General learning approach for event extraction: Case of management change event

Abstract

Starting from an ontology of a targeted financial domain corresponding to transaction, performance and management change news, relevant segments of text containing at least a domain keyword are extracted. The linguistic pattern of each segment is automatically generated to serve initially as a learning model. Each pattern is composed of named entities, keywords and articulation words. Some generic named entities like organizations, persons, locations, dates and grammatical annotations are generated by an automatic tool. During the learning step, each relevant segment is manually annotated with respect to the targeted entities (roles) structuring an event of the ontology. Information extraction is processed by associating a role with a specific entity. By alignment of generic entities to specific entities, some strings of a text are automatically annotated. An original learning approach is presented. Experiments with the management change event showed how recognition rates are improved by using different generalization tools.

Event detection based on open information extraction and ontology

“…This alleviates the problems with the statistical methods concerning the meaning of the text. For example, the GLAEE approach (Elloumi et al, 2013) is based on the generation of annotation patterns that involves a list of keywords and cue words. Doing so, we identify events in the learning phase, afterwards the annotation phase is performed by an alignment between the pattern and the new text.…”

“…We have studied their functionality and, according to their limits, we have introduced our new approach. GLAEE (Elloumi et al, 2013) is an approach based on the generation of annotation patterns which involves a list of keywords and cue words. It purposes to identify events by an alignment between the pattern and the new text.…”

“…The relation identification is also a worthy of mention task in the field of IE, which aims to find the mention of a binary relation between two entities in a text (Elloumi et al, 2013). Another specific type of knowledge that can be extracted from the text is the event and can be considered as an object that admits an existence in the time space and depends on other objects (Elloumi et al, 2013). The event extraction is domain dependent.…”

Where Is the News Breaking? Towards a Location-Based Event Detection Framework for Journalists

“…Natural language processing (NLP) has been used for event detection from an information extraction (IE) perspective [8]. While the use of NLP encourages the extraction of entities enclosed in text, a supervised learning approach can extract more information about entities and context.…”

Audio-Textual Classification System Design for Arabic News Videos

“…Jaam et al [8,9] have developed new algorithms for Arabic text summarization, and automatic data classification. Elloumi et al [10] developed data reduction and redundancy elimination algorithms, and knowledge extraction from Arabic and English news [11]. While technologies have been developed attempting to tackle the semantic analysis of digital content, there exists no global approach or solution to the low-level analysis of Arabic media.…”