CN115544508A - Computer system security management system and method based on big data - Google Patents

Abstract

The present invention relates to the technical field of computer security management, specifically a computer system security management system and method based on big data, including: an access management module, a data upload and use module, a database, a regular detection module and a feedback module; through the access The management module registers the visitor's identity information and sets open permissions, and sends the confirmed identity information to the database and the data upload and use module; manages the upload and use of data through the data upload and use module ; Store all identity information and uploaded and used data through the database; regularly detect the computer software and hardware through the regular detection module, and send the detection results to the feedback module; display the detection through the feedback module The result and judge whether the data is abnormal; further solve the security problem of the campus computer system.

Description

A computer system security management system and method based on big data

technical field

The invention relates to the technical field of computer security management, in particular to a computer system security management system and method based on big data.

Background technique

Computer system refers to the computer hardware, software and network system used for database management. It is characterized by accurate and fast calculation and judgment, good versatility, easy to use, and can be connected into a network. With the rapid development of the network, computer technology has become an indispensable course for college students. Therefore, the campus computer room has also become an indispensable part of campus construction. However, there are still certain hidden dangers when people use computers.

Since most of the objects faced on campus are students, it is necessary to avoid the appearance of bad web pages and bad advertisements. In addition, the emergence of viruses may cause the entire computer system to be paralyzed, which will cause a very significant impact. Therefore, the computer system on campus Security management still needs to be strengthened.

Therefore, people need a computer system security management system and method based on big data to solve the above problems. By using intelligent audit model and information filtering technology to audit and filter information, thereby reducing the occurrence of bad information; by using software simulation to scan There is no way to identify and kill viruses in the computer, so as to avoid the invasion of viruses.

Contents of the invention

The purpose of the present invention is to provide a computer system security management system and method based on big data, so as to solve the problems raised in the above-mentioned background technology.

In order to solve the above technical problems, the present invention provides the following technical solutions: a computer system security management system based on big data, the system includes: an access management module, a data upload and use module, a database, a regular detection module and a feedback module;

Registering the identity information of the visitor through the access management module and setting open permissions, and sending the confirmed identity information to the database and the data upload and use module;

Manage the upload and use of data through the data upload and use module;

Store all identity information and upload and use data through said database;

Regularly detect computer software and hardware through the regular detection module, and send the detection result to the feedback module;

The detection result is displayed through the feedback module to determine whether the data is abnormal, and if abnormal, an alarm is issued.

Further, the access management module includes an identity registration unit and an authority setting unit; the identity registration unit is used to register the identity information of the visitor, and confirm whether the visitor is a member of the school, and if the visitor is a member of the school, the The identity information is uploaded to the database, if it does not belong to the school member, then stop the visit; the authority setting unit is used to open different permissions of the computer according to the different identities of the visitors, and the open computer permissions specifically include if the identity of the visitor is confirmed If the visitor is a teacher, the teacher access mode is turned on; if the visitor is confirmed to be a student, the student access mode is turned on.

Further, the data upload and use module includes an upload management unit, an information filtering unit and a data backup unit; the upload management unit is used to detect whether the file uploaded by the visitor complies with the campus management norms and whether the file contains a virus; The information filtering unit is used to filter bad information encountered by visitors in the process of using data by using information filtering technology; the data backup unit is used to back up the data uploaded and used by visitors;

The upload management unit includes a content review subunit and a virus detection subunit; the content review subunit is used to review the uploaded file content using an intelligent review model to determine whether it meets the campus management norms, and if it meets the campus management norms, then When uploading a file, if it does not comply with the campus management norms, it will display that the upload fails; the virus detection subunit is used to detect whether the uploaded file data contains a virus by using a string-based search method.

Further, the database also includes a virus database and a sensitive word database; the virus database is used to store the characteristic strings of possible computer viruses; the sensitive word database is used to store sensitive word information that does not comply with campus management norms.

Further, the regular detection module includes a time control unit, a hardware detection unit and a software detection unit; the time control unit controls the startup time of the hardware detection unit and the software detection unit by setting a time valve; the The hardware detection unit is used to detect whether the computer hardware function can be used normally; the software detection unit is used to detect viruses in the computer system;

The software detection unit includes a virus recognition subunit, a virus killing subunit and a virus alarm subunit; the virus recognition subunit is used to identify viruses in the computer system; the virus killing subunit is used to clean up the virus The virus identified by the identification subunit; the virus alarm subunit is used for alarming the virus that cannot be removed, and sending the signal to the feedback module.

