Hotspot index pushing method based on user behavior analysis

Abstract

The invention discloses a hotspot index pushing method based on user behavior analysis, and the method comprises the steps: obtaining the basic information of non-visited users, determining a sharing authority level and sharable users, obtaining the historical behavior information of each sharable user in a historical period, according to the historical behavior information, determining and analyzing a rule index, a target time interval of the rule index, an irregular index and a target application evaluation value of the irregular index, and according to an analysis result, obtaining a hotspot index of an unaccessed user in the current time period; the method comprises the following steps: acquiring a real-time alarm index of a current time period, determining an associated alarm index according to an association relationship between the real-time alarm index and a hotspot index, and generating an index pushing sequence of a non-accessed user in the current time period based on the hotspot index and the associated alarm index, thereby improving the index pushing effect of the non-accessed user. And perception requirements of non-accessed users in different time periods are met.

Description

Technical Field

[0001] The present application relates to the technical field of hotspot indicator push technology, and in particular to a hotspot indicator push method based on user behavior analysis. Background Art

[0002] With the rapid development of information technology, users can intuitively and accurately view the operating status of the fuel supply system of group companies and power plants in real time by selecting corresponding indicators. By providing the ability to dynamically push indicators, users' personalized perception needs can be met.

[0003] In the existing technology, hot spots are pushed by analyzing the historical access data or concerned indicators of visited users, but the indicator push effect for non-visited customers is poor, and thus cannot meet the perception needs of corresponding users in different time periods. Summary of the Invention

[0004] To solve the above technical problems, the present application provides a hotspot indicator push method based on user behavior analysis. By determining the shareable users of non-visited users and analyzing the historical behavior information of the shareable users, regular indicators, target time intervals of regular indicators, irregular indicators and target application evaluation values ​​of irregular indicators are obtained, and the hotspot indicators of the current time period are determined. According to the correlation between the real-time alarm indicators and the hotspot indicators, the associated alarm indicators are determined, and an indicator push sequence of the hotspot indicators and the associated alarm indicators is generated, thereby improving the indicator push effect of non-visited users and meeting the perception needs of non-visited users in different time periods.

[0005] In some embodiments of the present application, a method for pushing hotspot indicators based on user behavior analysis is provided, including: Obtain basic information of users who have not accessed the site, determine the sharing permission level of the users who have not accessed the site and the users who can share the site based on the basic information, and obtain historical behavior information of each user who can share the site in a historical period, wherein the historical behavior information includes the historical number of visits and historical stay duration of preset indicators; Analyze historical behavior information over historical periods to determine regular indicators, target time intervals for regular indicators, irregular indicators, and target application evaluation values ​​for irregular indicators; Analyze the regularity indicator, the target time interval of the regularity indicator, the irregularity indicator, and the target application evaluation value of the irregularity indicator, and obtain the hotspot indicator of the non-visited users in the current period based on the analysis results; Obtain the real-time alarm indicators for the current period, determine the associated alarm indicators based on the correlation between the real-time alarm indicators and the hot indicators, and generate an indicator push sequence for non-visited users in the current period based on the hot indicators and the associated alarm indicators.

[0006] In some embodiments of the present application, determining the sharing permission level of the non-access user and the shareable users based on the basic information includes: Obtaining basic information of the non-visited user, including the non-visited user's department importance, role level, length of employment, and work performance; Generate the initial sharing permission level of the unaccessed user based on the department importance and role level of the unaccessed user, and generate the credibility of the unaccessed user based on the length of time the unaccessed user has been employed and work performance. Generate a sharing permission level for users who have not accessed the site based on the initial sharing permission level and credibility; Perform similarity analysis based on the basic information of non-visited users and the basic information of visited users to obtain the similarity level; The shareable users of the non-accessed user are obtained according to the similarity between the non-accessed user and the accessed user and the sharing permission level of the non-accessed user.

