Abstract: A method for dialogue processing, an electronic device and a storage medium are provided. The specific technical solution includes: obtaining a dialogue history; selecting a target machine from a plurality of machines; inputting the dialogue history into a trained dialogue model in the target machine to generate a response to the dialogue history, in which the dialogue model comprises a common parameter and a specific parameter, and different machines correspond to the same common parameter.

Abstract: The disclosure discloses a method for generating a conversation, an electronic device, and a storage medium. The detailed implementation includes: obtaining a current conversation and historical conversations of the current conversation; selecting multiple reference historical conversations from the historical conversations and adding the multiple reference historical conversations to a temporary conversation set; and generating reply information of the current conversation based on the current conversation and the temporary conversation set.

Abstract: Disclosed is a full-section anti-slide water-cut wall for treating giant landslides, which comprises at least one water-cut wall embedded in the landslide and arranged along the direction of blocking the movement of a sliding body; the water-cut wall comprises a plurality of anti-slide piles arranged at intervals, the bottom end of the anti-slide piles extend into the sliding bed, and a blocking wall is fixedly connected between adjacent anti-slide piles; the anti-slide pile and the blocking wall are provided with a plurality of drainage holes.

Abstract: A computer-implemented framework and/or system for cyberbullying detection is disclosed. The system includes two main components: (1) A representation learning network that encodes the social media session by exploiting multi-modal features, e.g., text, network, and time; and (2) a multi-task learning network that simultaneously fits the comment inter-arrival times and estimates the bullying likelihood based on a Gaussian Mixture Model. The system jointly optimizes the parameters of both components to overcome the shortcomings of decoupled training. The system includes an unsupervised cyberbullying detection model that not only experimentally outperforms the state-of-the-art unsupervised models, but also achieves competitive performance compared to supervised models.

Abstract: The present disclosure discloses use of a combination of an HIP-1? inhibitor and/or an AR inhibitor in inhibiting an .inflammatory cytokine storm,. treating or preventing viral pneumonia. The present disclosure finds that inhibitors HIF-1 a and AR prevent the transcription of inflammatory cytokine storm-related genes (IL6, MMP2, MMP13, ADAMTS4, ELN, VCAN. COL3A1 and VEGFA) in fibroblasts under low-oxygen environment by effectively inhibiting the interaction between HIF-1? and AR to reduce the expression quantity of the related genes, thereby influencing the activation of lung fibroblasts. Inhibition of activation of lung fibroblasts enables patients with acute respiratory distress syndrome (ARDS) due to severe respiratory tract infection to maintain the lung functions and improve clinical efficacy.

Abstract: A computer-implemented framework and/or system for cyberbullying detection is disclosed. The system includes two main components: (1) A representation learning network that encodes the social media session by exploiting multi-modal features, e.g., text, network, and time; and (2) a multi-task learning network that simultaneously fits the comment inter-arrival times and estimates the bullying likelihood based on a Gaussian Mixture Model. The system jointly optimizes the parameters of both components to overcome the shortcomings of decoupled training. The system includes an unsupervised cyberbullying detection model that not only experimentally outperforms the state-of-the-art unsupervised models, but also achieves competitive performance compared to supervised models.

Abstract: The present invention discloses an erasure code calculation method, including the following steps: S1) splitting original data, and building an original encoding matrix M; S2) acquiring a transverse exclusive OR encoding matrix M1; S3) acquiring a longitudinal exclusive OR encoding matrix M2; S4) acquiring an exclusive OR encoding matrix M3 according to the transverse exclusive OR encoding matrix M1 and the longitudinal exclusive OR encoding matrix M2; S5) transforming a data position of the transverse exclusive OR encoding matrix M1 to acquire a storage matrix M4; S6) judging whether storage nodes at which the last column of data of the storage matrix M4 is stored are damaged; S7) restoring the lost data according to a position 1 of the damaged node; and S8) restoring the lost data according to a position 2 of the damaged node. In the present invention, the operation is rapid, and calculation efficiency is high.

Abstract: The present invention discloses an erasure code calculation method, including the following steps: S1) splitting original data, and building an original encoding matrix M; S2) acquiring a transverse exclusive OR encoding matrix M1; S3) acquiring a longitudinal exclusive OR encoding matrix M2; S4) acquiring an exclusive OR encoding matrix M3 according to the transverse exclusive OR encoding matrix M1 and the longitudinal exclusive OR encoding matrix M2; S5) transforming a data position of the transverse exclusive OR encoding matrix M1 to acquire a storage matrix M4; S6) judging whether storage nodes at which the last column of data of the storage matrix M4 is stored are damaged; S7) restoring the lost data according to a position 1 of the damaged node; and S8) restoring the lost data according to a position 2 of the damaged node. In the present invention, the operation is rapid, and calculation efficiency is high.