Marketing customer archive information abnormity positioning method

Abstract

The invention relates to the technical field of data processing, in particular to a marketing customer archive information abnormity positioning method. The method comprises the following steps: acquiring short message sending states including an uplink short message state and a downlink short message state; users who have sent short messages and users who do not send short messages are screened according to the short message sending state, and whether the users who have sent short messages successfully send short messages or not is distinguished; wherein the mobile phone number of which the short message is not successfully sent is set as a missed sending mobile phone number; respectively carrying out downlink short message state verification on the missed sending mobile phone number and the wrong sending mobile phone number; according to the invention, through comprehensive monitoring and verification of the short message sending state, including careful screening of missed sending and wrong sending of mobile phone numbers, abnormal customer archive information can be accurately positioned and corrected, and key information such as customer mobile phone numbers and the like is ensured to be accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for locating abnormalities in marketing customer archive information. Background Art

[0002] Existing archive management solutions mainly focus on the following aspects: (1) Manual census and information comparison: This is the most common practice. Staff need to conduct on-site visits regularly or irregularly to check the information in customer files one by one, such as account name, contact information, electricity address, etc. This process is cumbersome and inefficient, especially when facing a large number of customers. It is difficult to ensure the timeliness and accuracy of information. (2) Telephone follow-up and email confirmation: In addition to on-site visits, power companies will also contact customers by phone or email to confirm the accuracy of archive information. Although this method can improve efficiency to a certain extent, it is still limited by customer response rate and the reliability of information transmission. Sometimes customers may ignore or forget to reply, resulting in delayed information updates.

[0003] The defects of existing technologies are mainly reflected in the following points: (1) Abnormal problems are difficult to detect: Due to the limitations of manual verification, some hidden or complex abnormal file problems are difficult to be discovered in time. (2) Large personnel input: Continuous human resource investment makes the cost of file management high, especially in large power companies, the labor cost required to maintain a large customer base cannot be underestimated. (2) Low labor efficiency: Manual operation is slow and prone to fatigue and negligence, which directly affects the efficiency and quality of file management. (3) Long governance cycle: From information collection to the final file update, the entire process may take weeks or even months, which is a major challenge for power services that need to respond to customer needs in a timely manner.

[0004] In summary, existing technologies have many shortcomings in customer archive management. There is an urgent need to use digital and automation technologies to achieve more efficient, accurate and automated methods for archive information to replace or supplement existing labor-intensive solutions. Summary of the invention

[0005] In order to solve the above-mentioned problems, the present invention provides a method for locating abnormalities in marketing customer file information.

[0006] In a first aspect, the present invention provides a method for locating abnormal marketing customer profile information, which adopts the following technical solution: A method for locating abnormality in marketing customer profile information, comprising: Get the SMS sending status, including the uplink SMS status and the downlink SMS status; Users who have sent text messages and users who have not sent text messages are filtered out based on the text message sending status. Among them, for users who have sent text messages, it is distinguished whether the text messages have been sent successfully; among them, the mobile phone numbers to which the text messages have not been sent successfully are set as missed mobile phone numbers; and the mobile phone numbers to which the text messages have been sent successfully are screened for wrong mobile phone numbers.

[0007] Perform status check of downlink SMS for missed and wrong mobile phone numbers; According to the status verification results, the abnormal location of the mobile phone number and the status correction are carried out.

[0008] Furthermore, the mobile phone numbers to which text messages are successfully sent are screened for incorrect mobile phone numbers, including screening out mobile phone numbers that actively report abnormalities from uplink text messages, and providing the abnormal data to the background; for mobile phone numbers that do not actively report abnormalities, cross-verification is performed with other external data. If the verification passes, it can be determined that the user name and mobile phone number match; if the verification fails, the wrong text message sending mobile phone number management process is entered for verification.

[0009] Furthermore, the downlink SMS status is checked for the missed mobile phone numbers and the wrongly sent mobile phone numbers, including number portability check, network status check, shutdown check and number cancellation check for the mobile phone numbers that failed to send the downlink SMS, and the mobile phone number status is corrected online based on the check results.

[0010] Furthermore, the number portability verification includes synchronizing the time according to the number portability database. When the operator's SMS channel is selected to send successfully after the number portability, the missed mobile phone number is judged to be a normal mobile phone number.

