Internal news portal data risk analysis method based on big data

Abstract

The invention belongs to the technical field of enterprise internal news risk analysis, and particularly discloses and provides an internal news portal data risk analysis method based on big data, which comprises the following steps of: performing news content misleading identification by detecting a bias reference risk, an internal propagation disorder risk and an employee behavior deviation influence of internal news release content, and obtaining a news content misleading result; whether the internal news is self-locked or not is judged, so that misleading propagation influence on the reading staff due to deviation influence of the news content is avoided; reading tracing of news transmission is carried out by identifying reading records of internal news in the earlier stage of locking, meanwhile, transmission word-of-mouth damage guidance of a reading group is analyzed, the error transmission influence degree of the internal news is reflected, and then the error transmission influence degree of the internal news is determined by analyzing the transmission word-of-mouth damage guidance of the internal news in the earlier stage of locking and the proportion of external transmission groups. And the data propagation uncontrollable risk trend of the news in the enterprise is determined, so that a reference basis can be provided for later news clarification measures.

Description

Technical Field

[0001] The present invention belongs to the technical field of internal news risk analysis of enterprises, and relates to a method for analyzing data risks of an internal news portal based on big data. Background Art

[0002] In today's digital age, the internal news portal of an enterprise, as a key hub for information exchange and dissemination, undertakes the task of massive information circulation within the organization. With the continuous expansion of the enterprise scale, the continuous expansion of business areas, and the rapid progress of information technology, the data scale of the internal news portal has shown an explosive growth trend.

[0003] In order to enable employees to promptly understand various key information such as the company's strategy, project progress, policies and regulations, etc., enterprises frequently release various news articles, notices, training materials, etc. through the internal news portal. These information sources are extensive, covering the feedback from various departments, management levels within the enterprise, and external cooperation units. The data types are complex and diverse. Behind this data prosperity, there are many potential risks. For example, in the data source link, due to the large number of information providers, there may be a risk of biased citation, that is, some news content may intentionally or unintentionally be biased towards a certain position or view, citing unreliable information sources, which is extremely likely to mislead employees' cognition and decision-making. However, traditional data risk analysis methods are unable to cope when faced with such a large and complex internal news portal data.

[0004] In the prior art, there are also some related solutions for internal news risk analysis of enterprises. For example, the patent with the Chinese patent publication number CN110502638A discloses a method for classifying enterprise news risks based on target entities. It adopts hierarchical classification, divides news into several major categories according to content, and there are several sub-categories under each major category. Each sub-category can specifically reflect the risks or development situation of the enterprise for the news; a statistical histogram and a probability graph of each enterprise entity in the news for the classification category are statistically calculated. Based on this statistical histogram, the risk statistical value of the news can be seen; the stored data is encrypted to improve the security of the stored data; functions such as risk category push and corresponding specific news content push are added to improve the user experience.

[0005] Although the above solution proposes a risk analysis and solution method for the impact of enterprise news content on the enterprise's own development, it still has the following limitations: specifically, it only analyzes the content published in the news itself, lacking the analysis of the relevance between internal news and the content sent through other enterprise communication channels, making it difficult to view the record of the news dissemination path. Inside large enterprises, there are diverse news dissemination channels, including emails, instant messaging tools, internal forums, etc. Especially when the dissemination involves multiple departments and levels, it is easy to have chaos in the dissemination link, and thus it is impossible to obtain the internal dissemination status and changes of the news in real time, making it difficult to discover and intervene in the disorder risks during the dissemination process in a timely manner. Moreover, it cannot comprehensively consider various risk factors, that is, it is difficult to track the dissemination behavior of news within the enterprise and its impact on employees' behavior, and even more impossible to give early warnings of potential news dissemination damage-oriented risks. Summary of the Invention

[0006] In view of this, to solve the problems raised in the above background technology, a method for analyzing the data risk of an internal news portal based on big data is proposed.

