A hot information extraction system based on big data

Abstract

The present invention relates to the technical field of hot information extraction, and in particular to a hot information extraction system based on big data, including a platform data collection module for collecting article information, user information and article access information in a social platform; a data numbering construction module for numbering article information and user information; an article classification processing module for classifying and processing article information; a user data analysis module for analyzing user preference characteristics and user interaction quality; an article data analysis module for analyzing content relevance and article content characteristics; an access data analysis module for analyzing article access characteristics; a heat comprehensive analysis module for analyzing article hot type; and a hot information extraction module for extracting article information with article hot types of category two and category three as hot information. The present invention realizes the accurate extraction of hot information in a social platform.

Description

Technical Field

[0001] The present invention relates to the technical field of hot spot information extraction, and in particular to a hot spot information extraction system based on big data. Background Art

[0002] Hot information extraction mainly relies on information acquisition, preprocessing and extraction technologies. These technologies locate and collect information through meta-search, etc. After preprocessing such as purification and word segmentation, machine learning algorithms are used to extract hot keywords to quickly and accurately obtain key information from massive data.

[0003] Chinese Patent Publication No.: CN115795175A discloses a multi-dimensional hotspot extraction method based on data analysis, including the following steps: Step 1: Extract dimensional hot information from different dimensions through a dimension selection unit. The present invention collects data on hotspot information of different dimensions, and divides the collected data in turn, thereby clearly dividing the categories of data, and extracting and processing the data, performing different types of data processing based on the divided data, thereby obtaining analysis coefficients and evaluation coefficients of corresponding types of data, converting the data into numerical values, increasing the intuitiveness of the data, performing comprehensive processing based on the different coefficients after conversion, and judging and selecting hotspot information of different dimensions based on the processing results, thereby selecting hotspots of different dimensions, selecting non-repetitive hotspot data, and generating extraction and transmission prompts, and extracting and transmitting according to the prompts. The invention realizes the extraction of hotspot data in combination with the user's access signal data and the set multi-dimensional data, but does not realize the comprehensive analysis of hotspot data extraction in combination with the user's preferences, the content relevance of the article and the user's comment data, and has the problems of low efficiency in analyzing user data and article data and inaccurate extraction of hotspot data. Summary of the invention

[0004] The object of the present invention is to provide a hot spot information extraction system based on big data to solve at least one of the problems existing in the prior art.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A hot information extraction system based on big data, comprising:

[0007] The platform data collection module is used to collect article information and user information on the social platform, and periodically collect article access information;

[0008] A data numbering construction module is used to number the article information and user information to obtain the article number and user number;

[0009] An article classification processing module is used to classify the article information to obtain the article category of the article information;

[0010] User data analysis module, used to analyze user preference characteristics and user interaction quality based on user information, article number, user number and article category;

[0011] Article data analysis module, used to analyze content relevance and article content characteristics based on article information and article categories;

[0012] Access data analysis module, used to analyze article access characteristics based on article access information, content relevance and article content characteristics;

[0013] Comprehensive analysis module for analyzing article hot spots based on article access characteristics, user preference characteristics, and user interaction quality;

[0014] The hot information extraction module is used to extract hot information according to the hot type of the article.

[0015] Furthermore, the article classification processing module matches the article subject with preset category keywords, and analyzes the article category based on the matching results. If there are preset category keywords in the article subject, the article classification processing module sets the article category to be the same as the preset category corresponding to the preset category keywords. If there are no preset category keywords in the article subject, the article classification processing module sets the article category to no category.

[0016] Furthermore, the user data analysis module is provided with a user preference analysis unit, which is used to count the number of different article categories corresponding to the article information in the user's browsing information as the browsing category number N1(j), count the number of different article information corresponding to each article category in the user's browsing information as the browsing preference number N(j,k), and analyze the user preference characteristics according to the browsing category number and the browsing preference number to obtain the user preference characteristics H(j).

[0017] Furthermore, the user data analysis module is also provided with a user comment analysis unit, which is used to analyze the user interaction quality according to whether the user comment content is a relevant comment, the number of relevant comments N2(j,i) and the number of comment likes B(j,i) to obtain the user interaction quality Q(j,i).

[0018] Furthermore, the article data analysis module is provided with a content association analysis unit, which is used to remove words in the article subject and article content that are the same as preset simplified keywords to obtain a simplified subject and simplified content, and use the words in the simplified subject as association analysis keywords. The content association analysis unit extracts the number of occurrences of the association analysis keywords in the simplified content as the content association number N3(i), and analyzes the content association degree based on the content association number to obtain the content association degree P(i).