Further, the feedback module displays the detection result of the periodic detection module, and judges the result, and if it is abnormal, performs alarm processing.

A computer system security management method based on big data, comprising the following steps:

S1: Register the visitor's identity information and set open permissions;

S2: Manage, detect and back up the data uploaded and used by visitors;

S3: store all identity information and uploaded and used data in the database;

S4: Regularly inspect the software and hardware of the computer system;

S5: Display the detection result and judge whether the data is abnormal, and alarm if abnormal.

Further, in step S1: register the identity information of the visitor, and judge whether the visitor belongs to the school member according to the member information recorded in the database, and if the visitor belongs to the school member, open the corresponding authority according to his identity information, if If you are not a member of the school, stop accessing.

Further, in step S2: In order to detect whether the files uploaded by the visitors conform to the campus management norms, use the intelligent audit model to review the uploaded data, if they conform to the campus management norms, then upload the files, if they do not conform to the campus management norms , the upload failed:

S201: First calculate the similarity between the keyword in the file content and the sensitive word information in the sensitive word database:

Use the word embedding algorithm to map the uploaded file content into a numerical vector space, establish a coordinate system, and process the uploaded file information to obtain vector set A, and obtain vector set B by processing the sensitive lexicon information , set γ as the similarity between vector set A and vector set B, at this time

If the γ value is greater than the set threshold m, it means that the sensitive words in the file content are similar to the sensitive words in the sensitive lexicon, and at this time it does not conform to the campus management norms; otherwise, it conforms to the campus management norms;

S202: Determine whether the content of the file complies with the campus management norms, if so, upload the file, if not, stop uploading the file, and display upload failure;

At the same time, in order to detect whether the uploaded file data contains a virus, a search method based on a character string is used to detect according to the virus database. The search method based on a character string belongs to the conventional technical means of those skilled in the art, therefore, there is no Make too many repetitions.

In order to prevent visitors from encountering bad information in the computer during the use of data, information filtering technology is used to filter:

S203: First extract the feature values of all words in the web page based on the vector space model:

Set the text set formed in the webpage as R, the i-th vocabulary in the text as R _i , and set the training text set in the database as M, then the logarithmic mutual information logI(R _i ,R) is:

Among them, P(R _i |R) represents the proportion of vocabulary R _i in the text set R, and p(R _i ) represents the proportion of vocabulary R _i in the training text set M; The amount of information is sorted, and some higher-ranked words are selected as feature items;

S204: Next, calculate the similarity between the extracted feature item and the information in the sensitive lexicon:

Set the selected feature item as W _i , and form a set W, β is the similarity between the feature item and the sensitive lexicon information:

S205: Then by setting the threshold n, determine whether the web page information is bad information, if it is greater than the set threshold n, it means that the web page contains bad information, otherwise, it means that the web page meets the requirements; and clear the web page containing bad information , leaving the normal use of the page.

Further, in step S4: by installing a time valve in the system to control the time to start the computer software and hardware detection, a time period is set in the time valve, such as 24 hours, at this time, it means that the computer system is checked every 24 hours A detection of software and hardware;

After the time valve is opened, then enter the hardware detection mode and the software detection mode; The hardware detection mode detects whether the computer hardware function can be used normally by starting a series of computer detection tools, and the computer detection tools include CPU-Z for detection processing HD TUNE is used to detect the hard disk transfer rate, hard disk capacity, health status, temperature and disk surface scanning; Super PI is used to detect CPU stability, etc.; the software detection mode is used to detect computer viruses by using the software simulation scanning method Identify, and use the anti-virus software installed in the computer to check and kill the identified virus. The software simulation scanning method belongs to the conventional technical means of those skilled in the art, so it is not repeated too much; if the anti-virus software cannot Thoroughly clean up the identified virus, enter the alarm system, and send the signal to the feedback module.

Further, in step S5: display system detection results, including hardware function detection results, software virus detection results, if the result is normal, then no alarm processing, if the result is abnormal, then remind and alarm.