[0011] Furthermore, the network status verification includes querying the network status of the missed mobile phone number, including normal network, unavailable network, shutdown and cancellation; wherein, the network status of the mobile phone numbers that are normally online and unavailable network are verified twice respectively. If the same network status verification result is still maintained, the missed mobile phone number is judged as an abnormal mobile phone number.

[0012] Furthermore, the shutdown verification includes performing a secondary online status verification on the shutdown mobile phone number according to the operator's shutdown rules, and performing corresponding processing according to the verification results. If the verification result is still shutdown, the online status verification is continuously repeated.

[0013] Furthermore, the cancellation verification includes determining that the missed mobile phone number has been cancelled based on the cancellation verification result, recording the mobile phone number and sending it to the background for correction.

[0014] Furthermore, the status verification of downlink SMS messages for missed mobile phone numbers and wrong mobile phone numbers respectively includes checking the data accuracy of uplink SMS messages with feedback, and performing three-factor verification, family member verification and intelligent outbound call verification on downlink SMS messages without feedback.

[0015] Furthermore, the family member verification includes modeling the passing data of the mobile phone number, and using the decision tree CHAID chi-square automatic cross-check algorithm to determine whether the mobile phone number and the corresponding ID card number are family members.

[0016] Furthermore, the intelligent outbound call verification automatically dials the user's mobile phone number through intelligent voice to confirm the user's file information, and confirms abnormal file information through voice recognition and NLP technology.

[0017] In summary, the present invention has the following beneficial technical effects: The present invention can accurately locate and correct abnormal customer file information through comprehensive monitoring and verification of the SMS sending status, including detailed screening of missed and wrongly sent mobile phone numbers, to ensure the accuracy of key information such as customer mobile phone numbers.

[0018] Accurate customer profile information helps improve the success rate and arrival rate of marketing SMS, thereby improving the effectiveness of marketing activities. At the same time, it reduces the trouble caused to customers by wrong or missed SMS, and enhances customer acceptance and satisfaction with corporate marketing activities.

[0019] Accurate customer positioning and information correction reduce invalid SMS sending and marketing costs. At the same time, it helps to establish a more complete, accurate and reliable customer profile database, providing strong support for the company's data analysis and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of a method for locating abnormal marketing customer profile information according to an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of the wrongly sent mobile phone number verification process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0023] Example 1 Reference Figure 1 A method for locating abnormal marketing customer profile information in this embodiment includes: Get the SMS sending status, including the uplink SMS status and the downlink SMS status; According to the SMS sending status, users who have sent SMS and users who have not sent SMS are screened, wherein for users who have sent SMS, it is distinguished whether the SMS has been sent successfully; wherein, the mobile phone number for which the SMS has not been sent successfully is set as the missing mobile phone number; and the mobile phone number for which the SMS has been sent successfully is screened for the wrong mobile phone number. The status of the missing mobile phone number and the wrong mobile phone number are respectively checked for the downlink SMS; and the abnormal location and status of the mobile phone number are corrected according to the status verification result. The wrong mobile phone number screening for the mobile phone number that successfully sent the SMS includes screening out the mobile phone numbers that actively feedback the abnormality from the uplink SMS, and providing the abnormal data to the background; for the mobile phone numbers that did not actively feedback the abnormality, cross-verification is performed with other external data. If the verification is passed, it can be determined that the user name and the mobile phone number match; if the verification fails, the SMS wrong mobile phone number management process is entered for verification. The downlink SMS status of the missing mobile phone number and the wrong mobile phone number are respectively checked, including the number portability verification, the online status verification, the shutdown verification and the number cancellation verification for the mobile phone number that failed to send the downlink SMS, and the mobile phone number status is corrected online based on the verification result. The number portability verification includes, according to the synchronization time of the number portability database, when the operator's SMS channel is successfully sent after the number portability, the missing mobile phone number is judged as a normal mobile phone number. The online status verification includes querying the online status of the missing mobile phone number, including normal online, unavailable online, suspended and cancelled; wherein, the mobile phone numbers that are normally online and unavailable online are respectively verified twice on the online status, if the same online status verification result is still maintained, the missing mobile phone number is judged as an abnormal mobile phone number. The suspension verification includes, according to the operator's suspension rules, conducting a secondary online status verification on the suspended mobile phone number, and performing corresponding processing respectively according to the verification results. If the verification result is still suspended, the online status verification is continuously cycled. The cancellation verification includes, according to the cancellation verification result, judging that the missing mobile phone number has been cancelled, recording the mobile phone number and sending it to the background for correction. The status verification of the downlink SMS is respectively performed on the missing mobile phone number and the wrongly sent mobile phone number, including the data accuracy verification of the uplink SMS with feedback, and the three-factor verification, family member verification and intelligent outbound call verification for the uplink SMS without feedback. The family member verification includes modeling the data of mobile phone numbers, using the decision tree CHAID chi-square automatic cross-check algorithm to determine whether the mobile phone number and the corresponding ID number are family members. The intelligent outbound call verification automatically dials the user's mobile phone number through intelligent voice to confirm the user's profile information, and confirms abnormal profile information through voice recognition and NLP technology. As a further embodiment, In this embodiment, regarding the management of abnormal mobile phone numbers, it is through the application of big data analysis and artificial intelligence technology, combined with number portability verification, online status verification, intelligent outbound calling and other services, using digital means, integrating internal and external data and service capabilities, starting from the global power customers, and classifying and grading the mobile phone numbers of electricity users. On the one hand, it is necessary to focus on global users and systematically solve the problem of mobile phone number management for existing customers; on the other hand, focus on selected high-risk users, review and confirm through intelligent outbound calling and other means, start from high-risk users, and implement targeted management for different levels of electricity users, so as to achieve accurate positioning and step-by-step progress, systematically solve the existing problems of mobile phone numbers of existing electricity users, prevent the emergence of incremental problems, and thus assist customers in solving related problems such as wrong or missed mobile phone calls.

