SMS interception method and system
By intercepting on the server of the SMS center, whether to intercept is determined based on the correlation between the SMS keyword and the user domain set, the problem of new users receiving irrelevant information after mobile phone number recycling is solved, and the accuracy and user experience of SMS interception are improved .
Patent Information
- Application Number
- CN202411295661.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-18
AI Technical Summary
After the mobile phone number is recycled, new users may receive sensitive information and unrelated notifications from their former users, resulting in privacy and security risks and communication experience issues.
By implementing the SMS interception method on the server of the SMS center, SMS keywords are extracted, the correlation degree with the user's domain set is calculated, and whether to intercept SMS is determined based on the preset threshold value is used to provide error correction opportunities to improve interception accuracy.
It realizes accurate interception when there are fewer SMSs that users have interacted or have not been unsubscribed, reduces server resource requirements, and is suitable for deployment in SMS centers, taking into account interception accuracy and user experience.
Smart Images

Figure CN118828518B_ABST
Abstract
Description
Background Art
[0002] Currently, users' mobile phone numbers are often bound to a variety of services, such as financial services, air travel, restaurant reservations, and various online and offline platforms. This makes mobile phone numbers an important means of identity authentication and plays an important role in a large number of daily lives and work. However, if users do not update the mobile phone numbers bound to various service agencies in a timely manner after canceling their numbers, they may encounter a series of problems. After the recycling period, the mobile phone number may be reallocated to another user by the operator. At this time, if the original service provider still sends messages to the number, it will cause two problems: first, the new user may inadvertently receive sensitive information from the previous user, thus posing a privacy and security risk; second, the new user will be constantly disturbed by notifications that are not related to him, which seriously affects his communication experience. For this reason, it is necessary to study a text message interception technology that can solve this problem. Summary of the invention
[0003] Multiple embodiments of this specification describe a text message interception method and system.
[0004] In a first aspect, an embodiment of the present specification provides a method for intercepting short messages, which is run on a server connected to a short message center and is used to intercept short messages of a receiving number of a changed user, and includes the steps of:
[0005] Step S1, extracting a number of keywords from a text message, and obtaining the domain of the text message according to the keywords;
[0006] Step S2, reading a user-related domain set, and calculating a correlation between the domain and the user-related domain set, wherein the user-related domain set is obtained based on SMS messages that the user has interacted with or has not unsubscribed;
[0007] Step S3: if the correlation is higher than a preset first threshold, forwarding the short message to a terminal corresponding to the receiving number; if the correlation is lower than a preset second threshold, intercepting the short message;
[0008] Step S4, if the correlation degree is between the preset second threshold value and the preset first threshold value, temporarily store the text message and feedback to the sending number that the text message has been delivered, wait for a preset time and monitor the receiving number, if the text message sent by the sending number is received again within the preset time, all text messages are forwarded to the terminal corresponding to the receiving number, if the receiving number does not receive the text message sent by the sending number again within the preset time, the text message is discarded.
[0009] In a second aspect, the embodiments of this specification provide a short message interception system, which is configured on a server connected to a short message center and is used to intercept short messages whose receiving numbers have been changed by a user, including:
[0010] An extraction module, extracting a number of keywords from a text message, and obtaining the domain of the text message according to the keywords;
[0011] A classification module, which reads a user-related domain set and calculates a correlation between the domain and the user-related domain set, wherein the user-related domain set is obtained based on SMS messages that the user has interacted with or has not unsubscribed;
[0012] A first interception module, which forwards the short message to a terminal corresponding to a receiving number if the correlation degree is higher than a preset first threshold, and intercepts the short message if the correlation degree is lower than a preset second threshold;
[0013] The second interception module temporarily stores the text message and feeds back to the sending number that the text message has been delivered if the correlation degree is between the preset second threshold and the preset first threshold, waits for a preset time and monitors the receiving number, and if the text message sent by the sending number is received again within the preset time, all text messages are forwarded to the terminal corresponding to the receiving number, and if the receiving number does not receive the text message sent by the sending number again within the preset time, the text message is discarded.
[0014] In a third aspect, an embodiment of this specification provides an electronic device, including a processor and a memory;
[0015] The processor is connected to the memory;
[0016] The memory is used to store executable program code;
[0017] The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in any one of the above aspects.
[0018] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any of the above aspects.
[0019] In a fifth aspect, an embodiment of this specification provides a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.
[0020] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least:
[0021] In multiple embodiments of the present specification, the provided SMS interception method obtains a user-involved domain set through SMS messages that the current user has interacted with or has not unsubscribed from. When intercepting SMS messages, it determines whether to intercept the SMS based on the comparison between the domain of the SMS and the user-involved domain set. This allows more accurate SMS interception to be provided when the number of SMS messages that the user has already been involved in is small.
[0022] It provides an error correction opportunity when the correlation is between the preset second threshold and the preset first threshold, so that the accuracy of SMS interception can be improved by combining user feedback. The interception judgment is made based on the SMS field, and the interception logic is simple to implement, with low requirements on server resources and fast processing speed. It is suitable for deployment on the server connected to the SMS center to judge whether to intercept all SMS.
[0023] The solution recorded in this embodiment only needs to match the domain with the user's domain set so that it will not be intercepted. It is suitable for new users in the initial use, and can collect user feedback through step S4, which takes into account a certain interception accuracy rate without causing large-scale SMS interception, but reducing user convenience.
[0024] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0026] Figure 1 This is a schematic diagram of the application environment architecture of the SMS interception method provided in the embodiments of this specification.
[0027] Figure 2 This is a flowchart of the SMS interception method provided in the embodiments of this specification.
[0028] Figure 3 A schematic diagram of a text message interception method provided in an embodiment of this specification.
[0029] Figure 4 A schematic diagram of a method for obtaining a text message domain provided in an embodiment of this specification.
[0030] Figure 5 A schematic diagram of another method for obtaining a text message domain provided in an embodiment of this specification.
[0031] Figure 6 A schematic diagram of a method for obtaining a user's domain set provided in an embodiment of this specification.
[0032] Figure 7 A schematic diagram of a text message interception system provided in an embodiment of this specification.