Compared with the prior art, the beneficial effects achieved by the present invention are:

By registering the identity information, the present invention effectively avoids the login of the computer by personnel who do not belong to the school; by setting the authority, it is more conducive to the use of the computer by different people; by using the intelligent audit model to audit data, it is more conducive to detection and uploading Whether the files conform to the campus management regulations; by using the string-based search method to detect whether the files contain viruses, effectively avoiding virus intrusion caused by file uploads; Bad information in the computer; by setting the time valve to control the time of hardware detection and software detection, the computer can be cleaned automatically on a regular basis, which is more conducive to avoiding virus intrusion; by timely alarming and processing of viruses that cannot be cleaned, it is more conducive to work Personnel find problems, thereby reducing the problem of computer system paralysis due to the emergence of viruses.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the attached picture:

Fig. 1 is a structural diagram of a computer system security management system based on big data of the present invention;

Fig. 2 is a flowchart of a computer system security management method based on big data in the present invention.

detailed description

The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

Referring to Fig. 1-Fig. 2, the present invention provides a technical solution: a computer system security management system based on big data, the system includes: an access management module, a data upload and use module, a database, a regular detection module and a feedback module;

Registering the identity information of the visitor through the access management module and setting open permissions, and sending the confirmed identity information to the database and the data upload and use module;

Manage the upload and use of data through the data upload and use module;

Store all identity information and upload and use data through said database;

Regularly detect computer software and hardware through the regular detection module, and send the detection result to the feedback module;

The detection result is displayed through the feedback module to determine whether the data is abnormal, and if abnormal, an alarm is issued.

Further, the access management module includes an identity registration unit and an authority setting unit; the identity registration unit is used to register the identity information of the visitor, and confirm whether the visitor is a member of the school, and if the visitor is a member of the school, the The identity information is uploaded to the database, if it does not belong to the school member, then stop the visit; the authority setting unit is used to open different permissions of the computer according to the different identities of the visitors, and the open computer permissions specifically include if the identity of the visitor is confirmed If the visitor is a teacher, the teacher access mode is turned on; if the visitor is confirmed to be a student, the student access mode is turned on.

Further, the data upload and use module includes an upload management unit, an information filtering unit and a data backup unit; the upload management unit is used to detect whether the file uploaded by the visitor complies with the campus management norms and whether the file contains a virus; The information filtering unit is used to filter bad information encountered by visitors in the process of using data by using information filtering technology; the data backup unit is used to back up the data uploaded and used by visitors;

The upload management unit includes a content review subunit and a virus detection subunit; the content review subunit is used to review the uploaded file content using an intelligent review model to determine whether it meets the campus management norms, and if it meets the campus management norms, then If the uploading of the file does not comply with the campus management norms, the uploading failure will be displayed; the virus detection subunit is used to detect whether the uploaded file data contains a virus by using a string-based search method.

Further, the database also includes a virus database and a sensitive word database; the virus database is used to store the characteristic strings of possible computer viruses; the sensitive word database is used to store sensitive word information that does not comply with campus management norms.

Further, the regular detection module includes a time control unit, a hardware detection unit and a software detection unit; the time control unit controls the startup time of the hardware detection unit and the software detection unit by setting a time valve; the The hardware detection unit is used to detect whether the computer hardware function can be used normally; the software detection unit is used to detect viruses in the computer system;

The software detection unit includes a virus recognition subunit, a virus killing subunit and a virus alarm subunit; the virus recognition subunit is used to identify viruses in the computer system; the virus killing subunit is used to clean up the virus The virus identified by the identification sub-unit; the virus alarm sub-unit is used to alarm the virus that cannot be cleared, and send the signal to the feedback module, and it is more conducive to timely alarm processing for the virus that cannot be cleaned. The staff finds problems, thereby reducing the problem of computer system paralysis due to the emergence of viruses.

Further, the feedback module displays the detection result of the periodic detection module, and judges the result, and if it is abnormal, performs alarm processing.

A computer system security management method based on big data, comprising the following steps:

S1: Register the visitor's identity information and set open permissions;

S2: Manage, detect and back up the data uploaded and used by visitors;

S3: store all identity information and uploaded and used data in the database;

S4: Regularly inspect the software and hardware of the computer system;

S5: Display the detection result and judge whether the data is abnormal, and alarm if abnormal.

Further, in step S1: register the identity information of the visitor, and judge whether the visitor belongs to the school member according to the member information recorded in the database, and if the visitor belongs to the school member, open the corresponding authority according to his identity information, and By setting permissions, it is more beneficial for different people to use the computer. If they are not members of the school, access will be stopped.

