2018
DOI: 10.1016/j.procs.2018.08.078
|View full text |Cite
|
Sign up to set email alerts
|

Hardcore Gamer Profiling: Results from an unsupervised learning approach to playing behavior on the Steam platform

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
11
0
2

Year Published

2020
2020
2024
2024

Publication Types

Select...
5
2

Relationship

1
6

Authors

Journals

citations
Cited by 33 publications
(14 citation statements)
references
References 24 publications
1
11
0
2
Order By: Relevance
“…First, means are heavily influenced by extreme scores. While the average esports gameplay hours were ∼7 h per week (Nielsen, 2018), according to more than 700,000 hardcore gamers profiles, a minimum 1-week play time was 20 h (Baumann et al, 2018). The high gameplay hours might pull the mean up so that average hours per week might not be an exact standard of the general frequency.…”
Section: Methodsmentioning
confidence: 99%
“…First, means are heavily influenced by extreme scores. While the average esports gameplay hours were ∼7 h per week (Nielsen, 2018), according to more than 700,000 hardcore gamers profiles, a minimum 1-week play time was 20 h (Baumann et al, 2018). The high gameplay hours might pull the mean up so that average hours per week might not be an exact standard of the general frequency.…”
Section: Methodsmentioning
confidence: 99%
“…A következő szempont a játékidő, ez alapján vizsgálták Baumann et al (2018) a Steam -játékokat egybegyűjtő szolgáltató adatait -ahol tíz meghatározó videójátékot írnak le a játszási idő alapján, amelyek kitűntek a megvizsgált 3 537 játék közül. Ezek közé tartozott a Dota 2, Team Fortress 2, Counter-Strike, Sid Meier's Civilization V., Counter-Strike: Source, Counter-Strike: Global Offensive, Garry's Mood, The Elder Scrolls: Skyrim, Call of Duty: Modern Warfare 2, Left 4 Dead.…”
Section: Mikro Szint éS Sportfogyasztás Micro Level and Sport Consuptionunclassified
“…ML methods are capable of identifying complex non-linear relationships in large amounts of data. ML approaches are subdivided in three learning categories: supervised, semi-supervised and unsupervised [27]. Supervised approaches consist of two steps for training: In the first step, the data must be labeled as, e.g., "acceptable" and "not acceptable".…”
Section: Introductionmentioning
confidence: 99%
“…This means before learning, the user must know and identify the defects and label them. An unsupervised approach is used without labeling of the data [27,28]. An algorithm tries to identify defects itself.…”
Section: Introductionmentioning
confidence: 99%