A method for improving call completion rate

Abstract

The application relates to a method for improving call connection rate, belonging to the field of voip voice data communication, which comprises the following steps: storing the signaling meeting preset conditions in the called side sip temporary response information when a call request into a signaling database; intelligently inferring and analyzing the listening behavior of a customer to an incoming call to obtain sensitive number information, storing the analyzed sensitive number information into a sensitive number information database; extracting desensitization number information, extracting a complaint customer number as sensitive number information and storing the sensitive number information into the sensitive number information database; writing sensitive customer number information into the sensitive number information database, simultaneously increasing a cache in the front end of the sensitive number information database; establishing a bloom filter in the front end of the cache during idle time of platform business; and realizing call behavior control according to the bloom filter, the cache and the query result of the sensitive number information database. The method can improve the call connection capability of a voice call platform and the telephone call connection rate, and reduce the complaint rate of the called side customer.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of voip voice data communication, and particularly relates to a method for improving call connection rate. BACKGROUND

[0002] The current voice call platform is an AS (Application Server) role in an IMS large network, and realizes call number changing service. Figure 1 As shown in the figure, the technical process is as follows: when the platform receives a call request sent by a core network device, the called number is extracted, and a local explicit virtual number associated with the called number is queried. Partial signaling conversion is realized according to the call indication information returned by the query to realize number changing outgoing call, so that the caller number displayed by the called terminal is a virtual number. At the same time, according to the routing information associated with the real called number in the call indication information, the call is sent to the called side network element until the called user terminal, so as to realize the protection of the privacy of the calling party and the called party and improve the call connection rate.

[0003] The existing platform number changing display virtual number technical scheme has the following problems:

[0004] (1) The growth of platform traffic causes a large number of number changing outgoing calls, and the capacity causes a large number of complaints from called users;

[0005] (2) A large number of outgoing calls during platform business operation cause repeated calls to sensitive called users to be easily hung up, resulting in low traffic connection rate;

[0006] (3) The business number of the platform is easily blacklisted or disabled by a third party, resulting in unavailable number resources.

[0007] (4) When the enterprise customers based on the voice cloud capability platform use virtual number changing to realize value-added services, the frequent calls during user business development will cause complaints from customers, causing the calling number to be blacklisted or complained, resulting in waste of platform business number and reduction of connection rate. SUMMARY

[0008] In view of the above problems of the prior art, the purpose of the present application is to provide a method for improving call connection rate, which can improve the call connection rate of the voice call platform, reduce the complaint rate of the called side customers, and maximize the protection of the call demand of the enterprise.

[0009] The present application provides a method for improving call connection rate, comprising:

[0010] S1, call signaling collection: storing the signaling in the signaling database that meets the preset condition in the called side sip temporary response information when the call request is received;

[0011] S2, Intelligent Reasoning and Analysis of Call Behavior: Based on the customer's answering behavior of incoming calls, intelligent reasoning and analysis are performed to obtain sensitive number information, and the analyzed sensitive number information is stored in the sensitive number information database;

[0012] S3, Extraction of de-identified number information: For the de-identified number information of the complaint platform, an intelligent analysis model is established based on the known information of the calling customer. The call traffic model and the rule-based call record library are used for matching analysis. If the complaint is matched with the corresponding call record according to the intelligent analysis model, it indicates that the call is real. The complaining customer's number is extracted as sensitive number information and stored in the sensitive number information library.

[0013] S4, Construct sensitive number data: Write sensitive customer number information into the sensitive number information database. At the same time, add a high-speed cache to the front end of the sensitive number information database. When each call to the platform queries whether the called number is a sensitive number, it first accesses the high-speed cache. If the high-speed cache does not exist, it then accesses the sensitive number information database for matching and comparison.

[0014] S5, Establish a Bloom filter: When the platform is idle, establish a Bloom filter in front of the cache. When a call request comes in to query and verify a sensitive customer number, if the Bloom filter does not contain the called number, the call will be returned directly; if the Bloom filter contains the called number, the cache will be queried further.