[0007] In some embodiments of the present application, analyzing historical behavior information of a historical period to determine regularity indicators and target time intervals of regularity indicators includes: Analyze the historical behavior information of each sharable user in the historical period, determine the historical number of visits of each preset indicator of each sharable user in the historical period, and obtain the historical time interval between adjacent historical visit numbers of each preset indicator; Analyze the historical time intervals between multiple adjacent historical access times of each preset indicator, and determine whether there is a time period relationship between the multiple historical time intervals corresponding to the preset indicator based on the analysis results; If there is a time period relationship and the occurrence ratio of the same preset indicator with a time period relationship is greater than the preset ratio threshold, the corresponding preset indicator is set as a regular indicator; Obtain the historical time intervals of the same regularity indicator for all sharable users, and assign corresponding weight coefficients to the historical time intervals of the same regularity indicator according to the similarity between the sharable users and the non-accessed users; A target time interval of the corresponding regular indicator is generated based on the historical time intervals of the same regular indicator of all sharable users and the corresponding weight coefficients.

[0008] In some embodiments of the present application, the irregularity indicator and the target application evaluation value of the irregularity indicator include: If there is no time period relationship, set the corresponding preset indicator as an irregular indicator; Consider the historical period as multiple single-day periods, divide each single-day period into multiple single-day periods based on the preset time interval, and construct the single-day period matrix T; ; in, is the jth single-day period in the i-th single-day period, i=1,2,…m, j=1,2,…n, m is the total number of single-day periods in the historical period; Obtain the historical number of visits to the irregular indicator of each shareable user in each single day period and the historical length of stay corresponding to the historical number of visits, and generate a first application evaluation value of the corresponding irregular indicator in each single day period based on the historical number of visits and the historical length of stay corresponding to the historical number of visits; The weight coefficient of the shareable user is set according to the similarity between the shareable user and the non-accessed user, and the second application evaluation value of the corresponding irregular indicator in each single day period is generated according to the first application evaluation value of each irregular indicator of all shareable users in each single day period and the weight coefficient of the corresponding shareable user; The second application evaluation values ​​of the irregularity indicator in multiple single-day periods in the same preset time interval are averaged to obtain the target application evaluation value of the corresponding irregularity indicator in each preset time interval.

[0009] In some embodiments of the present application, the calculation formula for the first application evaluation value is: ; Where Y1 is the first application evaluation value, z is the historical visit count of the irregular indicator in the corresponding single day time period, Ls is the historical stay time of the sth historical visit count of the irregular indicator in the corresponding single day time period, y1 is the first application conversion coefficient, and y2 is the second application conversion coefficient; The calculation formula of the second application evaluation value is: Wherein, Y2 is the second application evaluation value, a is the total number of sharable users, Y1c is the first application evaluation value of the irregularity index of the c-th sharable user in the corresponding single day period, and qc is the weight coefficient of the c-th sharable user; The calculation formula of the target application evaluation value is: ; Among them, Y0 is the target application evaluation value, and Y2i is the second application evaluation value of the irregular indicator in the i-th single day period corresponding to the preset time interval.

[0010] In some embodiments of the present application, the regularity indicator, the target time interval of the regularity indicator, the irregularity indicator, and the target application evaluation value of the irregularity indicator are analyzed, and the hotspot index of the non-visited users in the current period is obtained according to the analysis results, including: Obtain the previous adjacent historical access time node of each regular indicator, and predict the next access time node of the corresponding regular indicator based on the previous adjacent historical access time node and the target time interval of the corresponding regular indicator; Determine whether the next access time node of each regular indicator is in the current time period. If so, set the corresponding regular indicator as the hot indicator of non-visited users in the current time period, and set the label of the corresponding hot indicator as the period label; Determine the preset time interval in which the current time period is located, and obtain the target application evaluation value of each irregular indicator in the corresponding preset time interval; Presetting a preset application evaluation value threshold; If the target application evaluation value in the corresponding preset time interval is greater than the preset application evaluation value threshold, the corresponding irregular indicator is set as the hotspot indicator of non-visited users in the current period, and the label of the corresponding hotspot indicator is set as the hotspot label.