[0019] Furthermore, the article data analysis module is also provided with a content feature analysis unit, which is used to count the number of other article information that have the same article category as the currently analyzed article information and contain associated analysis keywords as the feature analysis number s(i), and analyze the article content features based on the feature analysis number and the content association number N3(i) to obtain the article content feature D(i).

[0020] Furthermore, the access data analysis module is provided with an access feature analysis unit, which is used to analyze the article access features according to the access volume W(i, t) and the article publishing time T to obtain the article access features A(i, t).

[0021] Furthermore, the access data analysis module is further provided with an access association analysis unit, which is used to compare the content association P(i) with the association threshold p, and adjust the analysis process of the article access feature according to the comparison result. When the content association does not meet the threshold, the analysis process of the article access feature is adjusted, and the adjusted article access feature is A1(i,t);

[0022] The access data analysis module is also provided with an article content analysis unit to compare the article content feature D(i) with the content feature threshold d, and optimize the adjustment process of the article access feature according to the comparison result. When the article content feature meets the threshold, the adjustment process of the article access feature is optimized, and the optimized article access feature is A2(i,t).

[0023] Furthermore, the comprehensive heat analysis module is provided with a comprehensive heat analysis unit, which is used to analyze the article heat parameters according to the article access characteristics A(i, t), the user preference characteristics H(j) and the user interaction quality Q(j, i) to obtain the article heat parameters F(i, t).

[0024] Furthermore, the comprehensive heat analysis module is also provided with a hotspot type judgment unit, which is used to compare the article heat parameter F(i, t) with the heat thresholds f1 and f2, and judge the article hotspot type according to the comparison result. The article hotspot types include one, two and three.

[0025] The beneficial effects of the present invention are as follows: through the platform data collection module to collect article information, user information and article access information in the social platform, and the comprehensive analysis of the collected information by other modules, it is possible to analyze the popularity of articles in combination with user preferences, article subject content relevance, article content characteristics and article access data, thereby extracting high-popularity articles as hot information, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot data extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0027] Figure 1 This is a schematic diagram of the structure of the hot spot information extraction system based on big data in this embodiment.

[0028] Figure 2 Schematic diagram of the structure of the user data analysis module of this embodiment.

[0029] Figure 3 Schematic diagram of the structure of the article data analysis module of this embodiment.

[0030] Figure 4 This is a schematic diagram of the structure of the access data analysis module in this embodiment.

[0031] Figure 5 Schematic diagram of the structure of the comprehensive heat analysis module of this embodiment. DETAILED DESCRIPTION

[0032] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the content specifically described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0033] It should be noted that, although the terms first, second, third, etc. may be used to describe in the embodiments of the present application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0034] See also Figure 1 As shown, it is a hot spot information extraction system based on big data in this embodiment, including:

[0035] The platform data collection module is used to collect article information and user information in the social platform, and periodically collect article access information, the article information includes article subject and article content, the user information includes user interaction information and user browsing information, the user interaction information includes user interaction behavior and user comment information, the user interaction behavior includes like, comment and share, the user comment information includes user comment content and the number of comment likes, the user browsing information is the information of articles browsed by the user, the article access information includes the number of visits and the length of time the article has been published, the collection frequency of the article access information in this embodiment is once per preset collection period, the preset collection period in this embodiment is set to 0.5 hours, the setting of the preset collection period is not specifically limited in this embodiment, and the technicians in this field can freely set it, as long as the collection of article access information is satisfied, for example, the preset collection period can also be set to 0.8 hours, 1 hour and 2 hours, etc., the article information, user information, and article access information are obtained by importing data from the social platform data management platform.

[0036] Specifically, this embodiment is applied to the management cloud of a social platform, and extracts hot data from the article data by analyzing user interaction data and article data to ensure that the hot data meets the needs of users with various preferences. The social platform includes Weibo, etc.

[0037] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0038] The data number construction module is used to number the article information and the user information to obtain the article number and the user number. The data number construction module is connected to the platform data acquisition module.

[0039] Specifically, the data numbering construction module in this embodiment numbers the article information in the order of collection time to obtain the article number, and sets the article number to i, i∈N + The data number construction module sets different user numbers for different user information, and sets the user number to j, j∈N + .

