DETECT. ADAPT. IMPACT.
ed.Detect is an advanced time series analysis solution based on machine learning and artificial intelligence. ed.Detect time series analysis identifies anomalies and conspicuous behavior intelligently. In several thousand time series simultaneously and fully automated.
Advanced time series analysis
In today’s data-controlled world advanced time series analysis is of crucial importance in order to identify patterns and trends in various data sources.
The good thing is:
- time series data is simple in structure
- time series data is generally available
The bad thing is:
- there are too many time series to monitor and to understand
Time series data is a sequence of data points which are recorded over certain time intervals. This data can come from different sources for example finance systems, sensors, payment systems or web analytics systems.
ed.Detect was specially created for time series analysis. For each time series it:
- detects anomalies
- forecasts future values
- provides intelligent and targeted notifications
The three parts – anomaly detection, time series forecasting and intelligent alerts – are of equal importance.
ed.Detect applies AI & ML to time series analysis
With machine learning algorithms and artificial intelligence ed.Detect detects fluctuations and anomalies in time series data. ed.Detect learns each time series and understands its structure. By doing so ed.Detect is able to identify unordinary changes and anomalies that can reveal potential problems or uplift potential for future optimization.
By integrating ed.Detect into the data analytics process companies get valuable insights into their data and can quickly react to abnormal behaviour. The tool provides an intelligent alerting engine, enabling non-data scientists to evaluate and asses detected anomalies as well as to take action based on the targeted information provided by ed.Detect.
Because every time series is a simple data structure, implementation of ed.Detect is not complex. The implementation does not require individual implementation efforts. The only requirement is access to the data ed.Detect shall analyze. All transactions that are possibly required are performed within ed.Detect. Furthermore the solution provides multiple methods for receiving data from different sources. The e-dynamics professional services team will assist the implementation of ed.Detect to ensure a fast, time and cost efficient setup.
The intelligent alert and forecasting modules of ed.Detect operate directly on the result set generated by the ed.Detect core engine.
BENEFITS
ed.Detect is interesting for:
- Organizations that collect and analyze large number (thousands) of time series data
- Data-driven users who want to automatically detect deviations and anomalies in their processes, systems or on their websites
Advantage: Up-to-date about deviations at any time!
- Simultaneous monitoring of a thousands of data streams without additional manual effort
- Intelligent and self learning anomaly detection that is optimal for both Small and Big Data
- Consideration of seasonal and trend-specific developments as well as special events, e.g. public holidays and external influences
Advantage: Precise predictions through machine learning!
- Individual alerting function for prospect and responsibility groups
- Integration into existing data and reporting solutions without the requirement of learning or using an additional tool
- ed.Detect result set is provided to third party systems
Alerting Noticication Mail
USE CASE: BOT DETECTION WITH ED.DETECT
Distinguishing between (well-disguised) bot traffic and human traffic is becoming increasingly difficult and posed a major challenge for our customer. This is because bot traffic not only places an unnecessary burden on the infrastructure, but also makes it difficult to assess paid traffic acquisition and distorts conversion measurement.
By using our AI-based solution ed.Detect, the manual effort could be significantly reduced and at the same time the speed and accuracy could be strongly increased through fully automated bot detection. ✔️
Click here for the use case bot detection with ed.Detect!
