Advanced data platform & AI solutions for anomaly detection

Advanced AI Behavior & Anomaly Modeling Ecosystem

AI Edge systems are industry, machine, system and operation agnostic. They are solely data driven, and consume clients side data to build and train custom AI models for their operations.
Our API friendly ecosystem offers a powerful data platform which automated the integration, formatting and processing of big data to deep data and intelligent features for AI processing.

Flexible AI Solutions as a SaaS

We provide the platform you need to creates value from your enterprise data: automation of data pipeline, data transformation and feature engineering, and vast array of AI algorithms.
A monthly cloud AI platform as a SaaS. Our data engineers and data scientists will work with you to configure the AI model of interest to your enterprise.

Sys Edge Data Hub

Operational, sensor, financial, transaction and supply chain data have disparate timestamps, formats, and types. They are too large in size, lack indexing and structuring for feature extraction in AI modeling as well as analytics and reporting.

Developing and integrating a custom enterprise data hub or warehouse is a significant investment for a company depending on the size and complexity of the data and the intended analytics and reporting required thereafter. Once the data hub is up and running, a sizable expert team is required to maintain the platform and provide support.

We propose analternative approach. Sys Edge Data Hub is a modern flexible data platform with 6 integrated microservices ready out-of-the-box, automating the entire data pipeline. They can be configured on the fly to automatically create a custom data hub for any operation.

Data Object Creation

Creation and management of entities, hierarchies of placeholder for data. For example, company, subdivisions, nested entities and their inner relationships, each leading to their own data type. These automatically create the objects that manage the Data Hub.

Data Model Creation

Configure on screen, the data attributes, e.g., timescale, label, types and subtypes, etc., for each one of the data types associated with the entities previously created. This automatically creates a database table for each data type, with AP endpoint connectors.

Data Fetching

For each type of data and source, e.g., API, web-scraping, FTP, csv, multi-tab Excel, configure the sequence of data fetching, mapping, transformation and insertion to our data model. These automatically fetch and insert the data in the Data Hub.

Data Governance

As data is being fetched it can undergo mathematical transformation, e.g., change of granularity, units, etc, and validation under certain rules. Data access and action are controlled by user groups. Audit Trail applies to access and changes.

Data Transform

The API service allows to perform data transformation on the fly, such as change grain data. E.g., load 1-min data from hourly data by some methods to fill the inner points. Or load hourly data from 1-min data by some method such as average, sum, min, mix etc.

API and Data Management

A flexible API service manages access and publication of all data in the Data Hun. A web dashboard allows authorized users to interact with any data in the Data Hub: search, view, upload, download, edit, delete, plot and visualize.

Advanced Data Architecture

We use Postgres Timescale DB which offers the best performance for high-frequency time series or sensor data thanks to its hypertable capabilities. Our data platform has an optimized architecture for big data. We use Json and Json BLOB (binary large object) format to store multiple data in one record rather than single entry data. For example, is we store weather indicators, temperature, cloudiness, UV, wind speed, etc, rather than storing these as individual data, we store them all as 1 single Json file entry with a tag for each indicator. The API service reads and writes individual Json tags. The Jason storage format allows much faster query and smaller storage space.

Advanced AI Platform

Our AI platform offers a range of Machine Learning and Deep Learning models suitable for AI modeling of behavior and anomaly detection in different settings. The AI models are supervised, unsupervised and semi-supervised. They forecast, classify and model scores.

An AI is as effective as its features. Our AI platform automates data processing and feature extraction with 100+ mathematical transformers that are especially configured to extract features from a wide range of data such as sensors, event logs, financial, sales, online systems, supply chain and production.
The AI platform is a no-coding, auto-ML platform, reducing AI modeling and data science feature extraction work from weeks to days, so that your AI models can go live in record time.
The entire platform can be deployed on private cloud or on-ite as per your preference.

Algorithms

AI models

We cover 95% of existing models and continuously update our listing

Service

Accelerate your transition to an intelligent digital enterprise