Internet-wide scanning is a commonly used research technique in various network surveys, such as measuring service deployment and security vulnerabilities. However, these network surveys are limited to the given port set, not comprehensively obtaining the real network landscape, and even misleading survey conclusions. In this work, we introduce PMap, a port scanning tool that efficiently discovers the most open ports from all 65K ports in the whole network. PMapuses the correlation of ports to build an open port correlation graph of each network, using a reinforcement learning framework to update the correlation graph based on feedback results and dynamically adjust the order of port scanning. Compared to current port scanning methods, PMapperforms better on hit rate, coverage, and intrusiveness. Our experiments over real networks show that PMapcan find 90% open ports by only scanning 125 ports (90%@125) to each address, which is 99.3% less than the state-of-the-art port scanning methods. It reduces the number of scanned ports to decrease the intrusive nature of port scanning. In addition, PMapis highly parallel and lightweight. It scans 500 networks in parallel, achieving a port recommendation rate of up to 18 million per second, consuming only 7GB of memory. PMapis the first effective practice for scanning open ports using reinforcement learning. It bridges the gap of existing scanning tools and effectively supports subsequent service discovery and security research.
Anomaly detection for network attacks has always been a very important part of intrusion detection. The current research focus is anomaly detection based on deep learning, which has two main problems. One is the lack of a large amount of labeled data in model training, and the other is difficult to detect unknown network attacks or variant attacks. To solve the above problems, an unsupervised anomaly detection model is constructed in this paper. The automatic encoder is used to learn normal traffic characteristics and detect abnormal traffic. Meanwhile, time correlation features and hierarchical clustering algorithm are used for data preprocessing to reduce time and space complexity, so as to further improve the efficiency of model detection. Due to the serious lack of verification data sets for unsupervised anomaly detection, this paper collects and organizes a large amount of data and designs four types of network attack data, including new attack means, worms, system vulnerabilities and botnets. The experimental results showed that the detection accuracy of worms and system vulnerabilities reached 98%, the detection accuracy of botnets reached 89%, and the attacks of the new OriginLogger software were detected.
Log parsing converts raw system logs into structured templates and is a key preprocessing step for Artificial Intelligence for IT Operations (AIOps). Existing parsers face a practical trade-off: rule-based methods offer high throughput but limited adaptability across heterogeneous log sources, whereas Large Language Model (LLM)-based parsers achieve broader semantic coverage at the cost of inference latency, privacy exposure, and cloud dependency. This paper presents FlanBC, a log parsing framework that formulates template extraction as a BIO (Beginning, Inside, Outside) sequence-labeling task and integrates a Flan-T5 semantic encoder, Bidirectional Long Short-Term Memory (BiLSTM) layers for local sequential modeling, and a Conditional Random Field (CRF) decoder for structured label prediction. Log-specific preprocessing and a subword-to-token alignment mechanism adapt the general-purpose encoder to semi-structured log data. A layer-freezing strategy reduces the number of parameters updated during training. The framework supports local inference without external API dependency. Experiments on three benchmark datasets from LogHub (HDFS, BGL, OpenStack) under a supervised random-split setup evaluate parsing accuracy, training efficiency, statistical stability across random seeds, and component contributions. FlanBC achieves a Group Accuracy of 99.32% on HDFS and 98.47% on BGL, with an inference throughput of 700+ logs/s on a consumer-grade GPU. On OpenStack, performance is lower (GA = 92.54%), reflecting the challenge that diverse natural-language-like logs pose for compact encoder-based models. Under a stricter template-disjoint split that prevents template overlap between training and test sets, FlanBC achieves an average Group Accuracy of 91.14%, indicating that the model generalizes to unseen templates beyond in-distribution recognition. Ablation results indicate that the semantic encoder, BiLSTM module, and CRF decoder each contribute to prediction accuracy. These findings suggest that domain-adapted semantic encoders combined with structured decoding offer a practical accuracy–efficiency balance for log parsing in settings where local, cloud-free inference is preferred.