[0040] Specifically, in this embodiment, the data numbering construction module is used to process the numbering of article information and user information to obtain article numbers and user numbers, thereby ensuring that the data is clear when analyzing article information and user information, facilitating the distinction between different data, thereby improving the system's analysis efficiency of user data and article data and improving the accuracy of hot spot data extraction.

[0041] Please continue reading Figure 1As shown, the hot spot information extraction system based on big data also includes:

[0042] The article classification processing module is used to classify the article information to obtain the article category of the article information. The article classification processing module is connected to the data number construction module.

[0043] Specifically, the article classification processing module described in this embodiment matches the article subject with preset category keywords, and analyzes the article category based on the matching results. If there are preset category keywords in the article subject, the article classification processing module sets the article category to be the same as the preset category corresponding to the preset category keywords. If there are no preset category keywords in the article subject, the article classification processing module sets the article category to no category.

[0044] Specifically, in this embodiment, the article classification processing module classifies the article information to analyze the article category, classifies the article information according to its corresponding field, increases the diversity of system analysis, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot data extraction.

[0045] Specifically, the preset categories described in this embodiment can be divided into one category, two categories, and three categories, etc. In this embodiment, the preset categories are not specifically limited, and technical personnel in this field can set them freely. For example, one category can be used to represent agriculture, and its corresponding keywords may include "planting", "solar terms" and "irrigation", etc., as long as the judgment of the article category is met.

[0046] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0047] The user data analysis module is used to analyze the user preference characteristics and user interaction quality according to the user information, article number, user number and article category. The user data analysis module is connected to the article classification processing module.

[0048] See also Figure 2 As shown, the user data analysis module includes:

[0049] The user preference analysis unit is used to analyze the user preference characteristics according to the user browsing information, article number, user number and article category, and use the user preference characteristics to represent the browsing preference characteristics of articles of different article categories in the user browsing information.

[0050] Specifically, the user preference analysis unit in this embodiment counts the number of different article categories corresponding to the article information in the user's browsing information as the number of browsing categories, counts the number of different article information corresponding to each article category in the user's browsing information as the number of browsing preferences, and analyzes the user preference feature according to the number of browsing categories and the number of browsing preferences, sets the user preference feature as H(j), and sets , where N(j,k) represents the number of browsing preferences, k represents the article category number, N1(j) represents the number of browsing categories, and N(j,k) max Indicates the maximum value of the number of browsing preferences corresponding to each article category.

[0051] Specifically, in this embodiment, the user preference analysis unit analyzes the user browsing information, article number, user number and article category to analyze the user preference characteristics, analyze the browsing habits of different users, and determine whether the categories of articles browsed by users are diverse, thereby improving the system's analysis efficiency of user data and article data and improving the accuracy of hot data extraction.

[0052] Please continue reading Figure 2 As shown, the user data analysis module also includes:

[0053] The user comment analysis unit is used to analyze the user interaction quality according to the user comment content and article category, and the user interaction quality is used to represent the relevance between the user's comment on each article and the article content. The user comment analysis unit is connected to the user preference analysis unit.

[0054] Specifically, the user comment analysis unit described in this embodiment matches the user comment content with the preset category keywords corresponding to the article category of the article information. If the preset category keywords exist in the user comment content, the user comment analysis unit determines that the user comment content is a relevant comment, and counts the number of occurrences of the preset category keywords in the user comment content as the number of comment relevance. If the preset category keywords do not exist in the user comment content, the user comment analysis unit determines that the user comment content is an irrelevant comment.

[0055] Specifically, when the user comment content is a relevant comment, the user comment analysis unit in this embodiment analyzes the user interaction quality according to the number of comment relevance and the number of comment likes, and sets the user interaction quality to Q(j,i), and sets Q(j,i)=lgB(j,i) 1 / 2 ×lgM1(j,i) / lg[N2(j,i)+1], when the user comment content is an irrelevant comment, the user comment analysis unit analyzes the user interaction quality according to the user comment content, sets the user interaction quality to Q(j,i), and sets Q(j,i)=lgB(j,i)1 / 2 ×lgM1(j,i) / lg[M(k)+2], where M1(j,i) represents the number of words in the user's comment content, N2(j,i) represents the number of related comments, M(k) represents the number of keywords in the preset category keywords corresponding to the article category of the article information, and B(j,i) represents the number of likes on the comment.