[0024] like Figure 1 As shown, based on the data of sent mobile phone text messages, by obtaining the text message sending status, screen the users who have sent text messages, and distinguish whether the text messages are sent successfully. The mobile phone numbers that have not been sent successfully are abnormal (i.e., missed text messages). For the mobile phone numbers with missed text messages, carry out number portability verification and online status verification and troubleshooting to locate the abnormal mobile phone numbers more quickly and accurately.

[0025] For mobile phone numbers that have successfully sent SMS messages, we focus on screening "wrong" numbers. First, we screen out customers who actively report abnormalities from the uplink SMS, and provide the data to the staff for management; for users who did not actively report abnormalities, we "cross-verify with other external data". If the verification passes, it can be determined that the user name and mobile phone number match each other; if the verification fails, we will enter the process of managing wrong SMS numbers and carry out verification.

[0026] 1. Manage the mobile phone number where SMS messages are missed After analysis, the situations of failed downlink SMS can be basically divided into five types: empty number, shutdown, outage, blacklist, and other abnormalities. Among them, the empty number problem accounts for about 40%, which is a large proportion. In most cases, the empty number is caused by number cancellation, expiration, blocking or blocking of calls, information errors or other technical problems. Among them, the number cancellation and expiration can basically determine that the user no longer uses the mobile phone number and needs to modify the user profile information; shutdown, outage, blacklist, and other abnormalities cannot determine whether the user continues to use the mobile phone number, and further judgment is required. In view of the above situations, a suspected wrong mobile phone number analysis model is constructed, and with the help of the operator's number portability, online status verification and other model capabilities, the cause of the failure to send SMS is accurately located, and the following solution is used to further analyze and process the user's mobile phone number.

[0027] ①Number portability verification For mobile phone numbers that fail to send marketing SMS, a full comparison is performed in real time on the dynamic verification database of number portability to determine whether the failure is caused by number portability. The current system's number portability comparison database is synchronized to the relevant database of the Ministry of Industry and Information Technology the next day if the user ports the number on the same day. The system synchronizes data on the third day, which can basically achieve quasi-real-time correctness of the data. If the SMS fails to be sent due to number portability, and the SMS channel selected by the operator after the number portability is successfully sent, then the user's mobile phone number is normal.