[0007] The object of the present invention can be achieved through the following technical solutions: The present invention provides a method for analyzing the data risk of an internal news portal based on big data, and the method includes the following steps: 1) Obtain the content of internal news releases, identify the external reference risk information of the internal news release content, and analyze the bias reference risk index ext_risk of the internal news release content.

[0008] 2) Detect the relevant information content sent through other internal communication channels, and analyze the internal dissemination disorder risk index inter_risk of the internal news.

[0009] 3) Detect the impact rate of employees' behavior deviation after the internal news is released, combine the bias reference risk index and the internal dissemination disorder risk index of the internal news release content, lock the viewing channels of the internal news, and obtain the internal news dissemination cycle.

[0010] 4) Obtain each access record of the internal news during the dissemination cycle, and identify the corresponding traceable target groups of each access record of the internal news during the dissemination cycle, including untraceable groups and traceable groups.

[0011] 5) Detect the dissemination reputation of the corresponding traceable target groups of each access record of the internal news during the dissemination cycle, and analyze the dissemination reputation damage-oriented weight of the access groups of each access record.

[0012] 6) Combine the dissemination reputation damage-oriented weights of the access groups of each access record, and analyze the uncontrollable risk trend of the data dissemination of enterprise internal news, including severe uncontrollable risk trend and mild uncontrollable risk trend.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By detecting the risk of biased citation, the risk of internal communication disorder, and the impact of employee behavior deviation in the internal news release content, the present invention timely identifies the misleading nature of the news content to determine whether to self-lock the internal news, avoiding the misleading spread impact on the viewing employees caused by the biased influence of the news content.

[0014] (2) By identifying the viewing records of internal news in the pre-lock period, the present invention traces the viewing of news dissemination. At the same time, it analyzes the damage orientation of the dissemination reputation of the viewing group, reflecting the degree of miscommunication impact of the internal news. Furthermore, by analyzing the damage orientation of the dissemination reputation and the proportion of external dissemination groups of internal news in the pre-lock period, the present invention determines the uncontrollable risk trend of data dissemination of enterprise internal news, which helps to provide a reference basis for later news clarification measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0016] Figure 1 It is a schematic diagram of the implementation steps flow of the method of the present invention.

[0017] Figure 2 It is a schematic diagram of the locking metric process of the viewing channel for evaluating internal news of the present invention.

[0018] Figure 3 It is a process display diagram of the traced orientation group of the access record of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0020] Please refer to Figure 1 As shown, the present invention provides a method for analyzing the data risk of an internal news portal based on big data. The method includes the following steps: 1) Obtain the internal news release content, identify the risk information of external citation of the internal news release content, and analyze the biased citation risk index ext_risk of the internal news release content.

[0021] The internal news, such as task-based news, application-based news, and event-based news, and the internal news release content includes: information such as the project objectives, task breakdown, time nodes, and responsible persons of task-based news; the introduction of new product technologies, comparison of product technical parameters, etc. of application-based news; and holiday time of event-based news.

[0022] In a preferred embodiment, the outward reference risk information of the internal news release content includes the link pointing to the website and the cited news information.

[0023] In a further preferred embodiment, the risk index of biased reference for analyzing the internal news release content includes: B1. Retrieve the website addresses pointed to by each link in the internal news release content, retrieve its source code through the website page, identify the alternative website addresses pointed to by the link and the alternative weights, and detect the release timestamp of each link-pointing website address and the corresponding release cycle of the current timestamp. Calculate the ratio of the corresponding release cycle of each link-pointing website address to the preset reference cycle, and accumulate the obtained result with the alternative weights of the alternative website addresses pointed to by the link to which each link-pointing website address belongs, to obtain the alternative pointing risk factor of the link-pointing website address.

[0024] It should be noted that the website addresses pointed to by each link in the internal news release content refer to the technical citation website addresses related to the news content or the publicity website addresses of national release regulations.