[0015] S6, Platform Outbound Call Processing: Based on the query results of the Bloom filter, cache, and sensitive number information database, the platform controls call behavior, including: for calls that do not exist in the Bloom filter, the caller ID is changed and the call is made directly; for calls that exist in the Bloom filter, if the called number is matched with the corresponding result in the cache and sensitive number information database, the call is disconnected and a prompt tone is played to inform the calling user.

[0016] Furthermore, in S1, the preset condition is: an 18X temporary response SIP message, carrying a signaling message with a reason header field indicating call rejected or user busy.

[0017] Furthermore, in S2, intelligent reasoning analysis is performed based on the customer's answering behavior to incoming calls, including:

[0018] Analysis rules and models are established for the signaling database. When it is detected that the platform makes multiple outbound calls to the same called number and the calls are all directly hung up, it indicates that the customer is sensitive or unwilling to answer calls from unknown numbers. The called customer's number is then set as a sensitive number.

[0019] Data is extracted from the call detail record (CDR) database. If the same called customer answers multiple calls from the platform's number change service and then hangs up within the first preset time period, it indicates that the customer is unwilling to answer or is not interested. The called customer's number is then set as a sensitive number.

[0020] Furthermore, the first preset duration is 1 to 3 seconds.

[0021] Furthermore, in S3, the known information of the calling customer includes: the calling customer's industry, service type, date of the call, and call detail record for that day.

[0022] Furthermore, in S3, when the complaint platform receives a customer's complaint, it sends the complaint content to the platform for verification. At this time, the customer's number in the complaint content has been anonymized. An intelligent analysis model is established based on the known information of the calling customer, and matching analysis is performed using the call traffic model and the rule-based dialogue record library, including:

[0023] S31, determine whether the complaint content contains the complainant's number. If so, further determine whether there is a date of the call. Otherwise, proceed to step S39.

[0024] S32, determine whether the complaint content includes the date of the call; if so, proceed to step S33; otherwise, proceed to step S39.

[0025] S33, determine whether the complaint content includes the time of the call. If so, proceed to step S34; otherwise, proceed to step S35.

[0026] S34, determine whether the incoming call time is valid. If it is, proceed to step S36; otherwise, proceed to step S35.

[0027] S35, query the call details for the day. If the call details for the day are found, proceed to step S38; otherwise, proceed to step S39.

[0028] S36, query call details within the second preset duration before and after the incoming call time;

[0029] S37. If the call details are found within the second preset duration before and after the incoming call time, proceed to step S38; otherwise, proceed to step S35.

[0030] S38, determine whether the complaining customer's number matches the first 3 digits and the last 2 digits. If so, extract the de-identified number from the surface.

[0031] S39, it was determined that no de-identified number was extracted.

[0032] Furthermore, the second preset duration is 10 minutes.

[0033] Furthermore, in step S4, if a sensitive number is matched in the sensitive number information database, the query result is returned, and the sensitive number retrieved from the sensitive number information database is stored in a cache for use in subsequent call queries.

[0034] Furthermore, in step S5, if the cache matches the query result, it returns the result; if the cache does not match the result, it queries the sensitive number information database and returns the matching result, while storing the matching result in the sensitive number information database into the cache.

[0035] Furthermore, in S6,

[0036] If the called number does not exist in the Bloom filter, it is determined that the called number is not in the sensitive number information database, and the call is made directly according to the platform's original number change process;

[0037] If the called number exists in the Bloom filter, the cache is queried again to determine if the number exists in the sensitive number information database; if the called number exists in the cache, the call is disconnected and a risk warning tone is played to the calling party; if the called number does not exist in the cache, the outbound call is continued according to the platform's original number change process.