[0033] Figure 8 A schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION
[0034] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.
[0035] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0036] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.
[0037] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.
[0038] Before introducing the technical solution of this embodiment, the application scenarios and related technologies of this embodiment are introduced.
[0039] With the rapid development of information technology, various online platforms and online services have gradually covered all aspects of consumers' lives. Mobile phone numbers are not only a communication tool, but also an important part of identity authentication. It is used to receive information confirmation codes for various services, bind social media accounts, and even serve as a means of security verification for online payments. Mobile phone numbers are provided by telecom operators. Telecom operators each have different number segments. There are certain rules for the allocation of numbers. Generally, telecom operators maintain a number pool that contains all numbers that can be allocated to user 60. When user 60 applies for a new number, the operator will allocate an unused number from the number pool to user 60. In addition, the operator will strictly monitor the application and use of mobile phone numbers and implement necessary security inspection procedures, such as registering the identity information of consumers.
[0040] However, as early as when mobile phones were just becoming popular, number resources had already shown their scarcity. Due to the large population base and the rapid growth of mobile phone users, the limited number resources have become increasingly scarce. In order to meet the growing demand, telecommunications management agencies began to explore how to allocate and manage number resources more efficiently. For example, the practice of canceling numbers and waiting for a period of time before re-entering the market for new users to choose was born in this context. When a user cancels a number, the number will not be immediately reallocated to other users. Operators usually set a "freeze period" during which the number will be temporarily frozen.
[0041] Although this process has achieved the effective recycling of resources, there are still some problems that need to be solved. For example, privacy and security issues. If the former user 61 does not properly handle personal information, it may lead to privacy leakage. The new user may also receive text messages 11 or calls that originally belonged to the former user 61, which will not only cause trouble to the new user, but may also involve the responsibility definition of the service provider. For example, there have been cases where new users received SMS reminders of bank account balances from the former user 61, and there have even been cases involving the automatic login of the social media account of the former user 61. These problems expose management loopholes and technical difficulties in the number cancellation process. For this reason, it is necessary to conduct in-depth research on technologies that ensure safe use during the reuse of mobile phone numbers.
[0042] Among them, SMS interception is a technology that ensures the recycling of mobile phone numbers. Through certain technologies, it is identified whether SMS 11 is sent to the previous user 61 or the current user 61. When it is identified that SMS 11 is sent to the previous user 61, SMS 11 is intercepted, thereby reducing the possibility of the current user 61 receiving irrelevant information, and assisting in ensuring the normal use of the current user 61. At present, the following possible research directions have been proposed in the field of SMS interception technology in this regard: keyword filtering, sender identification, behavior pattern analysis, and user feedback mechanism. However, technical research in this area is still in its infancy, and there is no mature technology available. For example, a machine learning algorithm is used to analyze the sending mode of SMS 11 to identify and intercept SMS, such as frequency, time period and other characteristics. However, its implementation is relatively complex and requires a large amount of data to train the model. In addition, there is a certain delay, and it may not be accurate enough in the early stage.
[0043] To this end, the present specification provides a method and system for intercepting short messages, which are used to intercept short messages 11 from circularly used numbers.
[0044] Please refer to the attached Figure 1, is a schematic diagram of the use scenario of the SMS interception technology disclosed in this application. During the SMS service process, the user's mobile phone at the sending end 40 sends the short message to the SMS center 10 through the base station 30 and the gateway. The SMS center 10 is a core component in the operator's network, responsible for receiving, storing and forwarding SMS 11, and all SMS 11 need to pass through the SMS center 10. Then the SMS center 10 sends the SMS to the base station 30 and the gateway where the user's mobile phone at the receiving end 50 is located, and finally received by the receiving end 50. Spam SMS can be intercepted on the user's mobile phone, base station 30, mobile switching center and SMS center 10. Installing SMS firewall software on the user's mobile phone can intercept SMS 11, but the disadvantage is that the user is not very motivated to install SMS firewall software, and it is difficult to ensure coverage. Deploying SMS interception systems in base stations 30 and mobile switching centers can cover all SMS 11 for effective interception. However, the base station 30 and mobile switching center systems are complex and the tasks are arduous. The system also undertakes other important functions, and the hardware equipment is often composed of different models from multiple manufacturers. Deployment and maintenance are time-consuming and labor-intensive, and it is easy to cause system instability and high risk. Therefore, the technical solution disclosed in this specification adopts the solution of setting up an interception system in the SMS center 10, and the server 20 where the interception system is located is independent of the SMS center 10. The server 20 where the interception system is located exchanges data with the SMS center 10. Before the SMS center 10 sends the SMS to the downlink, it submits the SMS to the interception system, and the interception system decides whether to intercept it. When intercepting the SMS, the interception instruction is fed back to the SMS center 10. The server 20 where the interception system is located can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The sending end 40 and the receiving end 50 can be smart phones, tablet computers, laptops, PDAs, personal computers, smart speakers, smart TVs, smart watches, vehicle-mounted devices, wearable devices, etc., but are not limited to this.
[0045] Since this application involves some professional terms, these professional terms will be introduced below.
[0046] SMS Center
[0047] The Short Message Service Center (SMSC) is an important component of the mobile communication network. It is mainly responsible for managing and controlling short message services. The SMS center plays a core role in the entire SMS delivery process. Its main responsibilities include: receiving SMS messages. When a user sends a SMS message 11, the SMS message 11 is first sent to the SMS center 10. Storing SMS messages. If the recipient is temporarily unable to receive SMS messages (for example, the recipient's mobile phone is turned off or out of service area), the SMS center 10 will temporarily store the SMS messages until the recipient reconnects to the network. Forwarding SMS messages. After the SMS center 10 confirms that the recipient can receive SMS messages, it forwards the SMS messages to the recipient's mobile phone. Status reporting. The SMS center 10 can provide reports on the SMS transmission status, such as "sent", "delivered" or "undelivered". Managing SMS messages, including managing the validity period of SMS messages, prioritizing, etc. Error handling. Handling various error situations that may occur during the sending process and taking corresponding measures, such as retrying to send.