[0025] The alternative website addresses pointed to by the link are such as website addresses with network attack records, revoked website addresses, and the method for obtaining the alternative weights is: match the alternative website addresses pointed to by the link to which each link-pointing website address in the internal news release content belongs with the preset alternative weights of different alternative website address categories, to obtain the alternative weights of the alternative website addresses pointed to by the link to which each link-pointing website address in the internal news release content belongs.

[0026] The purpose of detecting the alternative website addresses pointed to by the link and the alternative weights is to: promptly discover those cited website addresses that have been tampered with by hackers or maliciously constructed. If an employee clicks on these website addresses, it may lead to the download of malicious software, such as viruses, Trojans, ransomware, etc. Therefore, detecting these alternative website addresses can help the enterprise identify risks in advance, prevent employees from accessing websites that may pose security threats, and thus protect the internal network security of the enterprise.

[0027] B2. Detect the relevant release content of the cited news information, including technical citation content and information citation content, and analyze the reliability risk factor of the cited information.

[0028] It should be noted that the purpose of analyzing the technical feasibility risk factor of the technical citation information is to: avoid the risk of task misguidance or task obstruction caused to enterprise employees due to changes in the technical citation information.

[0029] B3. Multiply the off-path pointing risk factor that links to a website and the reliability risk factor of the reference information by their respective preset risk weights, and then sum them up to obtain the biased reference risk indicator ext_risk of the internal news release content.

[0030] In a further preferred implementation, the content of analyzing the reliability risk factor of the reference information includes: constructing a reference database based on the technical reference content and information reference content in the relevant release content of the reference news information, and then identifying the changes in the reference content of the reference database through the establishment of a reference content tracking mechanism to obtain the content change rate ch1_F of the reference news information.

[0031] Set the source qualification score of the reference news information, and take the inverse ratio with the preset source reference score to obtain the confirmation risk indicator ch2_F of the reference news information.

[0032] The detection method for setting the source qualification score of the reference news information is as follows: for the technical reference content, obtain its release enterprise source, and then retrieve the authoritative certificate of the enterprise source, and perform authoritative qualification scoring on different levels of authoritative certificates, such as setting a higher qualification score for national authoritative certificates compared to provincial authoritative certificates; for the information reference content, obtain the relevant release time of the reference news, and perform different qualification scoring according to different release times, such as setting a higher qualification score for newly released reference news compared to historically released reference news.

[0033] Analyze the reliability risk factor of the reference information

[0034] 2) Detect the relevant information content sent through other internal communication channels, and analyze the internal dissemination disorder risk indicator inter_risk of the internal news.

[0035] The other internal communication channels are such as enterprise emails, department notices, etc.

[0036] The cross-channel associated information of the other internal communication channels includes a set of paradoxical information words.

[0037] In a preferred implementation, the content of analyzing the internal dissemination disorder risk indicator of the internal news includes: obtaining the topics and pointed departments involved in the internal news release content of the enterprise, and counting the employees in each internal department pointed by the enterprise and each other internal communication channel related to the enterprise application, retrieving the relevant information content sent by the enterprise to the employees in each internal department through each other internal communication channel based on the involved topics, comparing it with the internal news release content of the enterprise, and identifying the set of paradoxical information words X to which the internal news release content of the enterprise and the relevant information content sent by the employees in each internal department through each other internal communication channel belong through natural language processing (NLP) technology rk, r represents the number of employees in each internal department, r = 1, 2,..., a, and k represents the number of each other internal communication channels, k = 1, 2,..., c.

[0038] The paradoxical information refers to the keywords in the enterprise internal news release content that are mutually exclusive with the information content views sent through other internal communication channels. For example, regarding the progress of new product R & D: the internal news release content announces that the progress of new product R & D is 70%, while the information content sent through a certain other internal communication channel shows that the progress of new product R & D is 60%.

[0039] Identify all keyword sets Z of the internal news release content through natural language processing (NLP) technology, and analyze the internal communication disorder risk indicators of the internal news Where ∩ represents the intersection symbol and ∪ represents the union symbol.