Further, in step S2: In order to detect whether the files uploaded by the visitors conform to the campus management norms, use the intelligent audit model to review the uploaded data, if they conform to the campus management norms, then upload the files, if they do not conform to the campus management norms , the upload failed:

S201: First calculate the similarity between the keyword in the file content and the sensitive word information in the sensitive word database:

Use the word embedding algorithm to map the uploaded file content into a numerical vector space, establish a coordinate system, and process the uploaded file information to obtain vector set A, and obtain vector set B by processing the sensitive lexicon information , set γ as the similarity between vector set A and vector set B, at this time

If the γ value is greater than the set threshold m, it means that the sensitive words in the file content are similar to the sensitive words in the sensitive lexicon, and at this time it does not conform to the campus management norms; otherwise, it conforms to the campus management norms;

S202: Determine whether the content of the file complies with the campus management norms, if so, upload the file, if not, stop uploading the file, and display upload failure;

At the same time, in order to detect whether the uploaded file data contains viruses, the search method based on character strings is used to detect according to the virus database, thereby effectively avoiding the problem of virus intrusion caused by the upload of file data. The string search method belongs to the conventional technical means of those skilled in the art, therefore, too many details are not given here.

In order to prevent visitors from encountering bad information in the computer during the use of data, information filtering technology is used to filter:

S203: First extract the feature values of all words in the web page based on the vector space model:

Set the text set formed in the webpage as R, the i-th vocabulary in the text as R _i , and set the training text set in the database as M, then the logarithmic mutual information logI(R _i ,R) is:

Among them, P(R _i |R) represents the proportion of vocabulary R _i in the text set R, and P(R _i ) represents the proportion of vocabulary R _i in the training text set M; The amount of information is sorted, and some higher-ranked words are selected as feature items;

S204: Next, calculate the similarity between the extracted feature item and the information in the sensitive lexicon:

Set the selected feature item as W _i , and form a set W, β is the similarity between the feature item and the sensitive lexicon information:

S205: Then, by setting the threshold n, it is judged whether the web page information is bad information, if it is greater than the set threshold n, it means that the web page contains bad information, otherwise, it means that the web page meets the requirements and no bad information appears; after that, And remove the web pages containing bad information, and keep the web pages in normal use.

Further, in step S4: by installing a time valve in the system to control the time to start the computer software and hardware detection, set a time period in the time valve, such as setting 24 hours as a time period, at this time, it means that every The software and hardware of the computer system are inspected 24 hours a day. By using the time valve to control the computer to automatically clean the system on a regular basis, it is more conducive to avoiding the emergence of computer viruses;

After the time valve is opened, then enter the hardware detection mode and the software detection mode; The hardware detection mode detects whether the computer hardware function can be used normally by starting a series of computer detection tools, and the computer detection tools include CPU-Z for detection processing HD TUNE is used to detect the hard disk transfer rate, hard disk capacity, health status, temperature and disk surface scanning; Super PI is used to detect CPU stability, etc.; the software detection mode is used to detect computer viruses by using the software simulation scanning method Identify, and use the anti-virus software installed in the computer to check and kill the identified virus. The software simulation scanning method belongs to the conventional technical means of those skilled in the art, so it is not repeated too much; if the anti-virus software cannot Thoroughly clean up the identified virus, enter the alarm system, and send the signal to the feedback module, and timely alarm the virus that cannot be cleaned up, which is more conducive to the staff to find the problem, thereby reducing the occurrence of the virus and causing the computer system to be paralyzed The problem.

Further, in step S5: display the result of system detection, including hardware function detection result, software virus detection result, if the result shows as normal, then do not carry out alarm processing, if the result shows as abnormal, then remind and report to the police, convenient Remind the staff to repair and deal with the computer system.

Embodiment one:

In step S1: register the visitor's identity information, and judge the visitor's identity according to the member information recorded in the database. At this time, the system receives that the visitor's identity information exists in the record table in the database, which is "student - Zhang San", the system opens the student access mode, so that the computer is in the student access interface;

In step S2: In order to detect whether the uploaded file of "Student-Zhang San" complies with the campus management norms, the uploaded data is reviewed using the intelligent review model:

S201: First calculate the similarity between the keyword in the file content and the sensitive word information in the sensitive word database:

Use the word embedding algorithm to map the uploaded file content into a numerical vector space, establish a coordinate system, and process the uploaded file information to obtain vector set A, and obtain vector set B by processing the sensitive lexicon information , set γ as the similarity between vector set A and vector set B, at this time

Through traversal, it is found that the similarity between the content information of the file and the information in the sensitive lexicon is 30%;

S202: Since the similarity between the content information of the file and the sensitive lexicon information is 30% and less than the set threshold of 0.9, at this time, the file complies with the campus management regulations, and the file can be uploaded to the system;

At the same time, in order to detect whether the uploaded file data contains viruses, the search method based on character strings is used to detect according to the virus database, so as to effectively avoid the problem of virus invasion caused by the upload of file data. Through system detection, it is found that No virus was identified. At this moment, it shows that the file was uploaded successfully.