[0048] field
[0049] Fields refer to specific topics or industries that users frequently come into contact with or are interested in in their daily lives and work. These fields reflect information about users' hobbies, professional backgrounds, living habits, etc. For example: the field of games includes electronic games, tabletop games, chess games, etc. The field of entertainment includes entertainment activities such as movies, music, drama, and variety shows. The field of sports includes various sports such as football, basketball, and tennis. The field of science and technology includes computer science, artificial intelligence, and Internet technology. The field of education includes online courses, education and training, and academic research. The field of tourism includes travel guides, attraction introductions, travel agency information, etc. The field of finance includes the stock market, investment and financial management, and banking services.
[0050] User-related domain set
[0051] The user's domain set represents the set of domains involved by the user. It represents the set of various domains that the user touches within the scope of his or her activities or interests, and this set constitutes part of the user portrait. The user's domain set helps to more accurately portray the user's personal preferences and professional background, thereby helping to make judgments on SMS interception and improving service effects. The user's domain set is useful for understanding user behavior patterns.
[0052] Large Language Model
[0053] Large Language Model (LLM) is a large-scale, multi-parameter artificial intelligence model, which is usually trained through deep learning technology. This type of model is designed to understand and generate human language and can handle a wide range of natural language tasks12. The characteristics of large language models21 include but are not limited to having billions or more parameters, using large-scale corpora for training, and having strong contextual understanding and text generation capabilities.
[0054] Natural language tasks
[0055] Natural language tasks refer to those designed to enable computers to understand and generate human natural language. These tasks cover a wide range of aspects, from simple part-of-speech tagging to complex dialogue generation, such as text classification, named entity recognition, machine translation, question-answering systems, text summarization, text generation, dialogue systems, semantic understanding, etc. Example 1
[0056] This specification first provides a method for intercepting short messages, which is run on a server 20 connected to a short message center 10, and is used to intercept short messages whose receiving numbers have been changed by a user. Figure 2 And attached Figure 3 , including the steps of:
[0057] Step S1, extracting a plurality of keywords from the short message 11, and obtaining the domain of the short message 11 according to the plurality of keywords.
[0058] In step S1, see the attached Figure 4 , extracting several keywords from the short message 11, and obtaining the field of the short message 11 according to the several keywords includes:
[0059] According to the response of the pre-connected large language model 21 to the natural language task 12 of extracting the keywords of the text message 11, a plurality of keywords are obtained;
[0060] Comparing a plurality of the keywords with a pre-configured domain reference table to obtain the domain of the short message 11;
[0061] The field reference table records a number of fields and a number of corresponding example keywords.
[0062] For example, a credit card bill notification SMS 11 is as follows: [XX Bank] Dear Mr. XX, your credit card bill has been issued. The amount due for this period is RMB 2345.00, and the minimum payment amount is RMB 235.00. The due date for repayment is October 10, 2024. Please repay on time to avoid late payment fees. If you have any questions, please log in to online banking or call customer service. Please reply to TD to cancel.
[0063] The key words extracted through the large language model 21 include: credit card, bill, amount due, minimum repayment amount, due date, late payment fee, online banking, and customer service.
[0064] Table 1 shows an exemplary domain reference table. The domain reference table records example keywords related to the gaming field, the entertainment field, the sports field, the science and technology field, the education field, the tourism field, and the finance field. By comparing the keywords of the SMS 11 extracted by the large language model 21 with the example keywords of the fields recorded in the domain reference table, the field that the SMS 11 best matches can be obtained.
[0065] Table 1 Field reference table used in this embodiment
[0066]
[0067] By comparing the keywords extracted from the text message: credit card, bill, amount due, minimum repayment amount, due date, late payment fee, online banking, customer service, with the sample keywords of each field in the field reference table, it can be found that the field of the text message is the financial field.
[0068] On the other hand, in another embodiment, please refer to the attached Figure 5 In step S1, a plurality of keywords are extracted from a text message, and a method for obtaining the field of the text message according to the plurality of keywords includes:
[0069] According to the response of the pre-connected large language model 21 to the natural language task 12 of extracting the keywords of the text message, a plurality of keywords are obtained;
[0070] A natural language task 12 for obtaining a domain based on a number of the keywords is generated and input into the large language model 21 to obtain the domain of the text message.
[0071] Similarly, taking the aforementioned credit card bill notification SMS as an example, after the keywords are extracted by the large language model 21, a natural speech model of the domain is generated based on the keywords credit card, bill, amount to be repaid, minimum repayment amount, due date, late payment fee, online banking, and customer service, and then input into the large language model 21, and the response obtained is "financial domain". Figure 4 and attached Figure 5 In the two technical solutions, Figure 4 The fields obtained must be within the scope of the fields listed in the field reference table. Figure 5In the embodied technical solution, the obtained fields are open. For example, based on the keywords: credit card, bill, online banking, the obtained fields may be the financial field or the financial field. Considering that the financial field and the financial field are highly correlated, it is very easy to obtain, so this situation does not affect the implementation of the technical solution disclosed in this specification.
[0072] Step S2: read the user's domain set, and calculate the correlation between the domain and the user's domain set, wherein the user's domain set is obtained based on the SMS messages that the user has interacted with or has not unsubscribed.
[0073] The method for obtaining a user-involved domain set based on text messages that the user has interacted with or has not unsubscribed includes: reading text messages that the user has interacted with; extracting several keywords from each text message respectively, and obtaining the domain of the text message based on several of the keywords; and the domains of several of the text messages constitute the user-involved domain set.
[0074] The recommended implementation method is to periodically obtain the user's domain set based on the text messages that the user has interacted with or has not unsubscribed. For example, the user's domain set may be updated every half month or every month. The user's domain set may also be updated in the following periodic manner: in the first year after the mobile phone number is registered, the user's domain set is obtained based on the text messages that the user has interacted with or has not unsubscribed, and then the user's domain set is obtained based on the text messages that the user has interacted with or has not unsubscribed within one year, using a yearly cycle.
[0075] For example, the user received the following notification SMS and did not unsubscribe:
[0076] SMS 1: Dear Mr. A, your credit card was used to purchase RMB 1,200 at [Merchant Name] at 15:30 today. If you have any questions, please call the customer service hotline [Customer Service Phone Number]. To cancel, please reply to TD.