[0028] ②Online status verification The online status of the mobile phone number that failed to send SMS is verified. The online status verification is to query the online status of the mobile phone number, and the returned content includes multiple states such as normal online, unavailable online, downtime, cancelled number (off the network), etc. For the processed mobile phone numbers that failed to send SMS, the mobile phone numbers that have completed the online status verification are generally not verified again (except for the cases of normal online / unavailable online, downtime, etc., which require re-verification of the online status), to avoid repeated verification waste.

[0029] The mobile number is normally online / unavailable online, both mean that the mobile number is online, but the SMS fails to be sent due to some abnormal situation. Among them, the failure to send SMS normally online may be due to abnormal situations such as the mobile number being turned off, blacklisted, or other abnormalities. The unavailable online may be due to the insufficient contact between the metal contacts of the SIM card and the mobile phone, the SIM card is damaged or there is a problem with the mobile phone, the current location is a signal blind area or there is no signal suddenly, etc.

[0030] After the network status verification, if the user's mobile phone number is returned as being normally online / unavailable, the smart outbound call verification method is used to confirm whether the mobile phone number is the user's mobile phone number in use. If the user information is correct, then the user's mobile phone number is normal. If the smart outbound call verification has no result, then the current mobile phone number status is recorded and marketing SMS messages continue to be sent according to the current mobile phone number. If SMS sending fails for three consecutive months, the mobile phone number will be verified for the second time. If the mobile phone number is verified twice and the results of the network status are both normally online / unavailable, then the mobile phone number is basically judged as an abnormal mobile phone number. The system records the status of the mobile phone number and feedbacks it to each unit for manual verification and modification.

[0031] ③Shutdown After the online status verification, if the user's mobile phone number is returned as being suspended, it is preliminarily determined that the mobile phone number is suspended due to arrears or other reasons. According to the operator's regulations, if the mobile phone number is suspended for 3 months, the mobile phone number will be cancelled and recycled. For the mobile phone number whose online status is verified as suspended, if the mobile phone number still fails to send text messages for 3 months, the online status of the mobile phone number will be re-verified, and the verification results will be processed accordingly according to the different results of normal online / unavailable online, suspended, and cancelled (off the network). Among them, if the result is suspended, SMS sending + online status verification will be repeated in a cycle of 3 months. If the result is cancelled (off the network), the system will record the status of the mobile phone number and feedback to each unit for manual verification and modification.

[0032] ④ Account cancelled (offline) After the online status verification, if the user's mobile phone number is returned as cancelled (off the network), it is preliminarily judged that the mobile phone number has been cancelled. The system records the status of the mobile phone number and feedbacks it to various units for manual verification and modification.

[0033] ⑤Manual verification and modification The system will simultaneously record the status of mobile phone numbers that have been cancelled after network status verification, or mobile phone numbers that have been verified twice as being online / unavailable and have failed to send text messages. These user mobile phone numbers cannot be confirmed or modified by smart outbound calls or by dialing the mobile phone numbers. Marketing-related grassroots staff need to verify with the customer offline based on the system's mobile phone number status information, and make file corrections after the customer's consent.

[0034] 2. Solutions for SMS messages sent to wrong mobile phone numbers Missent SMS messages are a key factor that causes a large number of customer service calls to report abnormal basic information. Based on the experience of business personnel, data analysis technology is used to identify key customers with high risk of abnormal mobile phone numbers, such as abnormal uplink SMS feedback, multiple numbers for one household, out-of-town mobile phone numbers, and other key customers who meet specific rules. Then, using service capabilities such as "three-factor authentication", "family member verification", and "intelligent outbound calls", we proactively check and accurately identify potential customer mobile phone number anomalies, and assist customers in manual verification and modification. The specific process is as follows: Figure 2 : (1) Uplink SMS feedback For SMS messages that customers actively provide feedback on, the accuracy of the data needs to be manually verified. If the data is accurate, it will be considered normal data; if the data is inaccurate, it needs to be modified to the correct data.

[0035] For text messages that users have not responded to, it is necessary to conduct correlation analysis with other external data, "three-factor verification", "family members" and "smart outbound calls" to locate the correct mobile phone number.

[0036] (2) Three-factor verification Based on the personal data information reserved in the marketing customer files, the three data elements of name, ID number and mobile phone number are verified for consistency and the results are fed back.