In order to prevent "student-Zhang San" from encountering bad information in the computer during the use of data, use information filtering technology to filter:

S203: First extract the feature values of all words in the web page based on the vector space model:

Set the text set formed in the webpage as R, the i-th vocabulary in the text as R _i , and set the training text set in the database as M, then the logarithmic mutual information logI(R _i ,R) is:

Among them, P(R _i |R) represents the proportion of vocabulary R _i in the text set R, and P(R _i ) represents the proportion of vocabulary R _i in the training text set M; The amount of information is sorted, and some higher-ranked words are selected as feature items;

S204: Next, calculate the similarity between the extracted feature item and the information in the sensitive lexicon:

Set the selected feature item as W _i , and form a set W, β is the similarity between the feature item and the sensitive lexicon information:

S205: Then, by setting a threshold of 0.8, it is judged whether the web page information is bad information, if it is greater than the set threshold of 0.8, the web page is cleared, and if it is less than the set threshold of 0.8, the web page is kept, thereby avoiding possible bad information in the computer The information page appears.

After the "student-Zhang San" finishes using the computer, the file data uploaded by the member and the web page data used are backed up and sent to the database.

In step S4: by installing a time valve in the system to control the time to start the computer software and hardware detection, a 12-hour cycle is set in the time valve, which means that the computer system software and hardware are detected once every 12 hours; the time valve is opened Afterwards, then enter hardware detection mode and software detection mode; Described hardware detection mode detects computer hardware function by starting a series of computer detection tools, and described computer detection tool comprises CPU-Z and is used for detecting processor and graphics card; HDTUNE is used for Detect hard disk transfer rate, hard disk capacity, health status, temperature and disk surface scanning; Super PI is used to detect CPU stability, etc.; the software detection mode identifies computer viruses by using software simulation scanning method, and uses The anti-virus software checks and kills the identified virus; if the anti-virus software cannot thoroughly clean up the identified virus, it enters the alarm system and sends a signal to the feedback module;

At this time, the system detects that the computer hardware is functioning normally, and the software part does not recognize the virus.

In step S5: the result of the system detection is displayed, the hardware function is in good condition, the software function is normal and there is no virus.

Finally, it should be noted that: the above is only a preferred example of the present invention, and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, it can still be The technical solutions recorded in the foregoing embodiments are modified, or some of the technical features are equivalently replaced. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A big data-based computer system security management system is characterized in that: the system comprises: the system comprises an access management module, a data uploading and using module, a database, a periodic detection module and a feedback module;

the identity information of an accessor is registered and an opening authority is set through the access management module, and the confirmed identity information is sent to the database and the data uploading and using module;

the data uploading and using module is used for managing the uploading and using of the data;

storing all identity information and uploaded and used data through the database;

the software and hardware of the computer are periodically detected by the periodic detection module, and the detection result is sent to the feedback module;

and displaying the detection result through the feedback module, judging whether the data is abnormal or not, and alarming if the data is abnormal.

2. The big data based computer system security management system of claim 1, wherein: the access management module comprises an identity registration unit and a permission setting unit; the identity registration unit is used for registering identity information of the visitor and confirming whether the visitor belongs to a school member or not; the permission setting unit is used for opening different permissions of the computer according to different identities of visitors.

3. The big data based computer system security management system of claim 1, wherein: the data uploading and using module comprises an uploading management unit, an information filtering unit and a data backup unit; the uploading management unit is used for detecting whether the file uploaded by the visitor meets the campus management specification or not and detecting whether the file contains viruses or not; the information filtering unit is used for filtering bad information encountered by a visitor in the data using process by using an information filtering technology; the data backup unit is used for backing up data uploaded and used by an accessor;

the uploading management unit comprises a content audit subunit and a virus detection subunit; the content auditing subunit is used for auditing the uploaded file content by using an intelligent auditing model and judging whether the uploaded file content conforms to the campus management specification; the virus detection subunit is used for detecting whether the uploaded file data contains viruses or not by using a search method based on character strings.