[0077] SMS 2: Dear student A, hello! The autumn painting class you signed up for will officially start at 9:00 am on September 10 (Monday). Please bring your personal painting supplies and attend on time. Location: [address]. Looking forward to your arrival!
[0078] SMS 3: Dear Mr. A, the stock [stock code] [stock name] you hold has increased by 10% today and is currently trading at [price] yuan. Please pay attention to market trends and changes in your personal investment portfolio. Please log in to [platform name] for details. Please reply to TD to unsubscribe.
[0079] SMS 4: Dear Mr. A, your hotel reservation has been confirmed successfully. Check-in date: September 15; Check-out date: September 17; Hotel name: [hotel name]; Room type: [room type]. Have a pleasant trip! Please reply to TD to cancel the reservation.
[0080] SMS 5: Dear Mr. A, your order for "Introduction to Oil Painting Techniques" has been successfully placed, and the order number is [order number]. The estimated delivery time is 3 working days, and the courier will contact you after arrival. Thank you for your purchase! Please reply TD to cancel the order.
[0081] The user received the following text message and interacted with it.
[0082] Interactive SMS 6: Dear Mr. A, hello! This is [Securities Company Name]. We noticed that the stock [Stock Code] [Stock Name] you hold has performed well recently. Do you want to learn more about the professional analysis of this stock? Reply "Y" to get more information, and reply "N" to ignore this message.
[0083] The user replies with the text message as follows: Y.
[0084] Then I received another text message: Dear customer, thank you for your reply. According to our analysis, [stock name] has risen by 15% in the past month and the industry outlook is promising. However, the stock market is risky and investment should be cautious. It is recommended that you pay attention to the company's latest financial report and industry trends. If you need further consulting services, please reply "consult" and we will arrange for a professional consultant to contact you. I wish you a smooth investment!
[0085] Interactive SMS 7: [Game Name] Dear player, congratulations on winning the exclusive game package! It contains rare equipment, 5,000 gold coins, and experience potions. Please log in to the game and claim it on the event page before September 10. Reply "Y" to get the package immediately, and reply "N" to give up this opportunity.
[0086] User replies text message: Y.
[0087] Then I received another text message: [Game Name] Thank you for your reply! The gift package has been successfully sent to your game account. Please log in to the game and go to your mailbox to claim it. I wish you a happy game! For more activities, please pay attention to the in-game announcement or visit the official website [Official Website Link].
[0088] Since SMS 2 does not support unsubscription, SMS 2 does not participate in the generation of the user's domain set. Other SMS messages participate in the generation of the user's domain set. By obtaining the domain of each SMS message, it can be concluded that the domain of SMS 1 is the financial domain, the domain of SMS 3 is the financial domain, the domain of SMS 4 is the tourism domain, the domain of SMS 5 is the education domain, the domain of interactive SMS 6 is the financial domain, and the domain of interactive SMS 7 is the game domain. Therefore, the user's domain set is obtained as {finance domain, tourism domain, education domain, game domain}.
[0089] SMS 1 to be determined whether to be intercepted:
[0090] Dear fans, hello! [Celebrity Name] will hold a fan meeting in [City Name] on the 10th of this month. There will be opportunities to get autographed photos and interact with fans up close. Ticket pre-sale will start soon. Please follow our official social media for more information [Official Social Media Link].
[0091] SMS 2 to be determined whether to be intercepted:
[0092] Dear user, hello! The new flagship phone is now on the market, equipped with the latest processor, supports ultra-fast charging technology, and brings the ultimate photo-taking experience. Welcome to visit the official website to make an appointment to purchase, or go to offline stores to experience it [official website link].
[0093] SMS 3 to be determined whether to be intercepted:
[0094] [XX Travel Agency] Dear tourists, hello! We recommend you to go to the beautiful [destination city], where you can enjoy charming natural scenery and rich cultural experience. For more travel information, please visit [official website] or reply "details" to get more information. I wish you a happy life!
[0095] By obtaining the fields of the text messages respectively, it can be obtained that the fields of the text messages 1 to 3 to be determined whether to be intercepted are the entertainment field, the technology field, and the tourism field respectively.
[0096] The method for calculating the degree of association between the field and the field set involved by the user includes:
[0097] Calculate the relevance between the field and each field in the field set involved by the user respectively;
[0098] The maximum value of the degree of association between the user-related field set and the field is obtained, and the maximum value is used as the degree of association between the field and the user-related field set.
[0099] That is, the correlation between the entertainment field, technology field, tourism field and the user's field set {finance field, tourism field, education field, game field} is calculated respectively.
[0100] Specifically, the correlation between the entertainment industry field of the SMS 1 to be determined whether to be intercepted and each field in the user's field set is as follows:
[0101] The correlation between the entertainment industry and the financial industry is 0.1, the correlation between the entertainment industry and the tourism industry is 0.4, the correlation between the entertainment industry and the education industry is 0.2, and the correlation between the entertainment industry and the gaming industry is 0.5. The maximum value is 0.5, that is, the correlation between the entertainment industry and the user's domain set is 0.5. The correlation range is [0,1].
[0102] The correlation between the technology field of the SMS 2 to be determined whether to be intercepted and each field in the user's field set is as follows:
[0103] The correlation between the technology field and the financial field is 0.7, the correlation between the technology field and the tourism field is 0.3, the correlation between the technology field and the education field is 0.6, and the correlation between the technology field and the game field is 0.8. The maximum value is 0.8, that is, the correlation between the entertainment field and the set of fields involved by users is 0.8.
[0104] The correlation degree between the tourism field of the text message 3 to be determined whether to be intercepted and the user's field set is 1, because it is included in the user's field set.
[0105] On the other hand, in another embodiment, the following improved scheme is adopted, and the method for calculating the degree of association between the field and the field set involved by the user includes:
[0106] respectively calculating the modified relevance between the field and each field in the field set involved by the user;
[0107] The maximum value of the corrected relevance between the user-related field set and the field is obtained, and the maximum value is used as the relevance between the field and the user-related field set.
