CNL 2014

Main | Program | Posters/Demos | Info for Participants | Journal Issue



We invite everybody to present their work in the form of a poster and/or demo. Just send us ( your names and affiliations, a short description (one paragraph), and optionally a web link. These demo/poster descriptions will be made available here (but are not included in the proceedings). The extended deadline for the submission of posters/demos is 19 August 2014.

Announced Posters/Demos

First-Order Reasoning for Attempto Controlled English

Norbert E. Fuchs, University of Zurich, Switzerland

RACE is a first-order reasoner for Attempto Controlled English (ACE) that can show the logical (in-) consistency of ACE texts, deduce one ACE text from another one, and answer ACE queries from ACE texts. In each case RACE gives a proof justification in ACE and full English. RACE covers almost all of ACE including arithmetic.

Fact extraction and reasoning using a Controlled Natural Language

David Mott, IBM United Kingdom Ltd
Stephen Poteet, Boeing Research & Technology, USA
Ping Xue, Boeing Research & Technology, USA
Anne Kao, Boeing Research & Technology, USA
Ann Copestake, University of Cambridge, UK
Cheryl Giammanco, US Army Research Laboratory

We describe our research on fact extraction and reasoning using a Controlled Natural Language undertaken as part of the International Technology Alliance programme (ITA). Fact extraction from unstructured sources in Natural Language (NL) and reasoning about the facts extracted are key aspects of collaborative problem solving and analysis performed by human analysts when seeking to obtain high value information. Our research seeks ways to facilitate these tasks by use of ITA Controlled English (ITA CE), a subset of natural English that can be written and understood by both man and machine. We use an open source NL parsing system, the DELPH-IN English Resource Grammar (ERG), integrated into an ITA CE reasoning system that transforms the linguistic output from the ERG into ITA CE facts expressed in the terms of the user's domain model. In addition, some sentences require transformation into rules rather than simple facts. These transformations are effected by a set of ITA CE rules expressing the linguistic reasoning, based upon a series of semantic mappings from linguistic representations, through a general semantic model, to the domain specific model. NL sentences to be interpreted may be written in a Microsoft Word document, and turned into ITA CE facts by use of Word commands, placing the extracted ITA CE back into the document. The Word documents also provide commands to facilitate the exploration of rationale for the reasoning and the development of new reasoning and user's domain models. Subsequent reasoning applied to these extracted facts is shown in the context of the "ELICIT identification task", where human analysts must determine key aspects of a pending "attack" from a set of NL sentences. The research focuses on solving the problem of "who" are the participants of the attack, and shows the extraction of ITA CE facts and rules from simplified versions of the original ELICIT sentences, the reasoning using an ITA CE domain model, problem solving strategy and background knowledge to determine the participants and the visualisation of the rationale for the conclusions.

Two views, one model: tools for working with contracts in CNL and as C-O diagrams

John J. Camilleri, Chalmers / University of Gothenburg, Sweden

We will demonstrate a prototype of a web-based tool for working with our custom CNL for contracts. In addition to composing contracts with completion and checking them for syntactic correctness, the CNL tool can also send the contract to a complementary diagram editor tool, which allows the contract model to be edited graphically before re-exporting as CNL.

Fluent Editor

Paweł Kapłański, Cognitum Sp. z o.o., Poland

We will present the construction process of a model of knowledge in Fluent Editor. Fluent Editor is an ontology editor developed at Cognitum. It is a comprehensive tool for editing and manipulating complex knowledge bases and makes usage of a version of Controlled Natural Language as a specific CNL/OWL interface. By design, this version is compatible with OWL2 ontology standard. We would like to show the real-life, Telecom Company example: how to build an ontology model for inferencing a best offer. In a modern rapidly changing business environment companies need to quickly respond to these changes to gain operational excellence. Typical Corporation has a central business analysts' team that is constantly monitoring overall business performance, benchmarking the market and adjusting up-to-date business strategy and sales tactics. This tactics needs further to be deployed and executed by the army of sales people widely spread geographically. The Telecommunication Company that is referenced here has more than 1200 POS (point of sale) in the whole country. Sales process per-formed at each of them shall be coordinated on a daily basis with the recent company sales strategy in accordance to the customer needs, up-to-date product and service portfolio, technical assessment, and various locally available sales-promotions. With CNL those issues could be addressed in a convenient way both for the human (business analysts) and computer (to run application assessing customer needs and automatically generate the best offer proposal) perspectives.

ISO Language Resource Management Technical Specification Proposal for Controlled Natural Language: Basic Concepts and General Principles

Hitoshi Isahara
Key-Sun Choi
Christian Galinski
Kiyong Lee
Cunningham Michael

This poster aims to lay the basis for the development of an international standard for language resource management focused on controlled natural language. There are a number of different interpretations of the notion of 'control' alone, and the construction of effective or useful CNLs and their evaluation therefore require a systematic approach to characterizing a variety of CNLs. As the first step to satisfy such a requirement, this proposal sets out general definitions and principles as a basis for standardizing purpose- and domain-oriented CNLs in precise and concrete terms.