[0037] (3) Family member verification The users' data are labeled in a standardized manner, and a model is established using TFIDF calculations of the number of calls and the number of calling users, as well as 17 indicators. The decision tree CHAID chi-square automatic cross-validation algorithm is used to determine whether the mobile phone number and ID card number belong to a family member. The accuracy has been verified to be 76.98%.

[0038] (4) Intelligent voice verification According to the actual needs of customers, the user's mobile phone number is automatically dialed through intelligent voice to confirm the user's file information, and abnormal file information is confirmed through voice recognition and NLP technology.

[0039] (5) Manual verification and modification Data that is actively fed back in the form of uplink SMS and fails verification through three methods, including three-factor authentication, family member verification, and smart outbound calling, can basically be judged as erroneous file data and require further manual verification and modification to ensure data accuracy.

[0040] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for locating abnormalities in marketing customer profile information, characterized in that: include Get the SMS sending status, including the uplink SMS status and the downlink SMS status; Users who have sent text messages and users who have not sent text messages are filtered out based on the text message sending status. Among them, for users who have sent text messages, it is distinguished whether the text messages have been sent successfully; among them, the mobile phone numbers to which the text messages have not been sent successfully are set as missed mobile phone numbers; and the mobile phone numbers to which the text messages have been sent successfully are screened for wrong mobile phone numbers. Perform status check of downlink SMS for missed and wrong mobile phone numbers; According to the status verification results, the abnormal location of the mobile phone number and the status correction are carried out.

2. A method for locating abnormal marketing customer profile information according to claim 1, characterized in that: The screening of wrongly sent mobile phone numbers for mobile phone numbers to which text messages are successfully sent includes screening out mobile phone numbers that actively report abnormalities from uplink text messages, and providing abnormal data to the backend; for mobile phone numbers that do not actively report abnormalities, cross-verification is performed with other external data. If the verification passes, it can be determined that the user name and mobile phone number match; if the verification fails, the wrongly sent text message mobile phone number management process is entered for verification processing.

3. A method for locating abnormal marketing customer profile information according to claim 2, characterized in that: The downlink SMS status verification is performed on the missed mobile phone numbers and the wrongly sent mobile phone numbers respectively, including performing number portability verification, network status verification, shutdown verification and number cancellation verification on the mobile phone numbers that failed to send the downlink SMS, and online correction of the mobile phone number status is performed based on the verification results.

4. A method for locating abnormal marketing customer profile information according to claim 3, characterized in that: The number portability verification includes synchronizing the time according to the number portability database. When the operator's SMS channel is selected to send successfully after the number portability, the missed mobile phone number is judged to be a normal mobile phone number.

5. A method for locating abnormal marketing customer profile information according to claim 4, characterized in that: The network status verification includes querying the network status of the missed mobile phone number, including normal network, unavailable network, shutdown and cancellation; wherein, the network status of the mobile phone numbers that are normally online and unavailable network are verified twice, and if the same network status verification result is still maintained, the missed mobile phone number is judged as an abnormal mobile phone number.

6. A method for locating abnormal marketing customer profile information according to claim 5, characterized in that: The shutdown verification includes performing a secondary online status verification on the shutdown mobile phone number according to the operator's shutdown rules, and performing corresponding processing according to the verification results. If the verification result is still shutdown, the online status verification is continuously repeated.

7. A method for locating abnormal marketing customer profile information according to claim 6, characterized in that: The cancellation verification includes determining that the missed mobile phone number has been cancelled based on the cancellation verification result, recording the mobile phone number and sending it to the background for correction.

8. A method for locating abnormal marketing customer profile information according to claim 7, characterized in that: The status verification of downlink SMS messages for missed mobile phone numbers and wrong mobile phone numbers also includes checking the data accuracy of uplink SMS messages with feedback, and performing three-factor verification, family member verification and intelligent outbound call verification on downlink SMS messages without feedback.

9. A method for locating abnormal marketing customer profile information according to claim 8, characterized in that: The family member verification includes modeling the passing data of the mobile phone number, and using the decision tree CHAID chi-square automatic cross-check algorithm to determine whether the mobile phone number and the corresponding ID card number are family members.

10. A method for locating abnormal marketing customer profile information according to claim 9, characterized in that: The intelligent outbound call verification automatically dials the user's mobile phone number through intelligent voice to confirm the user's file information, and confirms abnormal file information through voice recognition and NLP technology.

Images