[0108] The method for calculating the modified correlation between the field and the field in the field set involved by the user includes:
[0109] Calculate the correlation between the field and the field selected in the field set involved by the user, and record it as a first correlation;
[0110] Respectively calculating the correlation between the selected fields in the field set involved by the user and other fields in the field set involved by the user to obtain a plurality of second correlations;
[0111] After the plurality of second correlation degrees are multiplied by the first correlation degree respectively, the second correlation degrees are added to the first correlation degree to obtain a modified correlation degree, wherein the maximum value of the modified correlation degree is limited to 1.
[0112] Exemplarily, the user's domain set is {finance domain, tourism domain, education domain, game domain}. According to this improved solution, the process of calculating the correlation between the domain and the user's domain set is as follows:
[0113] The correlation between the field of the text message 1 to be determined whether to be intercepted, the entertainment field, and each field in the user's field set is as follows:
[0114] Calculate the modified correlation between the entertainment industry and each field in the user's field set:
[0115] First, calculate the corrected correlation between the entertainment industry and the financial industry. The process is as follows:
[0116] The correlation between the entertainment industry and the financial industry is: 0.1;
[0117] The correlation between the financial field and the tourism field is 0.4, the correlation between the financial field and the education field is 0.2, and the correlation between the financial field and the gaming field is 0.3.
[0118] Therefore, the corrected correlation between the entertainment industry and the financial industry is: 0.1+0.1×0.4+0.1×0.2+0.1×0.3=0.19.
[0119] Then calculate the corrected correlation between the entertainment industry and the tourism industry. The process is as follows:
[0120] The correlation between the entertainment industry and the tourism industry is: 0.4;
[0121] The correlation between the tourism field and the financial field is 0.4, the correlation between the tourism field and the education field is 0.2, and the correlation between the tourism field and the gaming field is 0.3.
[0122] Therefore, the corrected correlation between the entertainment industry and the tourism industry is: 0.4+0.4×0.4+0.4×0.2+0.4×0.3=0.76.
[0123] Then calculate the corrected correlation between the entertainment industry and the education industry. The process is as follows:
[0124] The correlation between the entertainment industry and education is: 0.2;
[0125] The correlation between the education field and the financial field is 0.1, the correlation between the education field and the tourism field is 0.2, and the correlation between the education field and the gaming field is 0.3.
[0126] Therefore, the corrected correlation between the entertainment industry and the education industry is: 0.2+0.2×0.1+0.2×0.2+0.2×0.3=0.32.
[0127] Then calculate the corrected correlation between the entertainment industry and the gaming industry. The process is as follows:
[0128] The correlation between the entertainment industry and the gaming industry is 0.5;
[0129] The correlation between the gaming field and the financial field is 0.3, the correlation between the gaming field and the tourism field is 0.3, and the correlation between the gaming field and the education field is 0.3.
[0130] Therefore, the corrected correlation between the entertainment industry and the gaming industry is: 0.5+0.5×0.3+0.5×0.3+0.5×0.3=0.95.
[0131] Finally, the corrected correlation between the entertainment industry and the financial sector is 0.19, the corrected correlation between the entertainment industry and the tourism sector is 0.76, the corrected correlation between the entertainment industry and the education sector is 0.32, and the corrected correlation between the entertainment industry and the gaming sector is 0.95.
[0132] That is, the correlation between the entertainment industry field finally obtained according to the corrected correlation and the field set involved by the user is 0.95. Its significance lies in that when the fields involved in the user's field set increase, the obtained correlation will become larger and larger after being corrected by the corrected correlation. Thus, the solution recorded in this specification is implemented to intercept more text messages when the user has interacted with or has not unsubscribed with fewer text messages. When the user has interacted with or has not unsubscribed with enough text messages, it is smoothly switched to the text message interception technology based on machine learning, thereby realizing the combination with the text message connection technology based on machine learning. When there is a lack of samples required for machine learning, the text messages are intercepted through the solution recorded in this specification. When the samples are sufficient, the solution recorded in this application will be less inclined to intercept the user's text messages due to the effect of the corrected correlation. Thus, the problem of low interception accuracy caused by the lack of samples in the early stage when machine learning technology intercepts text messages is made up.
[0133] Step S3: if the correlation degree is higher than a preset first threshold, the text message is forwarded to a terminal corresponding to the receiving number; if the correlation degree is lower than a preset second threshold, the text message is intercepted.
[0134] In this embodiment, it is generally required that the text message be directly included in the user's domain set, so the first threshold should be set to a higher value. In this embodiment, the first threshold is set to 0.9. Of course, it can also be set to 0.85, 0.95, etc. Setting it to 0.8 will result in some text messages that should be intercepted not being intercepted, but since the correlation with the user's domain set is high enough (reaching 0.8), it is generally believed that the user is also more interested, so it can also be set to 0.8, allowing some text messages in the field that have not yet been included in the user's domain set to be sent to the user.
[0135] The second threshold is recommended to be set to a value in the interval [0.6, 0.7]. Exemplarily, this embodiment is set to 0.6. Therefore, the correlation between the field of SMS 1 to be determined whether to be intercepted and the set of fields involved by the user is 0.5, which is lower than the second threshold, so it will be directly intercepted. The correlation between the field of SMS 3 to be determined whether to be intercepted and the set of fields involved by the user is 1, which is higher than the first threshold, so it will be directly forwarded to the terminal corresponding to the receiving number.
[0136] Step S4, if the correlation degree is between the preset second threshold value and the preset first threshold value, temporarily store the text message and feedback to the sending number that the text message has been delivered, wait for a preset time and monitor the receiving number, if the text message sent by the sending number is received again within the preset time, all text messages are forwarded to the terminal corresponding to the receiving number, if the receiving number does not receive the text message sent by the sending number again within the preset time, the text message is discarded.