[0040] 3) Detect the impact rate of employee behavior deviation after the internal news is released. Combine the bias citation risk indicator and the internal communication disorder risk indicator of the internal news release content to lock the viewing channels of the internal news and obtain the internal news dissemination cycle.

[0041] Please refer to Figure 2 As shown, in a preferred implementation manner, the detection of the impact rate of employee behavior deviation after the internal news is released includes: obtaining the corresponding release node of the internal news, extracting the corresponding progress efficiency of the news-related tasks pre-uploaded by each internal department employee at the release node as the numerator, detecting the corresponding progress efficiency of the news-related tasks of each internal department employee after the task is released as the denominator, and obtaining the news release progress impact rate of the corresponding news-related tasks of each internal department employee by taking the ratio, and then summing them up to obtain the impact rate bh of employee behavior deviation after the internal news is released.

[0042] The news-related tasks are such as the completion progress of the application functions of the product R & D project.

[0043] In a further preferred implementation manner, the locking of the viewing channels of the internal news and obtaining the internal news dissemination cycle include: evaluating the corresponding locking metric of the viewing channels of the internal news

[0044] , where ext1 and ext2 respectively represent the corresponding deviation threshold values of the preset first-level bias reference risk index and second-level bias reference risk index, inter1 and inter2 respectively represent the corresponding deviation threshold values of the preset first-level internal propagation disorder risk index and second-level internal propagation disorder risk index, ∨ represents the logical OR symbol, bh1 and bh2 respectively represent the corresponding deviation threshold values of the preset first-level employee behavior deviation influence rate and second-level employee behavior deviation influence rate. When q = 2, the viewing channel of the internal news is blocked and locked; when q = 1, the IPs of each access user of the internal news are marked for access records; when q = 0, the internal news is in the regular announcement period.

[0045] Obtain the real-time judgment timestamps of q = 2 and q = 1, and record their interval period as the internal news propagation cycle.

[0046] The present invention detects the bias reference risk, internal propagation disorder risk, and employee behavior deviation influence of the internal news release content, and timely identifies the misleading nature of the news content to determine whether to self-lock the internal news, so as to avoid the misleading propagation influence on the viewing employees caused by the biased influence of the news content.

[0047] 4) Obtain each access record of the internal news during the propagation cycle, and identify the corresponding traceable target groups of each access record of the internal news during the propagation cycle, including untraceable groups and traceable groups.

[0048] In a preferred embodiment, the identification of the corresponding traceable target groups of each access record of the internal news during the propagation cycle includes: detecting each access record of the internal news during the propagation cycle from the enterprise internal news release platform, and identifying the access group and access mark of each access record through a network viewing capture tool, where the access group includes the internal department employee group and the enterprise former employee group, and the access mark is the user screenshot record.

[0049] Track the propagation target groups of the access marks through the computer clients of the internal department employee group, including the external department employee group and the external enterprise personnel group.

[0050] Specifically, the method for obtaining the access group of each access record is: identify the user IP of each access record, and compare it with the pre-registered IP of the enterprise internal department employee group. If the user IP of a certain access record can match the pre-registered IP of the enterprise internal department employee group, the access group of this access record is the internal department employee group; similarly, compare it with the pre-registered IP of the enterprise former employee group to determine whether the access group of this access record is the enterprise former employee group.

[0051] The method for obtaining the target group of the spread of access tags by tracking the computer clients of the internal department employee group is as follows: Use a network browsing capture tool to capture the user IPs of the internal department employee group spreading the access tags in the enterprise internal software, and compare them with the pre-registered user IPs of the external department employee group of the enterprise. If the user IP of a certain access record can match the pre-registered IP of the external department employee group of the enterprise, the access group of this access record is the external department employee group; otherwise, it is the external personnel group of the enterprise.

