Improving multi-choice question answering by identifying essential terms in questions
Journal Title: Revista Romana de Interactiune Om-Calculator - Year 2018, Vol 11, Issue 2
Abstract
This paper describes a method for identifying the most relevant or essential terms in a question: the words that define the meaning of the question and whose removal makes it impossible to be answered correctly even by human agents. We use an artificial neural network architecture built upon semantic and syntactic features extracted from the question. The neural network prediction represents the degree to which the given term is essential to the question. The model has been trained and validated on a dataset consisting of 2233 questions and about human 18000 labeled terms with essentialness information. It achieves an F1 score of 0.80 which is similar to other state-of-the-art approaches but requires fewer features (15 compared to 120 features used by similar works) and is much easier to train. We further show how using only essential term information can improve the accuracy of a multi-choice question answering system, based on standard information retrieval (IR) techniques, by up to 4%.
Authors and Affiliations
George-Sebastian Pirtoaca, Stefan Ruseti, Traian Rebedea
Keywords
Related Articles
- EP ID EP673684
- DOI -
- Views 133
- Downloads 0
Câteva observaţii privind relaţia dintre profilul de personalitate al utilizatorului şi preferinţa pentru utilizarea Facebook
Cercetări recente analizeazeă cu atenţie utilizarea Facebook şi impactul său asupra vieţii psihologice şi sociale a milioane de tineri care accesează zilnic această platformă de socializare. Paginile de profil al utiliza...
ObDroid: An Android permanent monitoring application using the observer pattern
Nowadays, mobile devices are the most important gadgets from our lives. We use smartphones and tablets for many activities: communication, instant messaging, emails, multimedia content sharing, social networks, e-banking...
A Corpus of Multi-participant Conversations and an Editor for its Annotation
The paper presents the details and the history of building a corpus of instant messenger (chat) conversations with multiple participants used for Computer-Supported Collaborative Learning. Two XML annotation schemas are...
Video Game Development Concepts
he paper represents a case study which provides an overview of the main concepts regarding the development of video games, and of the methods used by independent game developers to put these concepts into practice. The n...
Interactive video interface for embedded systems
This paper presents an original solution to display information on embedded systems by generating a composite video signal. For simplicity, this signal is generated without the use of additional circuits. At present such...