• Corpus ID: 195791464

E-Sports Talent Scouting Based on Multimodal Twitch Stream Data

  title={E-Sports Talent Scouting Based on Multimodal Twitch Stream Data},
  author={Anna Belova and Wen He and Ziyi Zhong},
We propose and investigate feasibility of a novel task that consists in finding e-sports talent using multimodal Twitch chat and video stream data. In that, we focus on predicting the ranks of Counter-Strike: Global Offensive (CS:GO) gamers who broadcast their games on Twitch. During January 2019-April 2019, we have built two Twitch stream collections: One for 425 publicly ranked CS:GO gamers and one for 9,928 unranked CS:GO gamers. We extract neural features from video, audio and text chat… 

Figures and Tables from this paper

The ingredients of Twitch streaming: Affordances of game streams

Behind the game: Exploring the twitch streaming platform

It is found that Twitch is very different to existing video platforms, with a small number of games consistently achieving phenomenal dominance, and a complex game ecosystem combining consistently popular games over years, newly released games enjoying bursts of popularity, and even old games appearing on the platform.

Chat Speed OP PogChamp: Practices of Coherence in Massive Twitch Chat

This work examines communication practices in Twitch chats for the popular game Hearthstone, comparing massive chats with at least 10,000 concurrent viewers and small chats with fewer than 2,000 concurrently viewers.

Interactions of Twitch Users and Their Usage Behavior

The usage behavior and interactions of Twitch users, especially their wish to get involved in a live-stream, are analyzed, to get an insight of the interactions between Twitch users.

Watch Me Play

A look at the revolution in game live streaming and esports broadcasting Every day thousands of people broadcast their gaming live to audiences over the internet using popular sites such as Twitch,

Video Highlight Prediction Using Audience Chat Reactions

A novel dataset based on League of Legends championships recorded from North American and Taiwanese Twitch.tv channels is presented, and strong results on these using multimodal, character-level CNN-RNN model architectures are demonstrated.

Design Challenges for Livestreamed Audience Participation Games

To explore the game design space of APGs, three game development teams were provided with an audience participation interface development toolkit and six challenges were identified: latency, screen sharing, attention management, player agency, audience-streamer relationships, and shifting schedules.

Seeker: Topic-Aware Viewing Pattern Prediction in Crowdsourced Interactive Live Streaming

This paper closely examines the characteristics of interactive messages in the real-world datasets and designs a crowdsourced framework Seeker, which can predict aforementioned patterns in advance and achieve much higher performance than the time-series approaches.

Gendered Conversation in a Social Game-Streaming Platform

It is shown that gendered conversation and objectification is prevalent in chats, and disparities in social game-streaming platforms is a nuanced phenomenon that involves the gender of streamers as well as those who produce gendered and game-related conversation.

Social Talent Scouting: A New Opportunity for the Identification of Football Players?

Abstract This paper investigates the diffusion of digital technologies within the football talent scouting process. A qualitative exploration based on open discussions and unstructured interviews