Network learning forum posts topic discovery

Information plane analysis, describing the mutual information between the input and a hidden layer and between a hidden layer and the target over time, has recently been proposed to analyze the training of neural networks. Since the activations of a hidden layer are typically continuous-valued, this mutual information cannot be computed analytically and must thus be estimated, resulting in apparently inconsistent or even contradicting results in the literature. The goal of this paper is to demonstrate how information plane analysis can still be a valuable tool for analyzing neural network training. To this end, we complement the prevailing binning estimator for mutual information with a geometric interpretation. With this geometric interpretation in mind, we evaluate the impact of regularization and interpret phenomena such as underfitting and overfitting. In addition, we investigate neural network learning in the presence of noisy data and noisy labels.

Download Full-text

Automatic annotation of weakly-tagged social images on flickr using latent topic discovery of multiple groups

Proceedings of the 2009 workshop on Ambient media computing - AMC '09 ◽

10.1145/1631005.1631023 ◽

2009 ◽

Cited By ~ 1

Author(s):

Ye Chen ◽

Jian Shao ◽

Ke Zhu

Keyword(s):

Automatic Annotation ◽

Topic Discovery ◽

Social Images ◽

Latent Topic ◽

Multiple Groups

Download Full-text

Building Tag-Aware Groups for Music High-Order Ranking and Topic Discovery

International Journal of Multimedia Data Engineering and Management ◽

10.4018/jmdem.2010070101 ◽

2010 ◽

Vol 1 (3) ◽

pp. 1-18 ◽

Cited By ~ 2

Author(s):

Dimitrios Rafailidis ◽

Alexandros Nanopoulos ◽

Yannis Manolopoulos

Keyword(s):

Real Data ◽

Social Tagging ◽

Topic Discovery ◽

Retrieval Systems ◽

The Social ◽

Information Retrieval Systems ◽

Aware Group ◽

User Groups ◽

Music Information ◽

Ranking Analysis

In popular music information retrieval systems, users have the opportunity to tag musical objects to express their personal preferences, thus providing valuable insights about the formulation of user groups/communities. In this article, the authors focus on the analysis of social tagging data to reveal coherent groups characterized by their users, tags and music objects (e.g., songs and artists), which allows for the expression of discovered groups in a multi-aspect way. For each group, this study reveals the most prominent users, tags, and music objects using a generalization of the popular web-ranking concept in the social data domain. Experimenting with real data, the authors’ results show that each Tag-Aware group corresponds to a specific music topic, and additionally, a three way ranking analysis is performed inside each group. Building Tag-Aware groups is crucial to offer ways to add structure in the unstructured nature of tags.

Download Full-text

Innovative Applications Mode of Network Learning Space in Exercise Physiology Based on Ubiquitous Learning

International Journal of Emerging Technologies in Learning (iJET) ◽

10.3991/ijet.v14i04.10117 ◽

2019 ◽

Vol 14 (04) ◽

pp. 113 ◽

Cited By ~ 3

Author(s):

Jie Kong

Keyword(s):

Flipped Classroom ◽

Exercise Physiology ◽

Internet Technology ◽

Traditional Teaching ◽

Ubiquitous Learning ◽

Learning Space ◽

Network Teaching ◽

Network Learning ◽

Space Construction

With continuous development of internet technology, the concept of ubiquitous learning and network learning space have received more and more attention from scholars, and gradually become the research focuses. College classroom has turned to network teaching from traditional teaching. In this study, literature review and case study were combined with ubiquitous learning and network learning space construction to systematically discuss classification and concept models of network learning space under the perspective of ubiquitous learning. Meanwhile, four models based on network learning space were proposed, and flipped classroom network teaching model was applied in the course of Exercise Physiology. The study showed that, the model has the good teaching effect in course teaching. It not just improves students’ interest, but also lays a foundation for popularizing the teaching mode.

Download Full-text