"police arrests 4 people"
"7 people were arrested".

gen arrest= regexm(description, "(^[1-9][0-9]$)[ ]*(arrests|arrested)")

(d+)[^,.dn]+?(?=arrest|custody)|(?<=arrest|custody)[^,.dn]+?(d+)

(?:([1-9]?[0-9])[a-zA-Z ]{0,20}(?:arrests|arrested))|(?:(?:arrests|arrested)[a-zA-Z ]{0,20}([1-9]?[0-9]))