Grant: Discovery Launch Supplement
McMaster University
$12,500.00
Human Centered Design of Artificial Intelligence Systems for Computational Journalism
EF 2024-25
Résumé
Information-producing professionals like journalists and data story writers today rely on complex public data sources to find important stories on topics like climate change, world resource consumption, and the impact of economic reforms. This process of data-driven storytelling (DDS), is important for communicating complex issues of public interest. Generating these stories, however, involves a series of cognitively demanding tasks. The writer first analyzes and extracts relevant information from large data sources like social media platforms, and digitized public archives. Next, they carefully structure the narrative of the story that will communicate their findings. Finally, they articulate their thoughts using written words and designing effective visualizations. These tasks are also often resource-intensive (requiring data expertise and large teams to accomplish) and not many news organizations can afford to find and report high-quality data stories with speed and efficiency. Recent advances in Artificial Intelligence (AI) (e.g. ChatGPT) can support these cognitively demanding tasks, by automatically identifying and analyzing important trends to report, creating visualizations from raw data files, and rapidly generating written text. Automating these DDS tasks can help to significantly accelerate the speed, volume, and breadth of their daily production. However, algorithmically generated content is susceptible to factual and reasoning errors, biased narratives, and toxic arguments. When such stories infiltrate our news feeds, they risk trapping readers in misinformation bubbles, intolerant viewpoints, and biased opinions that pose a threat to information infrastructures and society. In this research program, I will investigate the design and development of AI-based writing technologies that can improve the speed of the writing process, while improving the efficiency of the writer, and the quality of the writing outcome.
This research program will focus on building human-centered AI systems that 1) understand a writer's individual goals, needs, and preferences, 2) support the writer in producing high-quality content with speed and ease, and, 3) promote collaboration and iterative feedback. These research objectives will be achieved by firstly, building techniques to measure the impact of AI systems on writing efficiency. Simultaneously I will investigate interactive AI tools that can improve the quality of writing outcomes. Finally, I will leverage AI provenance information to design writing technologies that promote rich collaborative feedback. The proposed research program contributes to the field of human-centered machine learning by designing human-centered methodologies for building ML-powered systems. While this research program focuses on information production tools for journalistic applications, our findings will apply to a wide range of human-AI interaction design problems for complex decision-making tasks.
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
DGECR-2024-00020
Name nom
Mishra, Swati
Department département
Computing and Software
Organization id
27
Institution établissement
McMaster University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
2709
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Other computing methods
Research subject fr
Autres méthodes informatiques
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Human Centered Design of Artificial Intelligence Systems for Computational Journalism
Keywords
interactive machine learning; human computer interaction; user modelling; ai based writing assistants; interactive interfaces; human centered design; intelligent user interfaces; computer supported collaborative work; user interface design; human centered machine learning
Application summary
Information-producing professionals like journalists and data story writers today rely on complex public data sources to find important stories on topics like climate change, world resource consumption, and the impact of economic reforms. This process of data-driven storytelling (DDS), is important for communicating complex issues of public interest. Generating these stories, however, involves a series of cognitively demanding tasks. The writer first analyzes and extracts relevant information from large data sources like social media platforms, and digitized public archives. Next, they carefully structure the narrative of the story that will communicate their findings. Finally, they articulate their thoughts using written words and designing effective visualizations. These tasks are also often resource-intensive (requiring data expertise and large teams to accomplish) and not many news organizations can afford to find and report high-quality data stories with speed and efficiency. Recent advances in Artificial Intelligence (AI) (e.g. ChatGPT) can support these cognitively demanding tasks, by automatically identifying and analyzing important trends to report, creating visualizations from raw data files, and rapidly generating written text. Automating these DDS tasks can help to significantly accelerate the speed, volume, and breadth of their daily production. However, algorithmically generated content is susceptible to factual and reasoning errors, biased narratives, and toxic arguments. When such stories infiltrate our news feeds, they risk trapping readers in misinformation bubbles, intolerant viewpoints, and biased opinions that pose a threat to information infrastructures and society. In this research program, I will investigate the design and development of AI-based writing technologies that can improve the speed of the writing process, while improving the efficiency of the writer, and the quality of the writing outcome.
This research program will focus on building human-centered AI systems that 1) understand a writer's individual goals, needs, and preferences, 2) support the writer in producing high-quality content with speed and ease, and, 3) promote collaboration and iterative feedback. These research objectives will be achieved by firstly, building techniques to measure the impact of AI systems on writing efficiency. Simultaneously I will investigate interactive AI tools that can improve the quality of writing outcomes. Finally, I will leverage AI provenance information to design writing technologies that promote rich collaborative feedback. The proposed research program contributes to the field of human-centered machine learning by designing human-centered methodologies for building ML-powered systems. While this research program focuses on information production tools for journalistic applications, our findings will apply to a wide range of human-AI interaction design problems for complex decision-making tasks.