[0052] Record the enterprise's former employee group and the external personnel group of the enterprise as non-traceable groups, and record the internal department employee group and the external department employee group as traceable groups. Based on this, identify the corresponding traceable target groups for each access record during the dissemination cycle of the internal news.

[0053] 5) Detect the spread word-of-mouth of the corresponding traceable target groups for each access record during the dissemination cycle of the internal news, and analyze the damage-oriented weight of the spread word-of-mouth of the access group for each access record.

[0054] Please refer to Figure 3 As shown, in a preferred implementation, the content of analyzing the damage-oriented weight of the spread word-of-mouth of the access group for each access record includes: obtaining the corresponding traceable target groups for each access record during the dissemination cycle of the internal news. For the enterprise's former employee group in the non-traceable group, obtain their on-the-job performance score d1 and the score d2 for the reason of leaving the job, and detect their access tags. For the internal department employee group, detect their access tags and import them into the discrimination formula for the damage-oriented weight of the spread word-of-mouth of the access group of the access record Derive the damage-oriented weight D of the spread word-of-mouth of the access group for each access record j 。

[0055] In the formula, P represents the access group, P1 represents that the access group is the enterprise's former employee group in the non-traceable group, P1 = 0 means that the enterprise's former employee group does not have an access tag, P1 = 1 means that the enterprise's former employee group has an access tag, P2 represents that the access group is the internal department employee group in the traceable group, P2 = 0 means that the spread target group of the corresponding access tag of the internal department employee group is the external department employee group, P2 = 1 means that the spread target group of the corresponding access tag of the internal department employee group is the external personnel group of the enterprise, d0, d3, and d4 respectively represent the preset word-of-mouth damage-oriented weights corresponding to different access groups and their different access tags, d0 < d3 < d4, j represents the number of each access record, j = 1, 2,..., m, and ∧ represents the logical AND symbol.

[0056] It should be noted that the on-the-job performance score and the score for the reason of leaving the job of the former employee group are recorded and uploaded by the enterprise management.

[0057] 6) Combine the word-of-mouth damage-oriented weights corresponding to the access groups of each access record, and analyze the uncontrollable risk trend of data dissemination of internal corporate news, including severe uncontrollable risk trends and mild uncontrollable risk trends. Furthermore, when there is a severe uncontrollable risk trend, rectify and maintain the internal corporate news release platform, such as issuing news clarification announcements and managing the permissions of the group of former employees of the enterprise.

[0058] In a preferred embodiment, the analysis of the uncontrollable risk trend of data dissemination of internal corporate news includes: obtaining the proportion ρ of the access records corresponding to the untraceable groups of internal corporate news, and combining the word-of-mouth damage-oriented weights D corresponding to the access groups of each access record j , and analyze the uncontrollable risk trend of data dissemination of internal corporate news un_1 represents a severe uncontrollable risk trend, un_2 represents a mild uncontrollable risk trend, e is the natural constant, and μ represents a preset data dissemination uncontrollable risk assessment threshold.

[0059] The present invention conducts reading traceability of news dissemination by identifying the reading records existing in the early stage of locking of internal news. At the same time, it analyzes the word-of-mouth damage orientation of the reading group, reflects the degree of misinformation dissemination impact of internal news, and then determines the uncontrollable risk trend of data dissemination of internal corporate news by analyzing the word-of-mouth damage orientation and the proportion of external dissemination groups in the early stage of locking of internal news, which helps to provide a reference basis for later news clarification measures.