[0011] In some embodiments of the present application, determining the associated alarm indicator based on the association relationship between the real-time alarm indicator and the hotspot indicator includes: Obtain the real-time alarm indicators for the current period and extract the real-time alarm content parameters of each real-time alarm indicator; Obtain the first characteristic parameter of the hotspot indicator with a set period label; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the first characteristic parameter of each hotspot indicator with a set periodic label to obtain a first correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the first correlation coefficient is greater than a preset first correlation coefficient threshold, a first correlation identifier of the real-time alarm indicator and the corresponding hotspot indicator is generated, and a first correlation coefficient difference of the first correlation identifier is calculated; Generate a first comprehensive correlation coefficient between the corresponding real-time alarm indicator and the hotspot indicator with a set periodic label according to the number of first correlation identifiers and the corresponding first correlation coefficient difference; Obtaining a second characteristic parameter of a hotspot indicator having a hotspot label set therein; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the second characteristic parameter of each hotspot indicator set with a hotspot label to obtain a second correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the second correlation coefficient is greater than the preset second correlation coefficient threshold, a second correlation identifier of the real-time alarm indicator and the corresponding hotspot indicator is generated, and a second correlation coefficient difference of the second correlation identifier is calculated; Generating a second comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hotspot indicator set with the hotspot label according to the number of the second correlation identifiers and the corresponding second correlation coefficient difference; Generate a comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hot spot indicator according to the first comprehensive correlation coefficient and the second comprehensive correlation coefficient; Pre-set comprehensive correlation coefficient threshold; A real-time alarm indicator whose comprehensive correlation coefficient is greater than a comprehensive correlation coefficient threshold is set as a correlation alarm indicator, and a label of the correlation alarm indicator is set as an alarm label.

[0012] In some embodiments of the present application, the calculation formula of the comprehensive correlation coefficient is: ; Among them, G is the comprehensive correlation coefficient, g1 is the weight coefficient of the first comprehensive correlation coefficient, w1 is the number of first correlation identifiers, and w2 is the total number of hot spot indicators with periodic labels. is the first correlation coefficient difference of the vth first correlation identifier, d1v is the weight coefficient of the vth first correlation identifier, g2 is the weight coefficient of the second comprehensive correlation coefficient, w3 is the number of second correlation identifiers, and w4 is the total number of hotspot indicators with hotspot labels. is the second correlation coefficient difference of the sth second correlation identifier, and d2s is the weight coefficient of the sth second correlation identifier.

[0013] In some embodiments of the present application, generating an indicator push sequence of non-visited users in the current period based on hot spot indicators and associated alarm indicators includes: The hot indicators with periodic labels are set to construct a regular indicator set, and the push order of each hot indicator in the regular indicator set is set according to the target time interval, and the regular indicator push sequence B1, B1 (b11, b12, ..., b1 j1 ), where b1 i1 is the i1th hot indicator in the regular indicator push sequence, j1 is the number of hot indicators in the regular indicator push sequence, i1=1,2,…,j1; The hot indicators with hot labels are set to construct an irregular indicator set, and the push order of each hot indicator in the irregular indicator set is set according to the target application evaluation value, and the irregular indicator push sequence B2, B2 (b21, b22, ..., b2 j2 ), where b1 i2 is the i2th hot indicator in the irregular indicator push sequence, j2 is the number of hot indicators in the irregular indicator push sequence, i2=1,2,…,j2; The associated alarm indicators with alarm labels are set to construct an alarm indicator set, and the push order of each associated alarm indicator in the alarm indicator set is set according to the comprehensive correlation coefficient to obtain the alarm indicator push sequence B3, B3 (b31, b32, ..., b3 j3 ), where b3i3 is the i3th associated alarm indicator in the alarm indicator push sequence, j3 is the number of associated alarm indicators in the alarm indicator push sequence, i3=1,2,…,j3; Based on the combination of indicators with the same serial number in the regular indicator push sequence, hot indicator push sequence and alarm indicator push sequence, multiple groups of indicator push sets are obtained. The multiple groups of indicator push sets are (b11, b21, b31, b12, b22, b32, ..., b1 j1 、b2 j2 、b3 j3 ); Setting the order of indicator pushes in the next set of indicator pushes based on the indicator access status of non-visited users to the indicators in each set of indicator pushes, wherein the indicator access status includes the indicator access order, the number of indicator accesses, and the indicator stay time in each set of indicator pushes; Generate an indicator push sequence for non-visited users in the current period according to the indicator push order of multiple groups of indicator push sets.

[0014] In some embodiments of the present application, before obtaining the historical behavior information of each sharable user in the historical period, the method further includes: Obtain the initial sharing permission level and credibility of users who have not accessed the site; If the initial sharing permission level is greater than the preset permission level threshold or the credibility is less than the preset credibility threshold, a verification instruction of the feature information is sent to the corresponding non-accessed user; If the verification result of the verification command is normal, the sharing permission level of the non-accessed user will not be modified; If the verification structure of the verification quality assurance is abnormal, the sharing permission level of the non-accessed user is set to the sharing permission level threshold.

