Temel Kavramlar
Large Language Models have opened opportunities for designing chatbots to enhance news engagement, but understanding user perspectives is crucial.
Özet
This content explores the design of question-answering chatbots for online news. It includes interviews with journalists and an online experiment with readers to understand their interactions with chatbots. The study reveals differences in how users engage with authors and chatbots, highlighting the importance of aligning chatbot functionalities with user expectations.
Introduction
Journalists face challenges in engaging with readers.
Chatbots offer a potential solution for improving communication.
Study with Journalists
Journalists communicate mainly through email and social media.
Challenges include maintaining discussions at scale.
Study with Audience/Readers
Readers ask more factual questions to chatbots than authors.
Questions to authors are more complex and emotionally charged.
Taxonomy of Questions
Categories include Information, Interpretation, and Others.
Authors vs Chatbot
Participants ask more detailed questions to authors than chatbots.
Authors receive more critical questions about journalistic integrity.
Effect of Perceived Quality and Personal Preferences
High-quality articles lead to more detailed questions to authors.
Low-quality articles prompt questioning of journalistic integrity.
Framework for Designing Chatbots
Considerations include functionalities desired by authors and readers, newsroom preferences, current technology capabilities, AI/technology policies, and regulatory policies.
İstatistikler
Large Language Models (LLMs) have created opportunities for designing chatbots that can support complex question-answering (QA) scenarios.
An online experiment was conducted with 124 participants to understand how readers want to interact with a QA chatbot.
Alıntılar
"A paradigm shift away from a ‘lecturing’ approach to a ‘dialogue’ approach is a key factor for journalism in a post-truth era." - Meier et al.