[0060] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A data risk analysis method for internal news portal based on big data, characterized in that: The method comprises the following steps: 1) Obtain internal news release content, identify the external citation risk information of the internal news release content, and analyze the biased citation risk indicator ext_risk of the internal news release content; 2) Detect the relevant information content sent by other internal communication channels and analyze the internal communication disorder risk indicator inter_risk of internal news; 3) Detect the impact rate of employee behavior deviation after the internal news is released, combine the biased citation risk index of the internal news release content and the internal communication disorder risk index, lock the reading channel of the internal news, and obtain the internal news communication cycle; 4) Obtain each access record of internal news during the dissemination cycle, and identify the corresponding traceability-oriented groups of each access record of internal news during the dissemination cycle, including non-traceable groups and traceable groups; 5) Detect the communication reputation of each access record corresponding to the traceability group of the internal news during the communication cycle, and analyze the corresponding communication reputation damage guidance weight of the access group of each access record; 6) Combined with the corresponding word-of-mouth damage-oriented weights of the visit groups of each visit record, the uncontrollable risk trends of data dissemination of internal corporate news are analyzed, including serious uncontrollable risk trends and mild uncontrollable risk trends.

2. According to the method for analyzing internal news portal data risk based on big data in claim 1, it is characterized in that: The external citation risk information of the internal news release content includes the link pointing to the website and the cited news information.

3. According to the method for analyzing internal news portal data risk based on big data in claim 2, it is characterized in that: The analysis of the biased citation risk indicators of internal news releases includes: B1. Search the URLs pointed to by each link in the internal news release content, and search its source code through the URL page, identify the URLs pointed to by different paths and their different path weights, and detect the release timestamp of each URL pointed to by the link and the release cycle corresponding to the current timestamp, compare the corresponding release cycle of each URL pointed to by the link with the preset reference cycle, and accumulate the obtained result with the different path weight of the URL pointed to by different paths to each URL pointed to by the link, so as to obtain the different path pointing risk factor of the URL pointed to by the link; B2. Detect the relevant published content of the quoted news information, including technical quoted content and information quoted content, and analyze the reliability risk factors of the quoted information; B3. Multiply the risk factor of the alternate path pointing to the URL and the reliability risk factor of the cited information by the corresponding preset risk weights and add them up to obtain the biased citation risk indicator ext_risk of the internal news release content.

4. According to the method for analyzing internal news portal data risk based on big data as described in claim 3, it is characterized in that: The reliability risk factor of the cited information is analyzed, including: constructing a citation database based on the technical citation content and information citation content in the relevant published content of the cited news information, and then identifying the change of the citation content in the citation database by establishing a citation content tracking mechanism to obtain the content change rate ch1_F of the cited news information; Set the source qualification score of the quoted news information, and inversely compare it with the preset source reference score to obtain the confirmation risk index ch2_F of the quoted news information; Analyze the reliability risk factors of citation information 5. According to the method for analyzing internal news portal data risk based on big data in claim 1, it is characterized in that: The analysis of the internal communication disorder risk index of internal news includes: obtaining the topics and departments involved in the internal news release content of the enterprise, and counting the internal department employees of the enterprise's departments, and other internal communication channels related to the enterprise application, retrieving the relevant information content sent by the enterprise to the corresponding other internal communication channels of the internal department employees based on the topics involved, comparing it with the internal news release content of the enterprise, and identifying the contradictory information word set X to which the internal news release content of the enterprise and the relevant information content sent by the corresponding other internal communication channels of the internal department employees belong through natural language processing (NLP) technology rk , r represents the number of employees in each internal department, r = 1, 2, ..., a, k represents the number of other internal communication channels, k = 1, 2, ..., c; Identify all keyword sets Z of internal news releases through natural language processing (NLP) technology and analyze the risk indicators of internal dissemination disorder of internal news Here, ∩ represents the intersection symbol and ∪ represents the union symbol.

6. The method for analyzing internal news portal data risk based on big data according to claim 5 is characterized in that: The method for detecting the impact rate of employee behavior deviation after the internal news is released includes: obtaining the corresponding release node of the internal news, extracting the corresponding progress efficiency of the news-related tasks pre-uploaded by the employees of each internal department at the release node and using it as the numerator, detecting the corresponding progress efficiency of the news-related tasks of the employees of each internal department after the task is released and using it as the denominator, comparing and obtaining the news release progress impact rate of the corresponding news-related tasks of the employees of each internal department, and then accumulating and obtaining the employee behavior deviation impact rate bh after the internal news is released.

