Fields of Interest
Infrastructure Engines
Fields of Interest
Infrastructure Engines
Correlation Systems > Technology & Innovation > Geographic Data Mining
Geographic Data Mining
In an era when knowledge is power and technologies are improving rapidly, information technology
(IT) is widespread and can be found in
almost every aspect of our lives - business, science, homeland
security and more.
Because of the enormous volume of data available to us (from varying
sources such as: satellite imagery, medical equipments, video
cameras and tracking systems), it is costly and often unrealistic to
examine the data in detail, resulting in the loss of essential
information. Moreover, in many cases, the obtained data is
characterized by spatial and temporal relationships between
entities, making it harder to comprehend and efficiently analyze.
Being an early adopter of data mining techniques, Correlation Systems advanced and developed the Geographic Data Mining (GDM) technology, which aims to automate the knowledge discovery process in data sets which possess spatial and temporal characteristics.
Correlation Systems' GDM tools enable the extraction of interesting spatial patterns and features as well as the discovery of intrinsic relationships between spatial and non-spatial data.
The GDM technology is applicable in many fields, such as environmental control and protection; efficiency and quality of service monitoring in public transportation; intelligent security systems; defense & home land security, and more.
Technology Overview
Based on traditional data mining techniques, our GDM tools utilize a combination of statistical analyses, modeling techniques, and database technologies in order to find patterns and subtle relationships in large data volumes and infer rules that allow the prediction of future results.
Meeting the special needs of spatial and temporal data, Correlation Systems has developed the following algorithms and techniques:
Spatio-temporal relationships between entities are detected by applying proprietary clustering algorithms, robust to information gaps and ambiguous data sets.
Machine learning techniques are used for improving the GDM's performance. Analyzing the GDM products and the effect they have on the overall system (including the user reactions) enables to calibrate the existing processes and generate automated models for discovering new patterns and relationships.
The GDM tools are adapted for both central and distributed systems.
When dealing with heterogeneous data sources Data Fusion techniques are applied in order to reformat the data and apply the semantics used by the GDM tools.