[0056] Specifically, in this embodiment, the user comment analysis unit analyzes the user comment content and article category to determine whether the user's comment is a relevant comment, thereby analyzing the user interaction quality and ensuring that the user interaction quality is related to the user comment content and the popularity of likes on its content, thereby improving the system's analysis efficiency of user data and article data and improving the accuracy of hot spot data extraction.

[0057] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0058] The article data analysis module is used to analyze the content relevance and article content characteristics according to the article information and article category. The article data analysis module is connected to the article classification processing module.

[0059] See also Figure 3 As shown, the article data analysis module includes:

[0060] The content relevance analysis unit is used to analyze the content relevance based on the article information.

[0061] Specifically, the content association analysis unit in this embodiment removes the words that are the same as the preset simplified keywords in the article subject and the article content to obtain the simplified subject and the simplified content, and uses the words in the simplified subject as the association analysis keywords. The content association analysis unit extracts the number of times the association analysis keywords appear in the simplified content as the content association number, and analyzes the content association degree according to the content association number, and sets the content association degree as P(i), and sets P(i)=e N3(i) / M2(i) -1; where N3(i) represents the number of content associations, and M2(i) represents the number of words in the simplified content.

[0062] Specifically, in this embodiment, the content association analysis unit analyzes the article information to analyze the content association, and uses the content association to represent the characteristic relationship between the key words in the article theme and the words in the article content, so as to determine whether the article content and the theme are appropriate, thereby improving the system's analysis efficiency of user data and article data and improving the accuracy of hot data extraction.

[0063] Please continue reading Figure 3 As shown, the article data analysis module also includes:

[0064] The content feature analysis unit is used to analyze the content features of the article according to the article information and the article category, and the content feature analysis unit is connected to the content association analysis unit.

[0065] Specifically, the content feature analysis unit described in this embodiment counts the number of other article information that have the same article category as the currently analyzed article information and have associated analysis keywords in the article information as the feature analysis number, and analyzes the article content features based on the feature analysis number and the number of content associations, and sets the article content features to D(i), setting D(i)=N3(i)×lg{S(k) / [s(i)+1]} / M2(i), wherein S(k) represents the number of article information corresponding to the article category of the currently analyzed article information, and s(i) represents the feature analysis number.

[0066] Specifically, in this embodiment, the content feature analysis unit analyzes the article information and article categories to analyze the article content features, and uses the article content features to represent the word frequency features of the current analyzed article content in other article information in its corresponding article category, thereby realizing the correlation analysis of article information in the same category, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot data extraction.

[0067] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0068] The access data analysis module is used to analyze the article access characteristics according to the article access information, content relevance and article content characteristics, and the access data analysis module is connected to the article data analysis module.

[0069] See also Figure 4 As shown, the access data analysis module includes:

[0070] The access feature analysis unit is used to analyze the article access features according to the article access information.

[0071] Specifically, the access feature analysis unit in this embodiment analyzes the article access feature according to the number of visits and the length of time the article has been published, and sets the article access feature to A(i, t), and sets A(i, t) = lg[W(i, t) / W(i, t-1)], where W(i, t) represents the number of visits, t represents the collection cycle number, and t∈N + , t=T / ω, T represents the time the article has been published, and ω represents the length of the preset collection cycle. The collection cycle number is defined as the number used to distinguish different collection times of article access information.

[0072] Specifically, in this embodiment, the access feature analysis unit analyzes the article access information to analyze the article access features, and uses the article access features to represent the changing characteristics of the article access volume over a period of time, thereby realizing the analysis of the changes in the number of article accesses, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot spot data extraction.

[0073] Please continue reading Figure 4 As shown, the access data analysis module also includes:

[0074] The access association analysis unit is used to adjust the analysis process of the article access feature according to the content relevance, so that the adjusted article access feature is related to the degree of relevance between the article's subject and content. The access association analysis unit is connected to the access feature analysis unit.

[0075] Specifically, the access association analysis unit in this embodiment compares the content association with the association threshold, and adjusts the analysis process of the article access feature according to the comparison result. If P(i) < p, the access association analysis unit determines that the content association does not meet the threshold, and adjusts the analysis process of the article access feature. The adjusted article access feature is A1(i, t), and A1(i, t)=A(i, t) / e is set. P(i) ; If P(i)≥p, the access association analysis unit determines that the content association meets the threshold value, and does not adjust the analysis process of the article access features; wherein p represents the association threshold, 0.05≤p≤0.1; the access association analysis unit analyzes the content association to adjust the analysis process of the article access features, and adjusts the article access features when the correlation between the article topic and the content is small, thereby reducing the access features of articles that may be of low quality, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot spot data extraction; it can be understood that the value of the association threshold is not specifically limited in this embodiment, and technical personnel in this field can freely set it as long as it satisfies the adjustment of the article access feature analysis process. The optimal value of the association threshold is; p=0.07.

