The application discloses a growth file dynamic recording and
tracking system for children with development disorders, relates to the technical field of children development disorder assessment and
rehabilitation, and comprises a
data acquisition module, a
feature extraction module, a
feature fusion module, a semantic projection module and a semantic output module; the
system collects behavior text records, audio and video streams, course types,
noise spectra, interactive objects and other environmental
context data, extracts behavior feature vectors and environmental feature vectors, and generates environmental-behavior fusion vectors by using a cross attention mechanism; the fusion vectors are projected to a
semantic space to calculate
semantic similarity, and the result confidence is judged according to the difference value between the highest and the second highest similarity; if the result confidence is insufficient, environmental features are re-extracted based on an
extended time window and are weighted and fused to generate environmental feature reinforcement vectors, and then fusion, semantic projection and semantic category output are performed again; the application can ensure that the growth file maintains context consistency and development coherence in long-term tracking.