To efficiently mine threat intelligence from the vast array of open-source cybersecurity analysis reports on the web, we have developed the Parallel Deep Forest-based Multi-Label Classification (PDFMLC) algorithm. Initially, open-source cybersecurity analysis reports are collected and converted into a standardized text format. Subsequently, five tactics category labels are annotated, creating a multi-label dataset for tactics classification. Addressing the limitations of low execution efficiency and scalability in the sequential deep forest algorithm, our PDFMLC algorithm employs broadcast variables and the Lempel-Ziv-Welch (LZW) algorithm, significantly enhancing its acceleration ratio. Furthermore, our proposed PDFMLC algorithm incorporates label mutual information from the established dataset as input features. This captures latent label associations, significantly improving classification accuracy. Finally, we present the PDFMLC-based Threat Intelligence Mining (PDFMLC-TIM) method. Experimental results demonstrate that the PDFMLC algorithm exhibits exceptional node scalability and execution efficiency. Simultaneously, the PDFMLC-TIM method proficiently conducts text classification on cybersecurity analysis reports, extracting tactics entities to construct comprehensive threat intelligence. As a result, successfully formatted STIX2.1 threat intelligence is established.

For dynamic application scenarios of Mobile Edge Computing (MEC), an Energy-efficient Multiuser and Multitask Computation Offloading (EMMCO) optimization method is proposed. Under the consideration of multiuser and multitask computation offloading, first, the EMMCO method takes into account the existence of dependencies among different tasks within an implementation, abstracts these dependencies as a Directed Acyclic Graph (DAG), and models the computation offloading problem as a Markov decision process. Subsequently, the task embedding sequence in the DAG is fed to the RNN encoder-decoder neural network with combination of the attention mechanism, the long-term dependencies among different tasks are successfully captured by this scheme. Finally, the Improved Policy Loss Clip-based PPO2 (IPLC-PPO2) algorithm is developed, and the RNN encoder-decoder neural network is trained by the developed algorithm. The loss function in the IPLC-PPO2 algorithm is utilized as a preference for the training process, and the neural network parameters are continuously updated to select the optimal offloading scheduling decisions. Simulation results demonstrate that the proposed EMMCO method can achieve lower latency, reduce energy consumption, and obtain a significant improvement in the Quality of Service (QoS) than the compared algorithms under different situations of mobile edge network.