Contact tracing pre-screening method for infectious disease susceptible people based on wifi matching

A contact tracing pre-screening method for infectious disease susceptible people based on WiFi matching, wherein, characterized in that the method comprises following steps:

• 1) matching with user anonymous identification code record:
• obtaining a unique anonymous identification code data table of confirmed users through a database, using hive sql tool to search for WiFi connection records corresponding to identification codes (IDs) in a total database, creating data table partitions according to dates, and saving the data table partitions as original data records with a total data time span of T days, and constructing a WiFi connection data set R={R1, R2, . . . , RT} of the confirmed users, wherein records of the ith day are Ri, and there are a total of ki records, and Ri={ri1, ri2, . . . , riki}, wherein each of the records rij=(gidij, confirmDayij, macij), gidij represents a unique anonymous identification code of a user, confirmDayij represents a confirmed date of the user, and macij represents a media access control (MAC) address of an WiFi to which the user connects;
• 2) pre-processing data:
• saving the WiFi connection data set R obtained in step (1) to local codes, and establishing a two-way mapping dictionary class, creating a new referencing symbol for any string according to a number of internal stored entries after obtaining the string, using the string and the referencing symbol respectively as a key and a value to establish a mapping;
• establishing a code two-way mapping dictionary dictuser for the anonymous identification codes of the users, recording a total number of the users as Nu, and mapping original anonymous identification code with uniform-length of the user to a code number uiu, wherein iu=1, 2, 3, . . . , Nu;
• establishing a code two-way mapping dictionary dictwifi of the WiFis, recording a total number of the WiFis as Nw, and mapping original MAC address of the WiFi as a code number wuw, wherein uw=1, 2, 3, . . . , Nw;
• recording a record data set obtained after mapping an original record data set as R;
• 3) constructing a dangerous WiFi database:
• using the record data set R after code number mapping in step (2), recording the dangerous WiFi database as dictdan, extracting the WiFi connection records of the confirmed users in R, taking the code number wuw of the WiFi in records and putting wuw in dictdan, and recording a date in which the WiFi is recorded in the dangerous WiFi database as a current date daynow;
• if the WiFi already exists, updating a recording date thereof to daynow, and then deleting a record of which a difference between a recording date dayuw and the current date daynow exceeds a dangerous WiFi disappearance threshold thrdan;
• 4) judging an infection risk:
• recording newly acquired WiFi connection records of the users every day as {tilde over (R)}, grouping {tilde over (R)} by the users, and for each of the users, matching all the connected WiFis thereof with dictdan, and recording a coincidence rate as ruser=|Cuser|/|Wuser|, wherein |Cuser| is a number of WiFis in dictdan that the user connects to in that date, |Wuser| is the total number of WiFis that the user connects in that date, and if ruser is higher than a judgment threshold thruser, judging the user as a dangerous user.