[0137] The correlation between the field of SMS 2 to be determined whether to be intercepted and the user-involved field set is 0.8, which is between the second threshold value 0.6 and the first threshold value 0.9. Therefore, wait for the preset time length and monitor the receiving number. If the SMS sent by the sending number is received again within the preset time length, both SMS messages will be forwarded to the terminal corresponding to the receiving number. It is worth noting that after both SMS messages are forwarded to the terminal corresponding to the receiving number, they will be considered when the user-involved field set is generated next time. Of course, if the user unsubscribes or does not interact, it will not be included in the scope of consideration for the next generation of the user-involved field set. If the receiving number does not receive the SMS sent by the sending number again within the preset time length, the SMS 2 to be determined whether to be intercepted is discarded.
[0138] When a text message is received again within a preset time, both text messages are forwarded to the terminal corresponding to the receiving number because the party sending the text message is unaware that the text message has been temporarily intercepted, that is, the party sending the text message assumes that the receiving number has received the text message. Therefore, the second text message is usually sent because the user has sent a message again, visited a website again, and other operations have triggered the sending of the second text message. Therefore, both text messages are forwarded to the terminal corresponding to the receiving number. Of course, if the party sending the text message directly sends two identical text messages in a short period of time, the interception scheme recorded in this embodiment will be bypassed, but a large number of such text message sending behaviors will be identified by other interception schemes that have been disclosed in the art, and the text message sending authority of the party sending the text message will generally be suspended. Therefore, this situation is not considered in this specification.
[0139] This embodiment can be used in conjunction with the technology disclosed in the art for implementing SMS interception using machine learning algorithms, and has a relatively better effect. Since machine learning requires more samples to ensure accuracy, when the number of samples in the early stage is insufficient, the technical solution recorded in this embodiment is adopted. Although the technical solution recorded in this embodiment also requires some SMS messages that have been interacted with or not unsubscribed, the required sample size is very small. Even if there are no existing SMS messages, step S4 can be executed by setting the second threshold to 0 for a period of time, thereby obtaining some SMS messages that can be used as samples, and after the number of SMS messages that have been interacted with or not unsubscribed reaches a preset value, the second threshold is set to a recommended value (such as 0.6 to 0.7). Thereby, a better SMS interception effect is obtained. When the number of SMS messages that have been interacted with or not unsubscribed is large enough, the accuracy of SMS interception achieved by machine learning technology is sufficient. At this time, this implementation scheme can no longer be used, and the SMS interception provided by machine learning technology can be used instead, thereby achieving a consistently better SMS interception effect.
[0140] The new technical effects achieved by this embodiment include: obtaining the user's domain set through the SMS that the current user has interacted with or not unsubscribed, and when intercepting SMS, judging whether to intercept the SMS according to the comparison between the domain of the SMS and the domain set involved by the user, and being able to provide more accurate SMS interception when the number of SMS involved by the user is small. It has the opportunity to provide error correction when the correlation is between the preset second threshold and the preset first threshold, so that the accuracy of SMS interception can be improved by combining user feedback. The judgment of interception by the domain of SMS is simple to implement, has low resource requirements for the server 20, and has a fast processing speed. It is suitable for being deployed on the server 20 connected to the SMS center 10 to judge whether to intercept all SMS. The scheme recorded in this embodiment only needs to match the domain set involved by the user, that is, it will not be intercepted, which is suitable for new users in the initial use, and can collect user feedback through step S4, that is, it takes into account a certain interception accuracy, and will not cause large-scale SMS interception, but reduce the convenience of user use.
[0141] Example 2
[0142] The difference between this embodiment and embodiment 1 is that this embodiment provides an improved solution for obtaining the domain set involved by the user. Figure 6 , the methods for obtaining the user's domain set include:
[0143] Step S201, collect the records of SMS messages that the user has interacted with or has not unsubscribed, and extract the text of the SMS message and the sending number. Here, it is necessary to establish a database of the sending number so that the sender name corresponding to the sending number can be obtained by combining multiple SMS messages. The sender name can be obtained by extracting the content enclosed in brackets in the SMS message.
[0144] Sample SMS 1: Dear Mr. A, hello! This is [Securities Company Name]. We noticed that the stock [Stock Code] [Stock Name] you hold has performed well recently. Would you like to learn more about the professional analysis of this stock? Reply "Y" to get more information, and reply "N" to ignore this message. The [Securities Company Name] can be used to obtain the sender's name.
[0145] Step S202: extract a number of keywords from the text, identify the topic of the text based on the keywords, and obtain a number of candidate fields based on the topic and a pre-configured field reference set.
[0146] The keywords extracted from the example SMS 1 are [stock code], [stock name], professional analysis, and good performance, and the topic is identified as stocks. In this embodiment, the pre-configured field reference set includes the game field, the entertainment field, the sports field, the technology field, the education field, the tourism field, and the finance field. By comparing the stock with the field reference set, it is easy to obtain several candidate fields including the technology field, the education field, and the finance field.
[0147] Step S203: Try to extract the sender's name from the text, and obtain the sender's industry according to the extracted sender's name and the sending number. The sender's industry can be obtained as the securities industry through the sender's name [Securities Company Name].
[0148] Step S204: Screen the candidate fields according to the sender's industry, and obtain a user-related field set based on the screened candidate fields. The rest is the same as in Embodiment 1.
[0149] By screening the candidate fields through the securities industry, the science and technology fields and the education fields can be screened out. Thus, the user-related field set obtained from the example SMS 1 includes the financial field. When multiple SMS messages perform the same steps, more fields will be added to the user-related field set.
[0150] The method of screening a plurality of candidate fields according to the sender industry includes: respectively calculating the correlation between the sender industry and the candidate fields;
[0151] The candidate fields whose relevance is lower than a preset threshold are deleted. The relevance between the sender's industry and the candidate fields can be directly calculated by the large language model 21 .
[0152] The correlation between the sender industry and the candidate field can also be calculated by the following method:
[0153] The industry correlation between the sender's industry and the associated industry of the candidate field is obtained, and the correlation between the sender's industry and the candidate field is obtained according to the industry correlation.
[0154] The pre-configured field reference set records a number of fields and related industries of the fields, and the user-related field set is a subset of the field reference set.
[0155] The new technical effects achieved in this embodiment are: by screening the alternative fields in combination with the sender's industry, the accuracy of the user's field set can be improved, which helps to intercept SMS more accurately, improves the convenience of user use, and helps to protect the privacy and security of the previous user 61.
