From zero-shot machine learning to zero-day attack detection

Machine learning (ML) models have proved efficient in classifying data samples into their respective categories. The standard ML evaluation methodology assumes that test data samples are derived from pre-observed classes used in the training phase. However, in applications such as Network Intrusion Detection Systems (NIDSs), obtaining data samples of all attack classes to be observed is challenging. ML-based NIDSs face new attack traffic known as zero-day attacks that are not used in training due to their non-existence at the time. Therefore, this paper proposes a novel zero-shot learning methodology to evaluate the performance of ML-based NIDSs in recognising zero-day attack scenarios. In the attribute learning stage, the learning models map network data features to semantic attributes that distinguish between known attacks and benign behaviour. In the inference stage, the models construct the relationships between known and zero-day attacks to detect them as malicious. A new evaluation metric is defined as Zero-day Detection Rate (Z-DR) to measure the effectiveness of the learning model in detecting unknown attacks. The proposed framework is evaluated using two key ML models and two modern NIDS data sets. The results demonstrate that for certain zero-day attack groups discovered in this paper, ML-based NIDSs are ineffective in detecting them as malicious. Further analysis shows that attacks with a low Z-DR have a significantly distinct feature distribution and a higher Wasserstein Distance range than the other attack classes.

     

Fuzzy Risk Assessment of a Stormwater Storage System in a Poorly Gauged Mine Site: The Case of the Golgohar Mine Site

Mine Water and the Environment - Although many approaches have been proposed for flood risk assessment in urban areas, an equivalent methodology for remote mine sites with poor hydrological data...     

Sustainable functionalization and modification of materials via multicomponent reactions in water

In materials chemistry, green chemistry has established firm ground providing essential design criteria to develop advanced tools for efficient functionalization and modification of materials. Particularly, the combination of multicomponent reactions in water and aqueous media with materials chemistry unlocks a new sustainable way for constructing multi-functionalized structures with unique features, playing significant roles in the plethora of applications. Multicomponent reactions have received significant consideration from the community of material chemistry because of their great efficiency, simple operations, intrinsic molecular diversity, and an atom and a pot economy. Also, by rational design of multicomponent reactions in water and aqueous media, the performance of some multicomponent reactions could be enhanced by the contributing “natural” form of water-soluble materials, the exclusive solvating features of water, and simple separating and recovering materials. To date, there is no exclusive review to report the sustainable functionalization and modification of materials in water. This critical review highlights the utility of various kinds of multicomponent reactions in water and aqueous media as green methods for functionalization and modification of siliceous, magnetic, and carbonaceous materials, oligosaccharides, polysaccharides, peptides, proteins, and synthetic polymers. The detailed discussion of synthetic procedures, properties, and related applicability of each functionalized/modified material is fully deliberated in this review.