The invention provides a non-intervention monitoring and evaluating method for hypertension under a non-clinical environment. The method comprises steps of segmenting acquired electrocardiogram data to obtain an electrocardiosignal sequence taking a time window of 5 minutes as a unit, positioning the time of each
heartbeat by using an overlapping sliding window
algorithm, and calculating a primaryheartbeat
interval sequence; correcting the acquired primary
heartbeat interval sequence by adopting a threshold method; extracting linear domain, non-linear domain and waveform distribution morphological features from the
heartbeat interval sequence of each time window, quantitatively calculating correlation between the
heart rate variability linear domain and the non-linear domain features through a Pearson
correlation analysis method, performing
feature aggregation based on the correlation strength between the features, and selecting features through
information gain to form feature subsets; and fusing the feature subset features, and constructing a hypertension evaluation model by using a
random forest method. According to the method, finer
granularity analysis is carried out, and evaluation accuracy rate of hypertension patients can reach 97.1%.