[0038] The beneficial effects of this invention are as follows:

[0039] (1) Incorporate modules such as intelligent analysis model of complaint information, intelligent perception of call behavior, high-speed caching of massive data, efficient Bloom retrieval and decoupling of number change external paging into voice calls to achieve almost zero latency, high efficiency and high connection rate of number change paging on the platform.

[0040] (2) Combining the advantages of platform number change paging and intelligent reasoning for sensitive customers, in addition to ensuring normal number change paging on the voice cloud platform, the platform can improve the business call answer rate of the voice cloud platform by using intelligent analysis of large model numbers to perceive sensitive customers, efficient retrieval of massive data, and the platform’s powerful number change paging capabilities, thus ensuring the platform’s efficient and secure operation.

[0041] (3) Improve the call connection capability of the voice call platform and increase the call connection rate of the platform;

[0042] (4) Reduce the complaint rate of called customers and protect the call needs of enterprises to the greatest extent possible;

[0043] (5) Improve the effective utilization rate of the contracted numbers on the voice cloud platform and reduce unnecessary loss of number resources;

[0044] (6) Strengthen the platform's compliance and security, improve the paging experience for enterprise users, and protect the legitimate rights and interests of called users;

[0045] (7) As the daily call volume of the voice cloud platform reaches nearly 200 million, a high-speed caching mechanism is introduced at the front end of the sensitive number information database and a Bloom filter is established during the platform's idle time to improve the platform's call connection rate and meet the requirements of high-concurrency call processing. Attached Figure Description

[0046] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. It is obvious that the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings.

[0047] Figure 1 Here is a flowchart of the existing normal call processing procedure;

[0048] Figure 2 This is a flowchart of a method for improving call connection rate according to an embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram of a temporary response 18X message according to an embodiment of the present invention;

[0050] Figure 4 This is a flowchart of the de-identified number extraction process according to an embodiment of the present invention;

[0051] Figure 5 This is a flowchart illustrating the process of querying sensitive numbers according to an embodiment of the present invention;

[0052] Figure 6 This is a schematic diagram illustrating the construction of a Bloom filter according to an embodiment of the present invention;

[0053] Figure 7 This is a flowchart illustrating the processing logic of the Bloom filter according to an embodiment of the present invention.

[0054] Figure 8 This is a flowchart illustrating the outbound calling process according to an embodiment of the present invention;

[0055] Figure 9 This is a flowchart illustrating the method for improving call connection rate according to an embodiment of the present invention. Detailed Implementation

[0056] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0057] Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts disclosed in this invention.

[0058] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The terms "installed," "connected," and "linked" should be interpreted broadly; for example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0059] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of methods and systems consistent with some aspects of the invention as detailed in the appended claims.

[0060] The technical terms involved in this invention are explained below:

[0061] SIP (Session Initialization Protocol): A session initiation protocol, refer to RFC3261. RFC3261 is the core specification of SIP, which defines various aspects of the SIP protocol in detail.

[0062] IMS (IP Multimedia Subsystem): IP Multimedia System is a network architecture that integrates mobile and fixed networks into a unified solution for voice, data, and video.

[0063] Bloom Filter: A long binary vector and a series of random mapping functions. This algorithm outperforms ordinary algorithms in terms of both space and time.

[0064] like Figure 2 and Figure 9As shown, the present invention proposes a method for improving call connection rate, comprising:

[0065] S1, Call Signalling Acquisition

[0066] Signaling messages that meet preset conditions in the temporary SIP response information of the called party during a call request are stored in the signaling database.

[0067] In this step, the default condition is: an 18X temporary response SIP message, carrying a reason header field indicating call rejected or user busy, such as... Figure 3 As shown.

[0068] Specifically, the voice call platform is an AS service application within the IMS network, implemented based on the SIP protocol. When the calling party sends a call request to the called party using an INVITE message, the called party, upon receiving the request, will see it as an unknown number or marked as a spam call and will hang up if they do not wish to answer. In this case, the called party will return an 18X temporary response SIP message with a reason header field indicating "call rejected" or "user busy." The platform's signaling media softswitch will then store this type of signaling in the signaling storage database.