[0156] Example 3
[0157] The difference between this embodiment and embodiment 1 is that before step S1, the following steps are further performed:
[0158] Read the name reserved by the current user for the receiving number;
[0159] Try to extract the name in the text message. If the name does not match the reserved name, intercept the text message and add the sending number of the text message to the blacklist. If the name matches the reserved name, forward the text message to the terminal corresponding to the receiving number. If the name is not extracted, go to step S1. The rest is the same as in Example 1.
[0160] The title can quickly identify the SMS to be intercepted. For example, the current user is called Mr. C. When you receive a text message: Dear Mr. A, hello! This is [Securities Company Name]. We noticed that the stock [Stock Code] [Stock Name] you hold has performed well recently. Do you want to know more about the professional analysis of this stock? Reply "Y" to get more information, and reply "N" to ignore this message. The [Securities Company Name] can get the sender's title.
[0161] Since the title Mr. A in the text message does not match the title Mr. C of the current user, the text message can be directly intercepted.
[0162] The new technical effects achieved by this improved embodiment include: quickly intercepting text messages by extracting the names, thereby improving the efficiency of text message interception.
[0163] Example 4
[0164] The difference between this embodiment and embodiment 1 is that between step S3 and step S4, the following steps are further included:
[0165] If the association degree is between the preset second threshold and the preset third threshold, feedback information is sent to the sending number, requesting that the first registration time of the receiving number be included in the text message, waiting for a preset time and monitoring the receiving number, and if no text message sent by the sending number is received again within the preset time, the text message is discarded;
[0166] If a text message with the first registration time of the receiving number is received and sent again by the sending number, determining whether the first registration time is earlier than the registration date;
[0167] If the first registration time is earlier than the registration date, the short message is intercepted, and if the first registration time is not earlier than the registration date, the process proceeds to step S4. The rest is the same as in embodiment 1.
[0168] The third threshold is greater than the second threshold and less than the first threshold. For example, the third threshold can be taken as 0.7. Exemplarily, when the correlation is 0.6, feedback information is sent to the sending number, requiring the first registration time of the receiving number to be attached in the text message. If the device corresponding to the sending number does not process the feedback information, the text message sent by the sending number will not be received again within the preset time. At this time, the text message is discarded, that is, the text message is intercepted. On the contrary, if the device corresponding to the sending number processes the feedback information, that is, the first registration time is attached as required, the first registration time refers to the earliest time when the receiving number is registered on the device, platform, or system corresponding to the sending number. If the first registration time is earlier than the registration date, it indicates that the registration behavior was performed by the previous user 61, so the text message is intercepted. On the contrary, if the first registration time is not earlier than the registration date, it indicates that the registration behavior was performed by the current user, so the text message is not intercepted.
[0169] The new technical effects achieved by this embodiment include: by attempting to send feedback information to the sending number requiring the first registration time of the receiving number to be included in the text message, if the relevant first registration time is obtained, it is possible to more quickly and accurately determine whether the text message should be intercepted, and it is possible to accurately determine whether the sending number needs to be added to the blacklist, providing reference information for subsequent text message interception.
[0170] On the other hand, this specification provides a short message interception system, which is configured on a server 20 connected to a short message center 10, and is used to intercept short messages of a receiving number of a changed user. Figure 7 ,include:
[0171] An extraction module 100 extracts a plurality of keywords from a text message, and obtains the domain of the text message according to the plurality of keywords;
[0172] The classification module 200 reads a user-related domain set and calculates a correlation between the domain and the user-related domain set, wherein the user-related domain set is obtained based on short messages that the user has interacted with;
[0173] The first interception module 300 forwards the short message to the terminal corresponding to the receiving number if the correlation degree is higher than a preset first threshold, and intercepts the short message if the correlation degree is lower than a preset second threshold;
[0174] The second interception module 400, if the correlation degree is between the preset second threshold and the preset first threshold, temporarily stores the text message and feeds back to the sending number that the text message has been delivered, waits for a preset time and monitors the receiving number, if the text message sent by the sending number is received again within the preset time, all text messages are forwarded to the terminal corresponding to the receiving number, if the receiving number does not receive the text message sent by the sending number again within the preset time, the text message is discarded.
[0175] See also Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification is shown.
[0176] like Figure 8 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. Among them, the communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. Among them, the user interface 1103 may include a button, and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface 1104 may include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. Among them, the processor 1101 may include one or more processing cores. The processor 1101 uses various interfaces and lines to connect various parts in the entire electronic device 1100, and executes various functions and processes data of the routing device 1100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1101 may integrate one or a combination of CPU, GPU and modem, etc. Among them, the CPU mainly processes the operating system, user interface and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication.
[0177] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.
[0178] Among them, the memory 1105 may include RAM or ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1105 may also be at least one storage device located away from the aforementioned processor 1101. The memory 1105 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application. The processor 1101 may be used to call the application stored in the memory 1105 and execute the methods in the above-mentioned multiple embodiments.
[0179] The embodiments of this specification also provide a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the multiple steps in the above embodiments. If the components of the above electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.
[0180] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.
[0181] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.
[0182] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of this specification is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server 20 or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server 20 or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server 20, a data center, etc. that includes multiple available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)), etc.
[0183] When implemented by hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, and its logic function is determined by the user programming the device. The designer programs by himself to "integrate" a digital system on a PLD, without asking a chip manufacturer to design and make a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing program development, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and HDL is not just one, but many. Those skilled in the art should also be aware that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages and program it into the integrated circuit to easily obtain the hardware circuit that implements the logic method flow.
[0184] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.
