Information and communication society led to the production of huge volumes of content. This content is still generally non-structured (text, images, videos) and the promises of a "Web of Knowledge" are still long ahead. This situation evolves with the development of Open Data portals or resources such as DBPedia, that have made easier the access to information stored in databases (economic or demographic statistics, world knowledge contained in Wikipedia infoboxes, etc). However, most of the knowledge is still produced by textual data. Among the information concerned by the difficulty of accessing textual data, those related to events are of great interest, notably in the context of the emergence of data journalism. Data journalism have been fed until now by publicly available, statistical data, but it has paradoxically made only little use of the very journalistic materials that are events. The project ASRAEL aims at bridging this gap. Our proposal comes within the scope of the general scientific framework of information extraction (IE). We aim at extracting events from a large set of textual documents, without prior knowledge about them, and at populating and publishing a knowledge base of events. This knowledge base will be the support of a dedicated event search engine. We define event in a traditional information extraction way. An event is a structured representation of something that happens, with a nucleus, a spatio-temporal context and some arguments. The "event type" gathers comparable instances of events, as "earthquake", "election" or "car race". Arguments are attribute/value pairs that characterize an event type (for an earthquake, its location, date, magnitude, casualties...). A template is the set of arguments that can describe an event type (earthquake template, election template). The generic representation of an event is based on the rule of the "5 Ws" (What, Who, Where, When, Why) that prevails in the "Anglo-Saxon" way of writing articles. This rule stipulates that a good description of an event must make these five elements explicit. In automatic information extraction, the information about "Who", "Where" and "When" are extracted by a traditional and quite generic named entity recognition approach. On the other hand, the "What" is very domain-specific. For this reason, traditional IE systems lean on templates predefined by experts and identify events in texts with either rule-based systems or statistical models. However, in the general domain, where the huge number of possible events makes the manual definition of these templates impossible, information retrieval ("bag of words") methods take over, but do not provide a structured answer. In this project, we aim to tackle the following challenges: - Discover automatically event templates from very large text corpora, and populate a knowledge base dedicated to events. This implies a mixture of supervised and non-supervised approaches, which is necessary as soon as one consider such a generic problem. - Use this knowledge base in order to build an event aggregator and a semantic search engine. With this engine, a user (either journalist or end-user) will be able to query for an event type (e.g. earthquake) and provide filters on attribute values (location = Turkey, magnitude > 8, etc). The knowledge base will also be published following the linked data principles for other to re-use.

The main objective of our project is the generation of innovative interfaces to display information according to temporal criteria. Although our goal is closed to applications such as timelines, unlike the latter, we plan to extract and use temporal informations from the texts in order to enrich the foreseen user interfaces. The manipulated objects, called « Event-based Chronologies », prepared from semi-automated position-finding of events and of datative temporal expressions in essentially “breaking news” type texts (written in French and in English), will be associated with visualisation (multimedia) widgets enabling to visualise events associated with a “mediatic event” in chronological order; wherein said event acts somehow as the “trigger” for information search so that said event is presented relative to a context forming the collection of events which may be associated therewith. AFP currently diffuses numerous Event-based Chronologies over a wide range of mediatic events via its information departments. They are currently handled manually, by copying breaking news or documentation transmitted previously and are purely textual (since provided for the press). There are hence unsuited to multimedia, Internet and mobile usage, which has now become the rule. The purpose of this project is to provide a solution to this situation by setting ourselves the global following objectives: 1. Assist semi-automatic construction of these Event-based Chronologies by using NLP (natural language processing) techniques ; 2. View and browse multimedia Event-based Chrononlogies by using visualisation technologies. Our working programme is hence organised quite simply in the light of both these objectives. More precisely, and this the original aspect of our approach from a conceptual angle as well as regards the applications contemplated, we combine items 1. and 2. while suggesting as follows: 1’. on the one hand taking into account the problematic of different levels of temporal referencing, associated with the different types of enunciative and modal managements which can be identified within the texts; 2’. and on the other hand to contemplate the development of tools enabling to anchor events along a “multilevel” temporal visualisation scale. In the first axis, the aim is to generate, in relation to a request (the name of an event, of a person, of a team associated with a competition, etc.), propositions of Event-based Chronologies which the user (the AFP journalist in that particular instance) may optionally modify before validation. This is hence an automatic processing step of the temporality of the texts, which should integrate not only the recognition, but also the analysis of a certain type of discursive organisation in the texts. The second axis concerns the visualisation of Event-based Chronologies, and the target this time is the end-user, i.e. the reader, the internaut or the owner of a multimedia telephone. Even if our project is ambitious, it remains that the work methodology that we suggest makes it “reachable“ in its objectives, in particular regarding the realisation of an effective processing chain. Indeed, we propose to anchor our working programme on the one hand in (i) the specification of a specific need and on the other hand in (ii) a close collaboration between the different partners for defining knowledge representation formats which are compatible with the knowledge extracted from texts as well as with the knowledge corpi to view.