Adaptive recommendation technology for remote sensing information based on behavior analysis
Caiping Li, Lei Chang, Jisheng Zhang, Xiaoming Zhou, Zhen Yu Hu, Yuanchen Song · 2019
In view of the changing interests of users of remote sensing information, this paper proposes a self-adaptive recommendation technology for remote sensing information based on behavior analysis. Through real-time collection and feedback of user's browsing, querying and downloading behavior data on the application platform, users can find new concerns and construct attenuation functions. Periodically, users' preferences are revised. The adaptive learning strategy adds users' new preferences to the key information structure, which strengthens the original information preferences. This technology not only takes full account of users' historical interest points, but also dynamically discovers users' new interests, and integrates them reasonably according to interest attenuation model, so as to improve the accuracy of recommendation.