Overview / Usage

With the advent of big data era, knowledge engineering has attracted wide attention, as mining knowledge from large-scale data is critical for big data analysis. Knowledge graph techniques provide a way to extract structured knowledge from large-scale texts and images, thus have wide application prospect.
Nowadays Knowledge graph techniques has already been used in many areas like Intelligent search, Character diagram building, Fraud detection and so on.
In data mining, anomaly detection (also outlier detection) is the identification of items, events or observations which do not conform to an expected pattern or other items in a dataset. Typically the anomalous items will translate to some kind of problem such as bank fraud, a structural defect, medical problems or errors in a text.In particular, in the context of abuse and network intrusion detection, the interesting objects are often not rare objects, but unexpected bursts in activity.
This pattern does not adhere to the common statistical definition of an outlier as a rare object, and many outlier detection methods (in particular unsupervised methods) will fail on such data, unless it has been aggregated appropriately.
Generally, Anomaly Dectection algorithms always based on linear models, proximity-based models, information-theoretic models and so on.
However, if we regards the scenario of anomaly detection as a network or graph, then the normal samples and outlier samples will be both a node in the graph. We can guess, the relationships and attribute of the two kinds of nodes will be somehow different, so we can use graph algorithm or knowledge graphs to find out the outlier samples.
This is the core of the project.
Now, it's still a cencept, and I hope I can try to make some experiments on this topic with Intel Technologies.
If you have any idea about this project, please feel free to contact me.

Technologies Used

Knowledge Graph

Other links

• My Blog Address