NICOP - Learning an effective representation for the hidden semantics

Abstract

The goal in this project is to develop an effective framework for learning a representationof the hidden semantics from massive tex"t collections. Unlike existing approaches, ourframework should be able to deal with different forms of semantics, such as wordmean""ings, themes, entity interaction, emotions, trends, social communities, etc. Manychallenges have to be resolved, such as the non-co""nvexity of the learning problem, theunstructured nature of text, the heterogeneity of texts, links, images, tags, etc.Nonetheless,"" a good way to represent and learn the hidden semantics will have asignificant impact in many areas. To this end, our methodology w""ill base on topicmodeling [Blei, 2012], neural networks [LeCun et al., 2015], manifold learning [Niyogi,2013], and stochastic opti""mization [Bottou, 1998]. The framework will be employed inpractical applications including text modeling, sentiment analysis, recom""mendersystems, social network analysis. Results from this project include published papers athigh-quality journals/conferences, an"d a library for public use.This research is relevant to ONRG with the focus area of Information Dominance~Cyber.

Document Details

Document Type
DoD Grant Award
Publication Date
Mar 26, 2018
Source ID
N629091812072

Entities

People

  • Khoat Than

Organizations

  • Hanoi University of Science and Technology
  • Office of Naval Research
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval
  • AI & ML - Neural Networks
  • Cyber