[0076] Please continue reading Figure 4 As shown, the access data analysis module also includes:

[0077] The article content analysis unit is used to optimize the adjustment process of the article access feature according to the article content feature, so that the optimized article access feature is related to the word frequency feature of the current analysis article category. The article content analysis unit is connected to the access association analysis unit.

[0078] Specifically, the article content analysis unit in this embodiment compares the article content feature with the content feature threshold, and optimizes the adjustment process of the article access feature according to the comparison result. If D(i)≤d, the article content analysis unit determines that the article content feature does not meet the threshold, and does not optimize the adjustment process of the article access feature. If D(i)>d, the article content analysis unit determines that the article content feature meets the threshold, and optimizes the adjustment process of the article access feature. The optimized article access feature is A2(i,t), and A2(i,t)=A1(i,t)×e is set. D(i) / d-1 ; Wherein, d represents the content feature threshold, 0.7≤d≤1; the article content analysis unit analyzes the article content features to optimize the adjustment process of the article access features, and realizes a comprehensive analysis of the feature relationship between the article information of each category, ensuring that the article access features are related to the relevance of the article content of the current category, thereby improving the system's analysis efficiency of user data and article data, and improving the accuracy of hot data extraction; it can be understood that in this embodiment, the value of the content feature threshold is not specifically limited, and technical personnel in this field can freely set it, as long as it meets the optimization of the article access feature adjustment process. The optimal value of the content feature threshold is: d=0.8.

[0079] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0080] The heat comprehensive analysis module is used to analyze the article hot spot types according to the article access characteristics, user preference characteristics and user interaction quality. The heat comprehensive analysis module is connected to the access data analysis module and the user data analysis module.

[0081] See also Figure 5 As shown, the heat comprehensive analysis module includes:

[0082] The comprehensive heat analysis unit is used to analyze article heat parameters based on article heat parameters, user preference characteristics and user interaction quality.

[0083] Specifically, the heat comprehensive analysis unit in this embodiment analyzes the article heat parameter according to the article access characteristics, user preference characteristics and user interaction quality, and sets the article heat parameter to F(i, t). , where U(i) represents the set of user numbers that have browsed the current article information, and NU(i) represents the number of elements in the set of user numbers that have browsed the current article information.

[0084] Specifically, in this embodiment, the comprehensive heat analysis unit analyzes the article heat parameters, user preference characteristics and user interaction quality to analyze the article heat parameters, and uses the article heat parameters to represent the analysis of the access heat of the article in the current time period based on comprehensive user preferences, article access data and article content, thereby improving the system's analysis efficiency of user data and article data and improving the accuracy of hot spot data extraction.

[0085] Please continue reading Figure 5 As shown, the heat comprehensive analysis module also includes:

[0086] The hotspot type judgment unit is used to analyze the hotspot type of the article according to the article heat parameters.

[0087] Specifically, the hotspot type judgment unit described in this embodiment compares the article heat parameter with the heat threshold, and analyzes the article hotspot type according to the comparison result. If F(i, t) < f1, the hotspot type judgment unit determines that the article hotspot type is one category; if f1≤F(i, t) < f2, the hotspot type judgment unit determines that the article hotspot type is two categories; if F(i, t)≥f2, the hotspot type judgment unit determines that the article hotspot type is three categories; wherein, f1 represents the first heat threshold, 0.7≤f1≤0.9, and f2 represents the second heat threshold, 1.5≤f2≤2. It can be understood that the value of the heat threshold is not specifically limited in this embodiment, and those skilled in the art can freely set it as long as it satisfies the analysis of the article hotspot type. The optimal value of the heat threshold is: f1=0.8, f2=1.8.

[0088] Specifically, in this embodiment, the article hotspot type of category one indicates that the article is non-hotspot information, the hotspot type of category two indicates that the article belongs to hotspot information in its corresponding category, and the hotspot type of category three indicates that the article is hotspot information.

[0089] Please continue reading Figure 1 As shown, the hot spot information extraction system based on big data also includes:

[0090] The hot information extraction module is used to extract the hot information according to the hot type of the article, and the hot information extraction module is connected to the heat comprehensive analysis module.

