Abstract: A method includes receiving input data including unstructured text representing one or more sequences of terms. For each respective sequence of terms, the method includes generating a corresponding line of therapy (LoT) pseudo-label indicating whether the respective sequence of terms includes LoT information, generating a corresponding LoT indicator predicting whether the respective sequence of terms includes LoT information, and determining a corresponding LoT indication loss based on the corresponding LoT pseudo-label and the corresponding LoT indicator. The method also includes fine-tuning a pre-trained transformer model based on the LoT indication losses determined for the one or more sequences of terms.

Abstract: Embodiments of the present application relate to methods, devices, apparatuses, and storage media for monitoring and analyzing air pollution. The methods includes acquiring air quality data of a monitored area, where the air quality data includes ground air quality data corresponding to the monitored area and N-component data collected by an air sensor in the monitored area; inputting the air quality data into a pollution analysis model, where the pollution analysis model is previously trained according to the ground air quality data and the N-component data; and determining whether the air quality data of the monitored area is abnormal according to an output result of the pollution analysis model. The methods may solve the technical problem of inaccurate monitoring results caused by the susceptibility of air quality monitoring in related technologies to human intervention.