4. The big data based computer system security management system of claim 1, wherein: the database also comprises a virus library and a sensitive word library; the virus library is used for storing characteristic character strings of possible computer viruses; the sensitive word bank is used for storing sensitive word information which does not accord with campus management specifications.

5. The big data based computer system security management system of claim 1, wherein: the periodic detection module comprises a time control unit, a hardware detection unit and a software detection unit; the time control unit controls the starting time of the hardware detection unit and the software detection unit by arranging a time valve; the hardware detection unit is used for detecting whether the hardware function of the computer can be normally used; the software detection unit is used for detecting viruses in the computer system;

the software detection unit comprises a virus identification subunit, a virus searching and killing subunit and a virus alarm subunit; the virus identification subunit is used for identifying a virus in the computer system; the virus searching and killing subunit is used for cleaning the virus identified by the virus identification subunit; the virus alarm subunit is used for carrying out alarm processing on the virus which cannot be cleared and sending the signal to the feedback module.

6. A computer system security management method based on big data is characterized in that: the method comprises the following steps:

s1: registering identity information of an accessor and setting an opening authority;

s2: managing and detecting data uploaded and used by an accessor, and backing up the data;

s3: storing all identity information and data uploaded and used to the database;

s4: regularly detecting software and hardware of a computer system;

s5: and displaying the detection result, judging whether the data is abnormal or not, and alarming if the data is abnormal.

7. The big-data based computer system security management method of claim 6, wherein: in step S1: and registering the identity information of the visitor, judging whether the visitor belongs to a school member or not according to the member information recorded in the database, if so, opening the corresponding authority according to the identity information, and if not, stopping the access.

8. The big-data based computer system security management method of claim 6, wherein: in step S2: in order to detect whether the file uploaded by the visitor meets the campus management specification, the uploaded data is audited by using an intelligent auditing model:

s201: firstly, calculating the similarity between the keywords of the file content and the sensitive word information in the sensitive word library:

mapping the uploaded file content to a numerical vector space by using a word embedding algorithm, establishing a coordinate system, processing the uploaded file information to obtain a vector set A, processing the sensitive word bank information to obtain a vector set B, and setting gamma as the similarity of the vector set A and the vector set B, wherein at the moment

If the gamma value is larger than the set threshold value m, the sensitive words representing the content of the file are similar to the sensitive words of the sensitive word bank, and the campus management standard is not met at the moment, otherwise, the campus management standard is met;

s202: judging whether the file content meets the campus management specification, if so, uploading the file, and if not, stopping uploading the file and displaying uploading failure;

meanwhile, in order to detect whether the uploaded file data contains viruses or not, detection is performed according to the virus library by using a character string-based search method.

9. The big data based computer system security management method of claim 8, wherein: in step S2: in order to avoid that visitors encounter bad information in the computer in the process of using data, the information is filtered by using an information filtering technology:

s203: firstly, extracting characteristic values of all words in a webpage based on a vector space model:

setting a text set formed in a webpage as R, and setting the ith vocabulary in the text as R _i And setting the training text set in the database as M, then the logarithm mutual information amount logI (R) of the vocabulary and the text set _i And R) is:

wherein, P (R) _i R) represents the vocabulary R _i Specific gravity in text set R, P (R) _i ) Representing the word R _i Specific gravity in the training text set M; at the moment, all the vocabularies are sequenced according to the mutual information quantity of the formula, and some vocabularies arranged at higher positions are selected as characteristic items;

s204: and then calculating the similarity between the extracted feature items and the information in the sensitive word stock:

setting the selected feature item as W _i And forming a set W, wherein beta is the similarity between the characteristic item and the sensitive word stock information:

s205: and then judging whether the webpage information is bad information or not by setting a threshold value n, removing the webpage containing the bad information, and reserving the normally used webpage.

10. The big data based computer system security management method of claim 6, wherein: in step S4: the method comprises the steps that a time valve is installed in a system to control the starting time of computer software and hardware detection, and after the time valve is started, a hardware detection mode and a software detection mode are started; the hardware detection mode detects whether the hardware function of the computer can be normally used by starting a series of computer detection tools; the software detection mode is used for identifying computer viruses by using a software simulation scanning method and searching and killing the identified viruses by using antivirus software installed in a computer; and if the antivirus software cannot thoroughly clean the identified virus, entering an alarm system and sending a signal to the feedback module.

Images