Discretization of Time-Series Behavioral Data and Rule Generation based on Temporal Context

Iqbal H. Sarker, Alan Colman, Jun Han and Paul Watters
Additional contact information
Iqbal H. Sarker: Swinburne University of Technology
Alan Colman: Swinburne University of Technology
Jun Han: Swinburne University of Technology
Paul Watters: Macquarie University

Chapter Chapter 5 in Context-Aware Machine Learning and Mobile Data Analytics, 2021, pp 75-92 from Springer

Abstract: Abstract In this chapter, we explore the discretization of the continuous time-series data to generate temporal segments according to the behavioral patterns of the users, which is used as the basis of generating rules based on temporal context. Although a static segmentation approach is straightforward to comprehend and can be beneficial for analyzing population behavior by comparing across individuals, the generated static segments do not always map to individual user activity patterns and subsequent behavior. As a result, we focus on dynamic segmentation by taking into account individuals’ various usage or activities with their phones during the week. We present a behavior-oriented time segmentation methodology that uses phone usage data to generate optimal time segments of individuals with similar behavioral characteristics. Moreover, how the generated segments can be used to produce a set of temporal behavioral rules according to users’ preferences, has been presented. Finally, some experimental results to show the effectiveness of the technique have also been provided.

Keywords: Time-series; Discretization; Machine learning; Temporal context; Time-series segmentation; Temporal rules; User behavior modeling; Time-based system (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

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:spr:sprchp:978-3-030-88530-4_5

Ordering information: This item can be ordered from
http://www.springer.com/9783030885304

DOI: 10.1007/978-3-030-88530-4_5

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().