Secure Data Analytics
The Industrial Internet of Things revolution brings a promise of using big amount of data for a better decision making and data analytics processing like AI, in order to achieve higher productivity and better business performance. While using BigData for AI has already proved to have benefits in other market segments, its potential is still somewhat ‘artificially limited’ in segments like the manufacturing industry.
Indusify’s platform applies data mining techniques to extract the value concealed in industry BigData while keeping data privacy, using a unique technology for ensuring a safe collaborated AI on the cloud.
Indusify’s platform collects data on customers facility or from customer’s cloud (in case it exists), masks it, so customer data is kept anonymous and private, and safely sends the data to a centralized location for processing while keeping customer’s data masked during the entire data processing phase. At this centralized location, an AI training process uses the data from all sources while it is kept in a protected masked format, to build a global model. This global model is much more accurate allowing all parties to benefit from the experience of each other, resulting in better insights and greater ROI.
The platform does not intend to replace any encryption of the data while in transit or any credential-based access control to access cloud storage that store the data. Rather, the platform adds an ultimate protection layer by keeping the data in a masked form at all times, making it void of any context, making it impossible to link to any specific organization, facility or industry.
Indusify’s patent-pending masking technology, allows sending organization’s data in an abstract form, stripped from its private data and context, thus making it robust to eavesdropping while in transit and also tolerant to an event in which data gets to the ‘wrong’ hands while stored in the cloud.
Our technology applies special masking algorithms to create a unique abstract form of the data
Anomaly detection (aka Outlier detection) is the identification of rare items, events or observations which raise suspicions by differing significantly from the majority of the data.
Typically, anomalous data can be related to some kind of problem or rare event such as bank fraud, structural defects, malfunctioning equipment etc. These relations (with significant events) makes it very important to be able to pick out which data points can be considered as anomalies, as identifying these events are typically very interesting from a business perspective. That brings us to one of the key objectives: How do we identify whether a data point is normal or anomalous? In some simple cases, as in the example below, data visualization can give us important information. Anomaly detection (aka Outlier detection) is the identification of rare items, events or observations which raise suspicions by differing significantly from the majority of the data.
IIOT software services
In addition to the platform and the solution that Indusify provides in the field of safe data collaborative analysis, the company also provides complementary professional software services in these fields. The software services may be combined with the company’s products or as a pure service to support and extend customer’s own IIOT platform.