DILiGENt: Domain-Independent Language Generation

We propose a two year project to develop a novel data-driven methodology to rapidly create high quality NLG systems for new domains, by combining recent advances in three domains:
(1) advances in statistical models for NLG,
(2) crowdsourcing methods for natural language data collection, which have shown first promising results in related fields, such as Machine Translation, and
(3) recently developed imitation learning algorithms for structured prediction.

The project team combines expertise of two leading research groups in these areas:
At Heriot-Watt University, we recently demonstrated the potential for data-driven statistical NLG in limited domains. In order to make this framework domain-independent we will leverage recent machine learning models, developed by researchers at the University College London. These models learn by imitating the actions a human expert would perform to generate NL utterances, which we collect via a tightly integrated crowdsourcing procedure. The outcome of this work is a framework which will allow the rapid development of NLG systems for new domains, and thus accelerate the impact NLG technology has on the market.

We will showcase this framework on a dataset provided by the BBC, where we address the problem of generating weather reports for over 20,000 individual locations. Currently, the BBC website features only 10 reports written by meteorologists. Each of these reports covers a rather large area of the country (e.g. East of England), and thus of little interest to their users who are usually interested in the weather in a particular location (e.g. Norwich).

In a second, more ambitious step, we will explore how this framework scales to more complex interactive dialogue settings, where generation has to account for discourse phenomena, such as long-distance discourse relations or syntactic coordination. This will be evaluated in a shared task challenge for generation in interactive systems, hosted by Heriot-Watt University.

In sum, this project will further our understanding of domain-independent language generation, as well as deliver substantial and novel resources to support future research in this area (in the forms of code and data), and practical implementations of NLG systems in a wide-range of domains, from weather reports to natural language interfaces.

(Please note that part of the following text is taken form the National Importance section of Part II)

Machine Learning for Natural Language Processing (NLP) is an area which has begun to generate positive economic impact, and start-ups and large companies such as Microsoft, Amazon, and Yahoo! are making substantial, new investments in this field. The UK has one of the highest proportion of world-leading NLG research centres. However, in contrast to other areas of NLP, the UK is still under-represented in using machine learning approaches for NLG. This proposal will help to strengthen this research strand in the UK.

From an end-user importance point of view, this work will be of interest to a wide range of businesses and companies in the UK (see letters of support). NLG creates more flexible output for interactive natural language interfaces. Repetitive linguistic output is one of the main problems for intelligent personal assistants (see for example Apple's Siri or Google Voice). As such, NLG technology has a strong potential for commercialisation, as the successful company ARRIA NLG has demonstrated.

The overall aim of this research - the rapid development of NLG systems for new domains - is to accelerate the impact that NLG technologies have on the market. To this end, we will assess our framework in a real-world applications, for example generating local weather reports for the BBC website. Furthermore, we will actively seek ways to commercialise this research, such as patenting and licensing of the framework that we develop.

From a societal importance point of view, NLG technology directly contributes to addressing key UK societal challenges: text simplification and data-to-text systems will be increasingly used to enable easy access to information across society. This will remove barriers to the use and understanding of information from large volumes of data. NLG systems have also been used in modern health technology, especially in the areas of personalised and localised healthcare.