Back to Search
Start Over
Constructing Topic Models of Internet of Things for Information Processing
- Source :
- The Scientific World Journal, Vol 2014 (2014)
- Publication Year :
- 2014
- Publisher :
- Wiley, 2014.
-
Abstract
- Internet of Things (IoT) is regarded as a remarkable development of the modern information technology. There is abundant digital products data on the IoT, linking with multiple types of objects/entities. Those associated entities carry rich information and usually in the form of query records. Therefore, constructing high quality topic hierarchies that can capture the term distribution of each product record enables us to better understand users’ search intent and benefits tasks such as taxonomy construction, recommendation systems, and other communications solutions for the future IoT. In this paper, we propose a novel record entity topic model (RETM) for IoT environment that is associated with a set of entities and records and a Gibbs sampling-based algorithm is proposed to learn the model. We conduct extensive experiments on real-world datasets and compare our approach with existing methods to demonstrate the advantage of our approach.
- Subjects :
- Technology
Medicine
Science
Subjects
Details
- Language :
- English
- ISSN :
- 23566140 and 1537744X
- Volume :
- 2014
- Database :
- Directory of Open Access Journals
- Journal :
- The Scientific World Journal
- Publication Type :
- Academic Journal
- Accession number :
- edsdoj.b392edccbf034488ba0e2b55b4b31c3f
- Document Type :
- article
- Full Text :
- https://doi.org/10.1155/2014/675234