1
$\begingroup$

I have a time series data. I have to identify the point in red. I know how to solve this using statistics. But if I had to use Machine Learning to detect these points how can I solve this.

enter image description here

How can I modify this data in order to feed it to machine learning model. Data is just one column readings as show in plot. Any idea?

$\endgroup$
1
$\begingroup$

Generally, You can use anomaly detection to detect a sudden change in Timeseries data. SOM is useful for detect it. Its like you want to capture an anomly in continuous distribution of data.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.