to an increasing set of regulatory requirements. These requirements are scattered throughout multiple documents expressing different levels of requirements or different kinds of requirements. Consequently, when licensees want to extract the set of regulations related to a specific concern, they lack explicit traces between all regulation documents and mostly get lost while attempting to compare two different regulatory corpora. This paper presents the regulatory landscape in the context of digital Instrumentation and Command systems in nuclear power plants. To cope with this complexity, we define and discuss challenges toward an approach based on information retrieval techniques to first narrow the regulatory research space into themes and then assist the recovery of these traceability links.

MatrixMiner: a red pill to architect informal product descriptions in the matrix

Domain analysts, product managers, or customers aim to capture the important features and differences among a set of related products. A case-by-case reviewing of each product description is a laborious and time-consuming task that fails to deliver a condensed view of a product line. This paper introduces MatrixMiner: a tool for automatically synthesizing product comparison matrices (PCMs) from a set of product descriptions written in natural language. MatrixMiner is capable of identifying and organizing features and values in a PCM – despite the informality and absence of structure in the textual descriptions of products. Our empirical results of products mined from BestBuy show that the synthesized PCMs exhibit numerous quantitative, comparable information. Users can exploit MatrixMiner to visualize the matrix through a Web editor and review, refine, or complement the cell values thanks to the traceability with the original product descriptions and technical specifications.

Model-based simulation of legal policies: framework, tool support, and validation

Simulation of legal policies is an important decision-support tool in domains such as taxation. The primary goal of legal policy simulation is predicting how changes in the law affect measures of interest, e.g., revenue. Legal policy simulation is currently implemented using a combination of spreadsheets and software code. Such a direct implementation poses a validation challenge. In particular, legal experts often lack the necessary software background to review complex spreadsheets and code. Consequently, these experts currently have no reliable means to check the correctness of simulations against the requirements envisaged by the law. A further challenge is that representative data for simulation may be unavailable, thus necessitating a data generator. A hard-coded generator is difficult to build and validate. We develop a framework for legal policy simulation that is aimed at addressing the challenges above. The framework uses models for specifying both legal policies and the probabilistic characteristics of the underlying population. We devise an automated algorithm for simulation data generation. We evaluate our framework through a case study on Luxembourg’s Tax Law.

Legal Markup Generation in the Large: An Experience Report

Legal markup (metadata) is an important prerequisite for the elaboration of legal requirements. Manually encoding legal texts into a markup representation is laborious, specially for large legal corpora amassed over decades and centuries. At the same time, automating the generation of markup in a fully accurate manner presents a challenge due to the flexibility of the natural-language content in legal texts and variations in how these texts are organized. Following an action research method, we successfully collaborated with the Government of Luxembourg in transitioning five major legislative codes from plain-text to a legal markup format. Our work focused on generating markup for the structural elements of the underlying codes. The technical basis for our work is an adaptation and enhancement of an academic markup generation tool developed in our prior research [1]. We reflect on the experience gained from applying automated markup generation at large scales. In particular, we elaborate the decisions we made in order to strike a cost-effective balance between automation and manual work for legal markup generation. We evaluate the quality of automatically-generated structural markup in real-world conditions and subject to the practical considerations of our collaborating partner.

From RELAW Research to Practice: Reflections on an Ongoing Technology Transfer Project

Over the past years, we have been studying the topic of automated metadata extraction from legal texts. While our research has been motivated primarily by RE problems, we have observed that the interdisciplinarity of the research on legal metadata, and indeed on several other topics considered by the RELAW community, has the potential to trigger innovation beyond the traditional RE. In particular, legal metadata is a key enabler for the rapidly-expanding field of Legal Technology (LegalTech). In this short paper, we describe the preliminary steps we have taken toward transitioning a prototype tool for legal metadata extraction (developed in our previous work) into a platform that is palatable to the LegalTech market. We hope that our findings would provide useful insights about the value chain for legal metadata and further offer a concrete example of a technology transfer attempt that is rooted in RELAW research.