Claims
1. A method for intercepting short messages, which runs on a server connected to a short message center, and is used to intercept short messages whose receiving numbers have been changed by a user, characterized in that: Includes steps: Step S1, extracting a number of keywords from a text message, and obtaining the domain of the text message according to the keywords; Step S2, reading a user-related domain set, and calculating a correlation between the domain and the user-related domain set, wherein the user-related domain set is obtained based on SMS messages that the user has interacted with or has not unsubscribed; Step S3: if the correlation is higher than a preset first threshold, forwarding the short message to a terminal corresponding to the receiving number; if the correlation is lower than a preset second threshold, intercepting the short message; Step S4, if the correlation degree is between the preset second threshold value and the preset first threshold value, temporarily storing the text message and feeding back to the sending number that the text message has been delivered, waiting for a preset time and monitoring the receiving number, if the text message sent by the sending number is received again within the preset time, forwarding all the text messages to the terminal corresponding to the receiving number, if the receiving number does not receive the text message sent by the sending number again within the preset time, discarding the text message; The method for calculating the degree of association between the domain and the domain set involved by the user includes: respectively calculating the modified relevance between the field and each field in the field set involved by the user; Obtaining a maximum value of the corrected relevance between the user-related domain set and the domain, and using the maximum value as the relevance between the domain and the user-related domain set; The method for calculating the modified correlation between the field and the field concentrated with the user includes: Calculate the correlation between the field and the field selected in the field set involved by the user, and record it as a first correlation; Respectively calculating the correlation between the selected fields in the field set involved by the user and other fields in the field set involved by the user to obtain a plurality of second correlations; After the plurality of second correlation degrees are multiplied by the first correlation degree respectively, the second correlation degrees are added to the first correlation degree to obtain a modified correlation degree, wherein the maximum value of the modified correlation degree is limited to 1.
2. A method for intercepting short messages according to claim 1, characterized in that: The following steps are also included, and the following steps are performed before step S1: Read the name reserved by the current user for the receiving number; Try to extract the name in the text message. If the name does not match the reserved name, intercept the text message and add the sending number of the text message to the blacklist. If the name matches the reserved name, forward the text message to the terminal corresponding to the receiving number. If the name is not extracted, enter step S1 to execute.
3. A method for intercepting short messages according to claim 1 or 2, characterized in that: In step S1, a method of extracting a plurality of keywords from a text message and obtaining the field of the text message according to the plurality of keywords includes: Obtaining a plurality of keywords according to a response of a pre-connected large language model to a natural language task of extracting keywords from the text message; Comparing a plurality of the keywords with a pre-configured domain reference table to obtain the domain of the short message; The field reference table records a number of fields and a number of corresponding example keywords; or, In step S1, a method of extracting a plurality of keywords from a text message and obtaining the field of the text message according to the plurality of keywords includes: Obtaining a plurality of keywords according to a response of a pre-connected large language model to a natural language task of extracting keywords from the text message; A natural language task for obtaining a domain based on a number of the keywords is generated and input into the large language model to obtain the domain of the text message.
4. A method for intercepting short messages according to claim 1 or 2, characterized in that: Methods for obtaining the user's domain set include: Collect records of SMS messages that users have interacted with or not unsubscribed from, and extract the text of the SMS messages and the sending number; Extracting a number of keywords from the text, identifying topics of the text based on the keywords, and obtaining a number of candidate fields based on the topics and a pre-configured field reference set; Try to extract the sender's name from the text, and obtain the sender's industry based on the extracted sender's name and the sending number; Screening a number of candidate fields according to the sender's industry, and obtaining a user-related field set according to the screened number of candidate fields; The pre-configured field reference set records a number of fields and related industries of the fields, and the user-related field set is a subset of the field reference set; The method of screening the candidate fields according to the sender industry includes: Calculate the relevance between the sender industry and the candidate fields respectively; Deleting the candidate fields whose correlation is lower than a preset threshold; The method for calculating the correlation between the sender industry and the candidate field includes: The industry correlation between the sender's industry and the associated industry of the candidate field is obtained, and the correlation between the sender's industry and the candidate field is obtained according to the industry correlation.
5. A method for intercepting short messages according to claim 1 or 2, characterized in that: Between step S3 and step S4, the method further includes the following steps: If the association degree is between the preset second threshold and the preset third threshold, feedback information is sent to the sending number, requesting that the first registration time of the receiving number be included in the text message, waiting for a preset time and monitoring the receiving number, and if no text message sent by the sending number is received again within the preset time, the text message is discarded; If a text message with the first registration time of the receiving number is received and sent again by the sending number, determining whether the first registration time is earlier than the registration date; If the first registration time is earlier than the registration date, the text message is intercepted; if the first registration time is not earlier than the registration date, the process proceeds to step S4 for execution.
6. A short message interception system, configured on a server connected to a short message center, for intercepting short messages whose receiving numbers have been changed by a user, characterized in that: include: An extraction module, extracting a number of keywords from a text message, and obtaining the domain of the text message according to the keywords; A classification module, reading a user-related domain set, and calculating a correlation between the domain and the user-related domain set, wherein the user-related domain set is obtained based on SMS messages that the user has interacted with or has not unsubscribed; A first interception module, which forwards the short message to a terminal corresponding to a receiving number if the correlation degree is higher than a preset first threshold, and intercepts the short message if the correlation degree is lower than a preset second threshold; The second interception module temporarily stores the text message and feeds back to the sending number that the text message has been delivered if the correlation degree is between the preset second threshold and the preset first threshold, waits for a preset time and monitors the receiving number, and forwards all text messages to the terminal corresponding to the receiving number if the text message sent by the sending number is received again within the preset time, and discards the text message if the receiving number does not receive the text message sent by the sending number again within the preset time; The method for calculating the degree of association between the domain and the domain set involved by the user includes: respectively calculating the modified relevance between the field and each field in the field set involved by the user; Obtaining a maximum value of the corrected relevance between the user-related domain set and the domain, and using the maximum value as the relevance between the domain and the user-related domain set; The method for calculating the modified correlation between the field and the field concentrated with the user includes: Calculate the correlation between the field and the field selected in the field set involved by the user, and record it as a first correlation; Respectively calculating the correlation between the selected fields in the field set involved by the user and other fields in the field set involved by the user to obtain a plurality of second correlations; After the plurality of second correlation degrees are multiplied by the first correlation degree respectively, the second correlation degrees are added to the first correlation degree to obtain a modified correlation degree, wherein the maximum value of the modified correlation degree is limited to 1.
7. An electronic device, characterized in that including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 5.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
