Yuta Hoshi

Japan Broadcasting Corporation (Japan)

Papers

4

Total Citations

13

H-Index

2

About

Yuta Hoshi is a human-computer interaction researcher whose work centers on social robotics, companion systems, and the evolving dynamics of media consumption. His research addresses a quietly significant modern problem: as streaming culture fragments the shared television-viewing experience, opportunities for spontaneous human connection are diminishing. Hoshi's primary contribution has been the design and development of TV-watching companion robots — systems capable of generating contextually appropriate utterances to simulate the warmth and engagement of watching television alongside another person. His most cited work, "TV-Watching Robot: Toward Enriching Media Experience and Activating Human Communication" (2022, 5 citations), exemplifies his approach of grounding robot behavior in careful analysis of real human dialogue. By systematically classifying the utterances exchanged between pairs of TV viewers — identifying eight distinct functional categories — Hoshi has established empirically informed design guidelines for conversational robots. His integration of open-domain chatbot technology, notably the KACTUS system, further demonstrates his commitment to building practically deployable companions rather than purely theoretical models. Though early in his citation trajectory, Hoshi's research sits at a compelling intersection of loneliness, technology, and everyday communication — making his work increasingly relevant as solo media consumption continues to rise globally.

Research Focus

Key Achievements

2
H-Index
4
Papers
13
Total Citations
3
Avg Citations/Paper
🏆 Most Cited Paper
TV-Watching Robot: Toward Enriching Media Experience and Activating Human Communication
5 citations · 2022
📈 Most Prolific Year: 2022 (2 Papers)
🤝 Key Collaborators: 16
🏛 Institutions: Japan Broadcasting Corporation (Japan)

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago