A junk information elimination system and method based on Chinese word segmentation recognition technology

By combining Chinese word segmentation technology, sensitive vocabulary and intelligent model in the spam recognition system, the problem of single recognition form in the existing technology is solved, the accuracy and efficiency of spam recognition are improved, and the intelligent terminal is ensured. Security.

CN112380323BActive Publication Date: 2024-11-05HEFEI D2S INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011391134.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-01
Publication Date
2024-11-05
Estimated Expiration
2040-12-01

AI Technical Summary

Technical Problem

The prior art has a single form when identifying spam messages, making it difficult to effectively identify spam messages of multiple arrangement rules, reducing practicality and accuracy.

Method used

The spam information removal system based on Chinese word segmentation recognition technology is adopted, and the initial screening is carried out through the SMS pre-processing module, and triple detection is carried out in combination with sensitive thesaurus and intelligent models, including short signal code matching, keyword matching and neural network model judgment to ensure the spam text messages. Accurate identification.

Benefits of technology

It improves the accuracy and work efficiency of spam information identification, enhances the ability to identify spam messages for multiple arrangement rules, and ensures that smart terminals are not intruded by spam messages.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112380323B_ABST
    Figure CN112380323B_ABST
Patent Text Reader

Abstract

The invention discloses a spam information elimination system and method based on Chinese word segmentation recognition technology, relates to the technical field of spam information identification, and solves the technical problems of low spam text message identification rate and low working efficiency in the prior art solutions; the invention first preliminarily screens text messages received by an intelligent terminal according to numbers from which the text messages are sent to obtain preliminarily screened text messages, then extracts verification keywords in the text messages through Chinese word segmentation technology, matches the verification keywords with a sensitive word library, and finally determines through an intelligent model; triple detection is set up, which is helpful to improve the accuracy of spam information identification and the working efficiency of the invention; the invention sets up a text message preprocessing module, which is helpful to improve the spam text message elimination efficiency in the invention; the invention sets up a text message analysis module, which is helpful to improve the recognition rate of spam text messages and ensure that the intelligent terminal is not disturbed by spam text messages.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of junk information recognition, and relates to Chinese word segmentation recognition technology, in particular to a junk information elimination system and method based on the Chinese word segmentation recognition technology. Background Art

[0002] As a basic service of mobile communication network, short message service (SMS) not only provides users with convenient message communication services, but also becomes a channel for sending some illegal SMS messages, causing many harms, such as using SMS to commit crimes such as extortion and fraud, and using SMS to spread false information and rumors.

[0003] The invention patent with publication number CN103874033A discloses a method for identifying irregular spam text messages based on Chinese word segmentation. For the same text message, Chinese word segmentation is first performed according to normal horizontal reading according to the content of the text message, and the weight is calculated according to the number of words in the word segmentation result; based on the characteristic that the number of characters in each line of text messages must be controlled for irregular text messages, the range of the irregular text message content is determined, and the characters within the range of the irregularly arranged text message content are converted vertically into horizontal arrangement, and then Chinese word segmentation is performed, and the weight is calculated according to the number of words in the overall word segmentation result; then, based on the comparison of the two weights, it is determined whether the text message is a normally arranged text message or an irregularly arranged text message.

[0004] The above scheme finally uses content analysis to match keywords according to the type of arrangement to identify whether it is a spam text message, thereby avoiding the omission of spam text messages and improving the recall and precision rates of spam text messages; however, the above scheme is mainly aimed at irregularly arranged spam text messages, resulting in a single form of identifying text messages, which reduces the practicality of the above scheme; therefore, the above scheme still needs further improvement. Summary of the invention

[0005] In order to solve the problems existing in the above scheme, the present invention provides a junk information elimination system and method based on Chinese word segmentation recognition technology.

[0006] The object of the present invention can be achieved by the following technical scheme: a spam information elimination system based on Chinese word segmentation recognition technology, comprising a processor, an IP analysis module, an information publishing module, a data storage module, a short message preprocessing module, an intelligent model module and a short message analysis module;

[0007] The SMS preprocessing module is used to preprocess the SMS received by the smart terminal to obtain the preliminary screening SMS, and send the preliminary screening SMS to the SMS analysis module through the processor;

[0008] The SMS analysis module analyzes the pre-screened SMS in turn through the intelligent model and keyword analysis technology, and screens out junk SMS according to the analysis results, and at the same time sends the IP analysis signal to the IP analysis module through the processor;

[0009] The intelligent model module is used to obtain an intelligent model;

[0010] The IP analysis module is used to analyze the IP addresses of spam text messages.

[0011] Preferably, the SMS preprocessing module is used to perform preliminary screening of SMS, including:

[0012] After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; the smart terminal includes a smart phone and a tablet computer;

[0013] After receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS and obtains the SMS tag database stored in the storage module through the processor;

[0014] Match the sending number with the numbers in the SMS database. When a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal. When no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor.

[0015] The sending record of the short message analysis signal is sent to the data storage module for storage through the processor.

[0016] Preferably, the SMS analysis module is used to analyze the preliminary selected SMS, including:

[0017] Acquiring, by a processor, a sensitive word library in a data storage module; the sensitive word library includes at least one keyword of a sensitive word type, and the sensitive word type includes drugs and pornography;

[0018] The verification keywords are extracted from the initial screening SMS through Chinese word segmentation technology, and the verification keywords are matched with the keywords in the sensitive word library. When the verification keywords match the results in the sensitive word library, the initial screening is determined to be junk SMS and automatically removed from the smart terminal; when the verification keywords cannot match the results in the sensitive word library, the intelligent model in the data storage module is obtained;

[0019] The initial screening SMS is converted into an input array and marked as a verification input array, and the verification input array is input into the intelligent model to judge the initial screening SMS;

[0020] When the initially screened text messages are determined to be spam, they are automatically removed from the smart terminal.

[0021] Preferably, the intelligent model module is used to train a neural network model to obtain an intelligent model, including:

[0022] Obtain a database of spam messages through the Internet, and number the spam messages; the specific number is a 5-digit number, such as 1+01+00; the position of 1 represents the arrangement rule of the spam messages, and the arrangement rule includes horizontal and vertical arrangement; the position of 01 represents the type of sensitive words, and 00 represents a mixed type, 01 represents drugs, and 02 represents pornography; the position of 00 represents the number of sensitive words;

[0023] After preprocessing, the junk text messages are converted into an input array of a neural network model, and the numbers corresponding to the junk text messages are used as the output array of the neural network to train the neural network model; the neural network model includes an error feedforward neural network and an RBF neural network;

[0024] The trained neural network model is marked as an intelligent model, and the intelligent model is sent to the data storage module for storage through the processor.

[0025] Preferably, the IP analysis module analyzes the IP address of the spam message after receiving the IP analysis signal, and when the number of spam messages sent by the IP address exceeds a preset number of spam messages, the IP address is added to the IP blacklist.

[0026] Preferably, the information publishing module is used to publish the results of removing spam text messages, and regularly publish the records of removing spam text messages to the smart terminal.

[0027] Preferably, the SMS tag database is generated through a third-party platform, including:

[0028] Generate an empty SMS tag library through the processor;

[0029] Obtain a harassing number statistics table through a third-party platform; the third-party platform includes China Mobile, China Unicom and China Telecom, and the numbers in the harassing number statistics table are numbers marked as harassing calls by users of the third-party platform;

[0030] Get the number of times the number is marked in the harassment number statistics table, and mark the number of times marked as BC;

[0031] When the number of marking times BC>L1, the number corresponding to the number of marking times is stored in the SMS marking library; where L1 is the preset number of marking times threshold, L1>0;

[0032] The short message tag library is sent to the data storage module for storage through the processor.

[0033] Preferably, the processor is communicatively connected to the IP analysis module, the information publishing module, the data storage module, the SMS preprocessing module, the intelligent model module and the SMS analysis module respectively, and the data storage module is communicatively connected to the information publishing module.

[0034] A method for removing junk information based on Chinese word segmentation recognition technology, the junk information removal method comprising the following steps:

[0035] Step 1: After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; after receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS, and obtains the SMS tag database stored in the storage module through the processor; the sending number is matched with the number in the SMS database, and when a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal; when no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor;

[0036] Step 2: Obtain the sensitive word library in the data storage module through the processor; extract the verification keyword from the initial screening text message through the Chinese word segmentation technology, match the verification keyword with the keyword in the sensitive word library, and when the verification keyword matches the result in the sensitive word library, it is determined that the initial screening is a spam text message and is automatically removed from the smart terminal; when the verification keyword does not match the result in the sensitive word library, obtain the intelligent model in the data storage module; convert the initial screening text message into an input array and mark it as a verification input array, and input the verification input array into the intelligent model to judge the initial screening text message; when the initial screening text message is determined to be a spam text message, it is automatically removed from the smart terminal.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. The present invention first preliminarily screens the short messages received by the smart terminal according to the number from which the short messages are sent to obtain the pre-screened short messages, then extracts the verification keywords in the short messages through the middle word segmentation technology, matches the verification keywords with the sensitive word library, and finally determines through the intelligent model; setting up triple detection is helpful to improve the accuracy of spam information identification and the working efficiency of the present invention;

[0039] 2. The present invention sets up a short message preprocessing module, which is used to perform preliminary screening of short messages; after receiving the short message, the smart terminal sends the short message to the short message preprocessing module; after receiving the short message, the short message preprocessing module obtains the sending number of the short message, and obtains the short message tag database stored in the storage module through the processor; the sending number is matched with the number in the short message database, and when the result is matched, the short message corresponding to the sending number is intercepted and automatically removed from the smart terminal; when the result is not matched, the short message is marked as a preliminary screening short message, and the preliminary screening short message and the short message analysis signal are sent to the short message analysis module through the processor; the short message preprocessing module realizes preliminary screening of the short message by identifying the number that sends the short message, which helps to improve the efficiency of removing junk short messages in the present invention;

[0040] 3. The present invention sets up a text message analysis module, which is used to analyze the preliminary selected text messages; obtain the sensitive word library in the data storage module through the processor; extract the verification keyword from the preliminary screening text messages through the Chinese word segmentation technology, match the verification keyword with the keyword in the sensitive word library, and when the verification keyword matches the result in the sensitive word library, it is determined that the preliminary screening is a spam text message and is automatically removed from the smart terminal; when the verification keyword cannot match the result in the sensitive word library, the intelligent model in the data storage module is obtained; the preliminary screening text message is converted into an input array and marked as a verification input array, and the verification input array is input into the intelligent model to judge the preliminary screening text message; when the preliminary screening text message is determined to be a spam text message, it is automatically removed from the smart terminal; the text message analysis module further screens the preliminary screening text messages through the Chinese word segmentation technology and the intelligent model in turn, and automatically removes the screened spam text messages, which helps to improve the recognition rate of spam text messages and ensure that the smart terminal is not disturbed by spam text messages. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a schematic diagram of the principle of the present invention. DETAILED DESCRIPTION

[0043] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 , a spam information elimination system based on Chinese word segmentation recognition technology, including a processor, an IP analysis module, an information publishing module, a data storage module, a short message preprocessing module, an intelligent model module and a short message analysis module;

[0045] The SMS preprocessing module is used to preprocess the SMS received by the smart terminal to obtain the preliminary screening SMS, and send the preliminary screening SMS to the SMS analysis module through the processor;

[0046] The SMS analysis module analyzes the initial screening SMS in turn through the intelligent model and keyword analysis technology, and filters out the junk SMS according to the analysis results. At the same time, the IP analysis signal is sent to the IP analysis module through the processor;

[0047] The intelligent model module is used to obtain the intelligent model;

[0048] The IP analysis module is used to analyze the IP addresses of spam messages.

[0049] Furthermore, the SMS pre-processing module is used to perform preliminary screening of SMS, including:

[0050] After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; the smart terminal includes smart phones and tablet computers;

[0051] After receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS and obtains the SMS tag database stored in the storage module through the processor;

[0052] Match the sending number with the numbers in the SMS database. When a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal. When no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor.

[0053] The sending record of the short message analysis signal is sent to the data storage module for storage through the processor.

[0054] Furthermore, the SMS analysis module is used to analyze the preliminary selected SMS, including:

[0055] Acquire a sensitive word library in the data storage module through a processor; the sensitive word library includes at least one keyword of a sensitive word type, and the sensitive word types include drugs and pornography;

[0056] The verification keywords are extracted from the initial screening SMS through Chinese word segmentation technology, and the verification keywords are matched with the keywords in the sensitive word library. When the verification keywords match the results in the sensitive word library, the initial screening is determined to be junk SMS and automatically removed from the smart terminal; when the verification keywords cannot match the results in the sensitive word library, the intelligent model in the data storage module is obtained;

[0057] The initial screening SMS is converted into an input array and marked as a verification input array, and the verification input array is input into the intelligent model to judge the initial screening SMS;

[0058] When the initially screened text messages are determined to be spam, they are automatically removed from the smart terminal.

[0059] Furthermore, the intelligent model module is used to train the neural network model to obtain the intelligent model, including:

[0060] Obtain a database of spam messages through the Internet and number the spam messages; the specific number is a 5-digit number, such as 1+01+00; the position of 1 represents the arrangement rule of spam messages, which includes horizontal and vertical arrangement; the position of 01 represents the type of sensitive words, and 00 represents mixed types, 01 represents drugs, and 02 represents pornography; the position of 00 represents the number of sensitive words;

[0061] After preprocessing, the junk text messages are converted into an input array of a neural network model, and the numbers corresponding to the junk text messages are used as the output array of the neural network to train the neural network model; the neural network model includes an error feedforward neural network and an RBF neural network;

[0062] The trained neural network model is marked as an intelligent model, and the intelligent model is sent to the data storage module for storage through the processor.

[0063] Furthermore, after receiving the IP analysis signal, the IP analysis module analyzes the IP address of the junk text message, and when the number of junk text messages sent by the IP address exceeds a preset number of junk text messages, the IP address is added to the IP blacklist.

[0064] Furthermore, the information publishing module is used to publish the results of removing junk text messages and regularly publish the records of removing junk text messages to the smart terminal.

[0065] Furthermore, the SMS tag database is generated through a third-party platform, including:

[0066] Generate an empty SMS tag library through the processor;

[0067] Obtain a statistical table of harassing numbers through a third-party platform; the third-party platforms include China Mobile, China Unicom and China Telecom. The numbers in the statistical table of harassing numbers are those marked as harassing calls by users of the third-party platform;

[0068] Get the number of times the number is marked in the harassment number statistics table, and mark the number of times marked as BC;

[0069] When the number of marking times BC>L1, the number corresponding to the number of marking times is stored in the SMS marking library; where L1 is the preset number of marking times threshold;

[0070] The short message tag library is sent to the data storage module for storage through the processor.

[0071] Furthermore, the processor is respectively connected in communication with the IP analysis module, the information publishing module, the data storage module, the SMS preprocessing module, the intelligent model module and the SMS analysis module, and the data storage module is connected in communication with the information publishing module.

[0072] A method for removing junk information based on Chinese word segmentation recognition technology, the junk information removal method comprising the following steps:

[0073] Step 1: After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; after receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS, and obtains the SMS tag database stored in the storage module through the processor; the sending number is matched with the number in the SMS database, and when a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal; when no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor;

[0074] Step 2: Obtain the sensitive word library in the data storage module through the processor; extract the verification keyword from the initial screening text message through the Chinese word segmentation technology, match the verification keyword with the keyword in the sensitive word library, and when the verification keyword matches the result in the sensitive word library, it is determined that the initial screening is a spam text message and is automatically removed from the smart terminal; when the verification keyword does not match the result in the sensitive word library, obtain the intelligent model in the data storage module; convert the initial screening text message into an input array and mark it as a verification input array, and input the verification input array into the intelligent model to judge the initial screening text message; when the initial screening text message is determined to be a spam text message, it is automatically removed from the smart terminal.

[0075] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0076] Working principle of the present invention:

[0077] After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; after receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS, and obtains the SMS tag database stored in the storage module through the processor; the sending number is matched with the number in the SMS database, and when a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal; when no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor;

[0078] The sensitive word library in the data storage module is obtained through the processor; the verification keyword is extracted from the initial screening text message through the Chinese word segmentation technology, and the verification keyword is matched with the keyword in the sensitive word library. When the verification keyword matches the result in the sensitive word library, the initial screening is determined to be a spam text message and is automatically removed from the smart terminal; when the verification keyword cannot match the result in the sensitive word library, the intelligent model in the data storage module is obtained; the initial screening text message is converted into an input array and marked as a verification input array, and the verification input array is input into the intelligent model to judge the initial screening text message; when the initial screening text message is determined to be a spam text message, it is automatically removed from the smart terminal.

[0079] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0080] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A spam elimination system based on Chinese word segmentation recognition technology, characterized in that: It includes a processor, an IP analysis module, an information publishing module, a data storage module, a SMS preprocessing module, an intelligent model module and an SMS analysis module; The SMS preprocessing module is used to preprocess the SMS received by the smart terminal to obtain the preliminary screening SMS, and send the preliminary screening SMS to the SMS analysis module through the processor; The SMS analysis module analyzes the pre-screened SMS in turn through the intelligent model and keyword analysis technology, and screens out junk SMS according to the analysis results, and at the same time sends the IP analysis signal to the IP analysis module through the processor; The intelligent model module is used to obtain an intelligent model; The IP analysis module is used to analyze the IP address of spam text messages; The SMS pre-processing module is used to perform preliminary screening of SMS, including: After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; the smart terminal includes a smart phone and a tablet computer; After receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS and obtains the SMS tag database stored in the storage module through the processor; Match the sending number with the numbers in the SMS database. When a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal. When no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor. The processor sends the sending record of the short message analysis signal to the data storage module for storage; The SMS tag database is generated through a third-party platform, including: Generate an empty SMS tag library through the processor; Obtain a statistical table of harassing numbers through a third-party platform; the third-party platforms include China Mobile, China Unicom and China Telecom. The numbers in the statistical table of harassing numbers are those marked as harassing calls by users of the third-party platform; Get the number of times the number is marked in the harassment number statistics table, and mark the number of times marked as BC; When the number of marking times BC>L1, the number corresponding to the number of marking times is stored in the SMS marking library; where L1 is the preset number of marking times threshold; Sending the SMS tag library to the data storage module for storage through the processor; The SMS analysis module is used to analyze the preliminary selected SMS, including: Acquiring, by a processor, a sensitive word library in a data storage module; the sensitive word library includes at least one keyword of a sensitive word type, and the sensitive word type includes drugs and pornography; The verification keywords are extracted from the initial screening SMS through Chinese word segmentation technology, and the verification keywords are matched with the keywords in the sensitive word library. When the verification keywords match the results in the sensitive word library, the initial screening is determined to be junk SMS and automatically removed from the smart terminal; when the verification keywords cannot match the results in the sensitive word library, the intelligent model in the data storage module is obtained; The initial screening SMS is converted into an input array and marked as a verification input array, and the verification input array is input into the intelligent model to judge the initial screening SMS; When the initially screened SMS is judged as junk SMS, it will be automatically removed from the smart terminal; The intelligent model module is used to train the neural network model to obtain the intelligent model, including: Obtain a spam SMS database through the Internet and number the spam SMS; the number is a 5-digit number, the first digit of which represents the arrangement rule of the spam SMS, the second and third digits represent the type of sensitive words, and the fourth and fifth digits represent the number of sensitive words; After preprocessing, the junk text messages are converted into an input array of a neural network model, and the numbers corresponding to the junk text messages are used as the output array of the neural network to train the neural network model; the neural network model includes an error feedforward neural network and an RBF neural network; The trained neural network model is marked as an intelligent model, and the intelligent model is sent to the data storage module for storage through the processor.

2. According to claim 1, a spam elimination system based on Chinese word segmentation recognition technology is characterized in that: The IP analysis module analyzes the IP address of the junk text message after receiving the IP analysis signal. When the number of junk text messages sent by the IP address exceeds a preset number of junk text messages, the IP address is added to the IP blacklist.

3. According to claim 1, a spam elimination system based on Chinese word segmentation recognition technology is characterized in that: The information publishing module is used to publish the results of removing junk text messages and regularly publish the records of removing junk text messages to the smart terminal.

4. A method for removing junk information based on Chinese word segmentation recognition technology, based on the operation of a junk information removal system based on Chinese word segmentation recognition technology as claimed in any one of claims 1 to 3, characterized in that: The junk information elimination method comprises the following steps: Step 1: After receiving the SMS, the smart terminal sends it to the SMS preprocessing module; after receiving the SMS, the SMS preprocessing module obtains the sending number of the SMS, and obtains the SMS tag database stored in the storage module through the processor; the sending number is matched with the number in the SMS database, and when a match is found, the SMS corresponding to the sending number is intercepted and automatically removed from the smart terminal; when no match is found, the SMS is marked as a preliminary screening SMS, and the preliminary screening SMS and SMS analysis signal are sent to the SMS analysis module through the processor; Step 2: Obtain the sensitive word library in the data storage module through the processor; extract the verification keyword from the initial screening text message through the Chinese word segmentation technology, match the verification keyword with the keyword in the sensitive word library, and when the verification keyword matches the result in the sensitive word library, it is determined that the initial screening is a spam text message and is automatically removed from the smart terminal; when the verification keyword does not match the result in the sensitive word library, obtain the intelligent model in the data storage module; convert the initial screening text message into an input array and mark it as a verification input array, and input the verification input array into the intelligent model to judge the initial screening text message; when the initial screening text message is determined to be a spam text message, it is automatically removed from the smart terminal.