[0069] S2, Intelligent Reasoning Analysis of Call Behavior

[0070] Based on customer call answering behavior and call duration in the call detail record database, intelligent reasoning analysis of call behavior is performed, and the analyzed sensitive number information is stored in the sensitive number information database.

[0071] Specifically, analysis rules and models are established for the signaling database. When it is detected that the platform makes multiple outbound calls to the same called number and the calls are all immediately disconnected, it indicates that the customer is sensitive or unwilling to answer calls from unknown numbers. The called customer's number is then marked as a sensitive number.

[0072] Data is extracted from the call detail record (CDR) database. If the same called customer answers multiple calls from the platform's number change service and then hangs up within the first preset time period, it indicates that the customer is unwilling to answer or is not interested. The called customer's number is then set as a sensitive number.

[0073] In this step, the first preset duration is 1 to 3 seconds, which can be set as needed.

[0074] S3, Extraction of de-identified phone number information

[0075] For the anonymized phone number information on the complaint platform, an intelligent analysis model is built based on the known information of the calling customer.

[0076] The known information of the calling customer includes: the calling customer's industry, service type, date of the call, and call details for that day.

[0077] Then, the call volume model and rule-based call record database are used for matching analysis. If the complaint is matched with the corresponding call record according to the intelligent analysis model, it indicates that the call is real. The complainant's number is then extracted as sensitive number information and stored in the sensitive number information database.

[0078] When the complaint platform receives a customer's complaint, it sends the complaint to the platform for verification. At this point, the customer's phone number in the complaint has been anonymized. Based on the known information of the caller, an intelligent analysis model is built, and matching analysis is performed using a call traffic model and a rule-based dialogue record database. Figure 4 As shown, it includes the following steps:

[0079] S31, determine whether the complaint content contains the complainant's number. If so, further determine whether there is a date of the call. Otherwise, proceed to step S39.

[0080] S32, determine whether the complaint content includes the date of the call; if so, proceed to step S33; otherwise, proceed to step S39.

[0081] S33, determine whether the complaint content includes the time of the call. If so, proceed to step S34; otherwise, proceed to step S35.

[0082] S34, determine whether the incoming call time is valid. If it is, proceed to step S36; otherwise, proceed to step S35.

[0083] S35, query the call details for the day. If the call details for the day are found, proceed to step S38; otherwise, proceed to step S39.

[0084] S36, query call details within a second preset duration before and after the incoming call time; where the second preset duration is 10 minutes;

[0085] S37. If the call details are found within the second preset duration before and after the incoming call time, proceed to step S38; otherwise, proceed to step S35.

[0086] S38, determine whether the complaining customer's number matches the first 3 digits and the last 2 digits. If so, extract the de-identified number from the surface.

[0087] S39, it was determined that no de-identified number was extracted.

[0088] In summary, when a customer receives a call from the platform with a spoofed number, such as a marketing call, and is sensitive to such calls, they may report it to the complaint platform. The complaint platform will then send the complaint to the platform for verification. However, the customer's number in the complaint will be anonymized (only the first three digits and the last two digits will be displayed, and the rest of the number, such as 181******31, will not be visible).

[0089] At this point, based on known information such as the date of the call, the call details of the day, and the user's industry, the call volume model and rules are used to match and analyze the massive call detail record database. If the complaint is matched with the corresponding call detail record according to the analysis model, it indicates that the call is real, and the complaining customer's number is added to the sensitive number information database.

[0090] S4, Construct sensitive number data

[0091] like Figure 5 As shown, a database and high-speed cache capable of handling massive amounts of sensitive number information are established. The sensitive customer number information identified by intelligent analysis is written into the sensitive number information database and persistently stored by being tagged according to rules.

[0092] Meanwhile, a high-speed cache is added to the front end of the database to improve the platform's concurrent processing capabilities. When each call to the platform queries whether the called number is a sensitive number, it first accesses the high-speed cache. If the high-speed cache does not contain the sensitive number, it then accesses the sensitive number information database for matching and comparison. If a sensitive number is matched in the sensitive number information database, the query result is returned, and the sensitive number retrieved from the information database is stored in the high-speed cache for use in subsequent call queries, thereby improving query efficiency and concurrency capabilities.

[0093] S5, Establish a Bloom filter

[0094] During off-peak hours, a Bloom filter is built using multiple hash algorithms at the front end of the cache, such as... Figure 6 As shown.

[0095] like Figure 7 As shown, when a call request comes in to query and verify a sensitive customer number, the called number is used as the keyword to query the Bloom filter. Based on the algorithm characteristics of the Bloom filter, if the number does not exist in the Bloom filter, the call connection is returned directly; if the Bloom filter exists, the cache is queried, and if the cache matches the query result, it is returned; if the cache does not match the result, the sensitive number information database is queried, and the matching result is returned. At the same time, the matching result in the database is stored in the cache.

[0096] S6, Platform Outbound Call Behavior Processing

[0097] Call behavior control is implemented based on the results of Bloom filters, caches, and sensitive number database queries. This includes: for calls not found in the Bloom filter, the caller ID is changed and the call is made directly; for calls found in the Bloom filter, the call is disconnected and a prompt tone is played to inform the caller. This aims to improve the connection rate of number-changing calls, reduce customer complaint rates, and improve platform resource utilization.

[0098] like Figure 8 As shown, if the called number does not exist in the Bloom filter, it is determined that the called number is not in the sensitive number information database, and the call is made directly according to the platform's original number change process.

[0099] If the called number exists in the Bloom filter, the cache is queried again to determine if the number exists in the sensitive number information database; if the called number exists in the cache, the call is disconnected and a risk warning tone is played to the calling party; if the called number does not exist in the cache, the outbound call is continued according to the platform's original number change process.

[0100] The method provided by this invention intelligently infers the customer's answering behavior and extracts sensitive customer numbers into a sensitive number information database when a voice cloud platform makes outbound calls with a changed number, based on the "reason" field in the temporary response signaling 18x message when the called party rejects the call and the call duration after the call is answered, and then stores these numbers in the database. Simultaneously, for anonymized number information from complaint platforms, a sensitive number information database is extracted based on the enterprise user's industry, call time, and call content. Considering the platform's requirement for a large number of changed number calls, a high-speed cache and Bloom filter are added before the sensitive number information database, improving the platform's high-concurrency call processing performance, connection efficiency, and security.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for improving call connection rate, characterized in that, include: S1, Call Signaling Collection: Store the signaling that meets the preset conditions in the temporary SIP response information of the called side when making a call request into the signaling database; S2, Intelligent Reasoning and Analysis of Call Behavior: Based on the customer's answering behavior of incoming calls, intelligent reasoning and analysis are performed to obtain sensitive number information, and the analyzed sensitive number information is stored in the sensitive number information database; S3, Extraction of de-identified number information: For the de-identified number information of the complaint platform, an intelligent analysis model is established based on the known information of the calling customer. The call traffic model and the rule-based call record library are used for matching analysis. If the complaint is matched with the corresponding call record according to the intelligent analysis model, it indicates that the call is real. The complaining customer's number is extracted as sensitive number information and stored in the sensitive number information library. S4, Construct sensitive number data: Write sensitive customer number information into the sensitive number information database. At the same time, add a high-speed cache to the front end of the sensitive number information database. When querying whether the called number is a sensitive number for each call sent to the complaint platform, the high-speed cache is accessed first. If the high-speed cache does not exist, the sensitive number information database is accessed for matching and comparison. S5, Establish a Bloom filter: When the voice call platform is idle, establish a Bloom filter in front of the cache. When a call request comes in to query and verify a sensitive customer number, if the Bloom filter does not contain the called number, return to call setup directly; if the Bloom filter contains the called number, further query the cache. S6, Outbound Calling Behavior Processing of Voice Call Platform: Controlling call behavior based on the query results of Bloom filter, cache and sensitive number information database, including: for calls that do not exist in the Bloom filter, the caller ID is changed and the call is made directly; for calls that exist in the Bloom filter, the caller ID is matched with the corresponding result in the cache and sensitive number information database, the call is disconnected and a prompt tone is played to inform the calling user.

2. The method for improving call connection rate according to claim 1, characterized in that, In S1, the preset condition is: an 18X temporary response SIP message, carrying a signaling message with a reason header field indicating call rejected or user busy.

3. The method for improving call connection rate according to claim 1, characterized in that, In S2, intelligent reasoning analysis is performed based on the customer's answering behavior to incoming calls, including: Analysis rules and models are established for the signaling database. When it is detected that the voice call platform makes multiple outbound calls to the same called number and the calls are all directly hung up, it indicates that the customer is sensitive or unwilling to answer calls from unknown numbers. The called customer's number is then set as a sensitive number. Data is extracted from the call detail record (CDR) database. If the same called customer answers multiple calls from the voice call platform with a changed number and then hangs up within the first preset time period, it indicates that the customer is unwilling to answer or is not interested. The called customer's number is then set as a sensitive number.

4. The method for improving call connection rate according to claim 3, characterized in that, The first preset duration is 1 to 3 seconds.

5. The method for improving call connection rate according to claim 1, characterized in that, In S3, the known information of the calling customer includes: the calling customer's industry, service type, date of the call, and call detail record for that day.

6. The method for improving call connection rate according to claim 5, characterized in that, In step S3, when the complaint platform receives a customer's complaint, the complaint is sent back to the complaint platform for verification. At this time, the customer's phone number in the complaint has been anonymized. An intelligent analysis model is established based on the known information of the calling customer, and matching analysis is performed using the call traffic model and the rule-based dialogue record library, including: S31, determine whether the complaint content contains the complainant's number. If so, further determine whether there is a date of the call. Otherwise, proceed to step S39. S32, determine whether the complaint content includes the date of the call; if so, proceed to step S33; otherwise, proceed to step S39. S33, determine whether the complaint content includes the time of the call. If so, proceed to step S34; otherwise, proceed to step S35. S34, determine whether the incoming call time is valid. If it is, proceed to step S36; otherwise, proceed to step S35. S35, query the call details for the day. If the call details for the day are found, proceed to step S38; otherwise, proceed to step S39. S36, query call details within the second preset duration before and after the incoming call time; S37. If the call details are found within the second preset duration before and after the incoming call time, proceed to step S38; otherwise, proceed to step S35. S38, determine whether the complainant's number matches the first 3 digits and the last 2 digits. If so, it indicates that the anonymized number has been extracted. S39, it was determined that no de-identified number was extracted.

7. A method for improving call connection rate according to claim 6, characterized in that, The second preset duration is 10 minutes.

8. The method for improving call connection rate according to claim 1, characterized in that, In step S4, if a sensitive number is matched in the sensitive number information database, the query result is returned, and the sensitive number retrieved from the sensitive number information database is stored in the cache for use in subsequent call queries.

9. A method for improving call connection rate according to claim 1, characterized in that, In step S5, if the cache matches the query result, it returns the result; if the cache does not match the result, it queries the sensitive number information database and returns the matching result, while storing the matching result in the sensitive number information database into the cache.

10. A method for improving call connection rate according to claim 1, characterized in that, In S6, If the called number does not exist in the Bloom filter, it is determined that the called number is not in the sensitive number information database, and the call is made directly according to the original number change process of the voice call platform; if the called number exists in the Bloom filter, the cache is queried again to determine whether the number exists in the sensitive number information database; if the called number exists in the cache, the call is disconnected and a risk warning tone is played to the calling party. If the called number is not found in the cache, the outbound call will be continued according to the original number change process of the voice call platform.

Images