top of page
  • Writer's pictureDale Montrone

IoT Data Anomaly Detection Using Machine Learning

In our previous Blog “Blockchain Powered IoT: Anomaly Detection”, we talked about anomaly detection in the time series data generated by the Internet of Things (IoT). We pointed out the analytical challenges posed by certain anomalous situations which require machine learning techniques for analysis.

Time Series is defined as a set of observations taken at a particular period of time. [1] For example, having a set of login details at regular intervals of time of each user can be categorized as a time series. On the other hand, when the data is collected only once or irregularly, it is not taken as a time series data.

Time series is a sequence that is taken successively at the equally pace of time. It appears naturally in many application areas such as economics, science, environment, medicine, etc. With the use of time series, it becomes possible to imagine what will happen in the future as future event depends upon the current situation.

In order to analyze the time series data, there is a need to understand the underlying pattern of data ordered at a particular time. This pattern is composed of different components which collectively yield the set of observations of time series. The Components of a time series data stream could be as follows:

1) Trend: Is a long pattern present in the time series.

2) Cyclical: Is a pattern that exhibits up and down movements around a specified trend.

3) Seasonal: Is a pattern that reflects regular fluctuations. It always consists of a fixed and known period.

4) Irregular: It is an unpredictable component of time series. These are short term fluctuations that are not systematic in nature and have unclear patterns.

There are two common approaches used for analyzing time series data: (a) Statistical approach and (b) Machine Learning based approach. Machine learning techniques are more effective as compared with the statistical techniques. This will be the subject of our next Blog.



338 views0 comments
bottom of page