Web applications are popular targets for cyber-attacks because they are network-accessible and often contain vulnerabilities. An intrusion detection system monitors web applications and issues alerts when an attack attempt is detected. Existing implementations of intrusion detection systems usually extract features from network packets or string characteristics of input that are manually selected as relevant to attack analysis. Manually selecting features, however, is time-consuming and requires in-depth security domain knowledge. Moreover, large amounts of labeled legitimate and attack request data are needed by supervised learning algorithms to classify normal and abnormal behaviors, which is often expensive and impractical to obtain for production web applications. This paper provides three contributions to the study of autonomic intrusion detection systems. First, we evaluate the feasibility of an unsupervised/semi-supervised approach for web attack detection based on the Robust Software Modeling Tool (RSMT), which autonomically monitors and characterizes the runtime behavior of web applications. Second, we describe how RSMT trains a stacked denoising autoencoder to encode and reconstruct the call graph for end-to-end deep learning, where a low-dimensional representation of the raw features with unlabeled request data is used to recognize anomalies by computing the reconstruction error of the request data. Third, we analyze the results of empirically testing RSMT on both synthetic datasets and production applications with intentional vulnerabilities. Our results show that the proposed approach can efficiently and accurately detect attacks, including SQL injection, cross-site scripting, and deserialization, with minimal domain knowledge and little labeled training data.

The Clarujust optimization method of programming CIs shows promise to improve patient performance and increase patient satisfaction in a shorter clinical test time.

No abstract

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. AbstractModern elements of military intelligence and decision making require predictions of adversary force actions and reactions to provide a complete and realistic viewpoint. Current methods for providing realistic adversary force simulation are largely manual processes. Adversarial simulation requires continual assessment of friendly courses of action and is currently "human assessment capability" limited. To develop a computational model of dynamic adversary behaviors that includes the ability to integrate with intelligence and mission data sources, computational models must address operational patterns, behaviors, or doctrines of present-day adversaries (terrorist cells, local insurgents, guerillas, and armed thugs) as well as more conventional force elements. The dynamic nature of adversary force behavior with respect to the changing capabilities, biases, beliefs, goals, intentions, and perceptions of friendly force actions must be addressed. The Emergent Adversarial Modeling System (EAMS) addresses these elements through explicit focus on adversarial intent as a driver for adversarial response. Specific capabilities address the changing nature of adversary composition. This paper will discuss the results of the ongoing EAMS research project into adversarial modeling and adversarial response simulation.

No abstract