7. The method for analyzing internal news portal data risk based on big data according to claim 6 is characterized in that: The locking of the reading channel of the internal news and obtaining the internal news dissemination cycle include: judging the corresponding locking metric of the reading channel of the internal news Among them, ext1 and ext2 represent the preset first-level biased citation risk index and the corresponding deviation threshold of the second-level biased citation risk index, inter1 and inter2 represent the preset first-level internal communication disorder risk index and the corresponding deviation threshold of the second-level internal communication disorder risk index, ∨ represents the logical OR symbol, bh1 and bh2 represent the preset first-level employee behavior deviation impact rate and the corresponding deviation threshold of the second-level employee behavior deviation impact rate, respectively. When q=2, the reading channel of internal news is blocked and locked; when q=1, the IP of each user accessing the internal news is marked and the access record is made; when q=0, the internal news is in the regular announcement period; The real-time evaluation timestamps of q=2 and q=1 are obtained, and the interval period is recorded as the internal news dissemination cycle.

8. The method for analyzing internal news portal data risk based on big data according to claim 1 is characterized in that: The identification of the corresponding tracing guidance groups of each access record of the internal news in the dissemination cycle includes: detecting each access record of the internal news in the dissemination cycle from the enterprise internal news release platform, and identifying the access group and access mark of each access record through a network reading capture tool, wherein the access group includes an internal department employee group and a group of employees who have resigned from the enterprise, and the access mark is a user screenshot record; Through the computer client of the internal department employee group, track the target group of the communication of the access mark, including the external department employee group and the group of people outside the enterprise; The group of employees who have resigned from the company and the group of people outside the company are recorded as non-traceable groups, and the group of employees in internal departments and the group of employees in external departments are recorded as traceable groups. Based on this, the corresponding traceability-oriented groups of each access record during the dissemination cycle of internal news are identified.

9. The method for analyzing internal news portal data risk based on big data according to claim 8 is characterized in that: The analysis of the corresponding communication reputation damage orientation weight of the visitor group of each visit record includes: obtaining the corresponding traceability orientation group of each visit record of internal news within the communication cycle, obtaining the on-the-job performance score d1 and the resignation reason score d2 of the enterprise resigned employee group in the non-traceable group, and detecting their visit marks; for the internal department employee group, detecting their visit marks, and importing the visit record The corresponding communication reputation damage orientation weight judgment formula Derive the corresponding word-of-mouth damage-oriented weight D of the visitor group for each visit record j ; Wherein, P represents the visit group, P1 represents the visit group is the group of employees who have left the enterprise in the untraceable group, P1=0 represents the group of employees who have left the enterprise does not have an visit mark, P1=1 represents the group of employees who have left the enterprise does have an visit mark, P2 represents the visit group is the group of internal department employees in the traceable group, P2=0 represents the target group of the corresponding access mark of the internal department employee group is the group of employees outside the enterprise, P2=1 represents the target group of the corresponding access mark of the internal department employee group is the group of people outside the enterprise, d0, d3, d4 represent the preset word-of-mouth damage guidance weights corresponding to different visit groups and their different access marks, d0<d3<d4, j represents the number of each visit record, j=1,2,...,m, ∧ represents the logical AND symbol.

10. The method for analyzing internal news portal data risk based on big data according to claim 9, characterized in that: The analysis of the uncontrollable risk trend of data dissemination of internal enterprise news includes: obtaining the corresponding access record proportion ρ of the untraceable group of internal enterprise news, combining the corresponding dissemination reputation damage orientation weight D of the access group of each access record j , analyze the uncontrollable risk trend of data dissemination of internal corporate news un_1 represents a serious uncontrollable risk trend, un_2 represents a mild uncontrollable risk trend, e is a natural constant, and μ represents the preset data propagation uncontrollable risk assessment threshold.

Images