[0091] Specifically, the hot information extraction module in this embodiment extracts article information of article hot types of category two and category three as hot information.

[0092] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the embodiments here. All obvious changes or modifications derived from the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A hot information extraction system based on big data, characterized in that: include: The platform data collection module is used to collect article information and user information on the social platform, and periodically collect article access information; A data numbering construction module is used to number the article information and user information to obtain the article number and user number; An article classification processing module is used to classify the article information to obtain the article category of the article information; User data analysis module, used to analyze user preference characteristics and user interaction quality based on user information, article number, user number and article category; Article data analysis module, used to analyze content relevance and article content characteristics based on article information and article categories; Access data analysis module, used to analyze article access characteristics based on article access information, content relevance and article content characteristics; Comprehensive heat analysis module, used to analyze article hot spots based on article access characteristics, user preference characteristics and user interaction quality; The hot information extraction module is used to extract hot information according to the hot type of the article.

2. The hot spot information extraction system based on big data according to claim 1 is characterized in that: The article classification processing module matches the article subject with preset category keywords, and analyzes the article category based on the matching results. If there are preset category keywords in the article subject, the article classification processing module sets the article category to be the same as the preset category corresponding to the preset category keywords. If there are no preset category keywords in the article subject, the article classification processing module sets the article category to no category.

3. The hot spot information extraction system based on big data according to claim 2 is characterized in that: The user data analysis module is provided with a user preference analysis unit, which is used to count the number of different article categories corresponding to the article information in the user's browsing information as the browsing category number N1(j), count the number of different article information corresponding to each article category in the user's browsing information as the browsing preference number N(j,k), and analyze the user preference characteristics according to the browsing category number and the browsing preference number to obtain the user preference characteristics H(j).

4. The hot spot information extraction system based on big data according to claim 3 is characterized in that: The user data analysis module is also provided with a user comment analysis unit, which is used to analyze the user interaction quality according to whether the user comment content is a relevant comment, the number of relevant comments N2(j,i) and the number of comment likes B(j,i) to obtain the user interaction quality Q(j,i).

5. The hot spot information extraction system based on big data according to claim 4 is characterized in that: The article data analysis module is provided with a content association analysis unit, which is used to remove words in the article subject and article content that are the same as preset simplified keywords to obtain a simplified subject and simplified content, and use the words in the simplified subject as association analysis keywords. The content association analysis unit extracts the number of times the association analysis keywords appear in the simplified content as the content association number N3(i), and analyzes the content association degree based on the content association number to obtain the content association degree P(i).

6. The hot spot information extraction system based on big data according to claim 5 is characterized in that: The article data analysis module is also provided with a content feature analysis unit, which is used to count the number of other article information that have the same article category as the currently analyzed article information and contain associated analysis keywords as the feature analysis number s(i), and analyze the article content features based on the feature analysis number and the content association number N3(i) to obtain the article content feature D(i).

7. The hot spot information extraction system based on big data according to claim 6 is characterized in that: The access data analysis module is provided with an access feature analysis unit, which is used to analyze the article access features according to the access volume W(i, t) and the article publishing time T to obtain the article access features A(i, t).

8. The hot spot information extraction system based on big data according to claim 7 is characterized in that: The access data analysis module is further provided with an access association analysis unit, which is used to compare the content association P(i) with the association threshold p, and adjust the analysis process of the article access feature according to the comparison result. When the content association does not meet the threshold, the analysis process of the article access feature is adjusted, and the adjusted article access feature is A1(i,t); The access data analysis module is also provided with an article content analysis unit to compare the article content feature D(i) with the content feature threshold d, and optimize the adjustment process of the article access feature according to the comparison result. When the article content feature meets the threshold, the adjustment process of the article access feature is optimized, and the optimized article access feature is A2(i,t).

9. The hot spot information extraction system based on big data according to claim 8 is characterized in that: The heat comprehensive analysis module is provided with a heat comprehensive analysis unit, which is used to analyze the article heat parameters according to the article access feature A(i, t), the user preference feature H(j) and the user interaction quality Q(j, i) to obtain the article heat parameter F(i, t).

10. The hot spot information extraction system based on big data according to claim 9, characterized in that: The comprehensive heat analysis module is also provided with a hotspot type judgment unit, which is used to compare the article heat parameter F(i, t) with the heat thresholds f1 and f2, and judge the article hotspot type according to the comparison result. The article hotspot types include one, two and three categories.

Images