Analysis of Solution Diversity in Topic Models for Smart City Applications

Toshio Uchiyama and Tsukasa Hokimoto

A chapter in Sustainable Smart Cities - A Vision for Tomorrow from IntechOpen

Abstract: Topic models are known to be useful tools for modeling and analyzing high-dimensional count data such as documents. In a smart city, it is important to collect and analyze citizens' voices to discover their concerns and issues. Topic modeling is effective for the above analysis because it can extract topics from a collection of documents. However, when estimating parameters (solutions) in topic models, various solutions are reached due to differences in algorithms and initial values. In order to select a solution suitable for the purpose from among the various solutions, it is necessary to know what kind of solutions exist. This chapter introduces methods for analyzing diverse solutions and obtaining an overall picture of the solutions.

Keywords: topic model; diversity of solution; normalized mutual information; typification of solutions; topic distribution; word distribution; information-theoretic clustering (search for similar items in EconPapers)
JEL-codes: Q56 (search for similar items in EconPapers)
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.intechopen.com/chapters/82818 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ito:pchaps:274299

DOI: 10.5772/intechopen.106009

Access Statistics for this chapter

More chapters in Chapters from IntechOpen
Bibliographic data for series maintained by Slobodan Momcilovic ().