Time series Anomaly Detector App using Azure Cognitive Services

Anomaly detection (AD) in time series is significant in many industrial, medical, and scientific applications. For instance, undetected anomalies in water treatment facilities or chemical plants can bring harm to millions of people. Time series analysis can be useful to see how a given asset, security, or economic variable changes over time. It can also be used to examine how the changes associated with the chosen data point compare to shifts in other variables over the same time period.

Anomaly detection is the identification of rare events, items, or observations which are suspicious because they differ significantly from standard behaviors or patterns. Anomalies in data are also called standard deviations, outliers, noise, novelties, and exceptions. Our Time-Series Anomaly Detector App detects any sudden spikes or anomalies from the manual or uploaded data, these data can be of any type e.g., bank statements, the leaves for a company, personal expenses, or values in a report, etc. So, the API will take numeric values and check if there’s any irregular pattern or spike.

In this system, the user will need to register to log in to the system with their basic details. The user can add a series of numeric data manually or upload an excel file to the system and select a column. The system will show the data in a graphical representation from the spikes, dips or deviations that’s been automatically detected. The user can save all the data.

They can easily view and delete these data from the list of saved data. This project is written in Dart and the database used here is MSSQL, and it is based on the flutter framework. Dart is a programming language that Google developed and keeps up with. A cross-platform framework for building high-performance mobile apps is called Flutter. Cognitive Services are a set of machine learning algorithms that Microsoft has developed to solve problems. It will allow scanning of the device folders.

Advantages

• It is easy to maintain.
• It is user-friendly.
• The system will automatically detect anomalies from any type of data.
• It will also display the reports in the form of graphical representation.