Research pick: Paper late, getting the news of the world - "Optimising news dissemination pathways in the media convergence era: an interactive digital media technology approach"

Research in the International Journal of Information and Communication Technology discusses a new framework that could be used to define how news and information are delivered. The approach could help us overcome some of the associated problems of media saturation and information overload. The approach allows for real-time, adaptive decisions about the flow of information in an age of media convergence as traditional journalism, digital platforms, and interactive technologies become increasingly intertwined.

The new approach side-steps the static, off-the-shelf distribution methods of conventional media and develops a system that can adjust output based on audience behaviour dynamically, tailoring content presented as user one’s interests change.

The approach integrates multimodal perception, the ability to interpret data from multiple sources such as text, images, and user interaction patterns, with reinforcement learning, a branch of artificial intelligence in which systems learn by trial-and-error guided by feedback. This combination allows the system to detect and respond to the gradual change in what an individual finds engaging over time, a concept known in psychology interest drift.

The researchers explain that their system works on three interconnected levels. First, it uses temporal attention mechanisms, which monitor how an individual’s focus changes over time, along with deep feature extraction to identify subtle behavioural patterns. This allows the system to anticipate shifts in audience engagement and adjust accordingly.

Second, it employs a hierarchical reinforcement learning architecture. In this design, complex decision-making is broken into layers, making it easier to balance competing objectives. Using a blend of deep learning methods and evolutionary algorithms, which mimic natural selection to find optimal solutions, the system maximises audience reach, ensures timely delivery, and minimises computational and network resource use.

The third layer introduces an adaptive regulation process using mathematical optimisation techniques. This component fine-tunes the balance between performance and resource consumption in real time, enabling the system to remain efficient even under fluctuating conditions.

The implications for journalism in areas where timeliness and accuracy are essential, such as public health, politics, and in emergencies and crises could be enormous. An adaptive, responsive delivery model should improve audience engagement and comprehension, the work suggests.

Zhang, Y., Liu, Y. and Guo, Z. (2025) ‘Optimising news dissemination pathways in the media convergence era: an interactive digital media technology approach’, Int. J. Information and Communication Technology, Vol. 26, No. 29, pp.110–126.