[0015] Compared with the prior art, the hotspot indicator push method based on user behavior analysis in the embodiment of the present application has the following beneficial effects: By determining the shareable users of non-visited users and analyzing the historical behavior information of shareable users, regular indicators, target time intervals of regular indicators, irregular indicators, and target application evaluation values ​​of irregular indicators are obtained, and the hot indicators of the current period are determined. According to the correlation between real-time alarm indicators and hot indicators, the associated alarm indicators are determined, and an indicator push sequence of hot indicators and associated alarm indicators is generated to improve the indicator push effect of non-visited users and meet the perception needs of non-visited users in different time periods. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a hotspot indicator push method based on user behavior analysis in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0017] The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0018] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0021] like Figure 1 As shown, the hotspot indicator push method based on user behavior analysis of the preferred embodiment of the present application includes: Step S101: Obtain basic information of non-accessed users, determine the sharing permission level of the non-accessed users and the users who can share based on the basic information, and obtain historical behavior information of each user who can share in a historical period, wherein the historical behavior information includes the historical number of visits and historical stay duration of preset indicators; Step S102: Analyze historical behavior information of a historical period to determine regularity indicators, target time intervals of regularity indicators, irregularity indicators, and target application evaluation values ​​of irregularity indicators; Step S103: analyzing the regularity index, the target time interval of the regularity index, the irregularity index, and the target application evaluation value of the irregularity index, and obtaining the hotspot index of non-visited users in the current period according to the analysis results; Step S104: obtaining the real-time alarm indicator of the current period, determining the associated alarm indicator according to the association relationship between the real-time alarm indicator and the hotspot indicator, and generating an indicator push sequence for non-visited users in the current period based on the hotspot indicator and the associated alarm indicator.

[0022] In this embodiment, the preset indicators refer to key indicators in multiple links of the fuel supply system, including but not limited to coal storage, coal consumption, coal quality structure, etc.

[0023] In this embodiment, a regular indicator refers to the existence of time periodicity in the historical time nodes of the historical number of visits to a preset indicator. For example, the corresponding preset management indicator is accessed at 8 o'clock every day in the historical period, or the corresponding preset indicator is accessed in a two-day time cycle. The target time interval refers to the time interval between adjacent visits to the regular indicator. An irregular indicator refers to the absence of time periodicity in the historical time nodes of the historical number of visits to the preset indicator. The target application evaluation value refers to the application evaluation value of the irregular indicator in each preset time interval.

[0024] In some embodiments of the present application, determining the sharing permission level of the non-access user and the shareable users based on the basic information includes: Obtaining basic information of the non-visited user, including the non-visited user's department importance, role level, length of employment, and work performance; Generate the initial sharing permission level of the unaccessed user based on the department importance and role level of the unaccessed user, and generate the credibility of the unaccessed user based on the length of time the unaccessed user has been employed and work performance. Generate a sharing permission level for users who have not accessed the site based on the initial sharing permission level and credibility; Perform similarity analysis based on the basic information of non-visited users and the basic information of visited users to obtain the similarity level; The shareable users of the non-accessed user are obtained according to the similarity between the non-accessed user and the accessed user and the sharing permission level of the non-accessed user.

[0025] In this embodiment, the importance of a department is set based on the number of associated sub-departments of the department and the importance of the business involved in the department. The more associated sub-departments there are and the greater the importance of the business involved, the greater the importance of the department. The role level refers to the position level of the role in the department. The higher the position, the greater the role level. The greater the importance of the department of the unaccessed user and the greater the role level, the greater the corresponding initial sharing permission level.

[0026] In this embodiment, the longer the non-accessed user has been employed and the better his / her work performance, the greater the corresponding credibility. Credibility refers to the degree of trustworthiness of the non-accessed user's initial sharing permission level. The initial sharing permission level is modified according to the credibility to obtain the sharing permission level.

[0027] In this embodiment, an accessed user whose similarity is greater than a preset similarity threshold and whose sharing permission level is greater than that of the accessed user is set as a shareable user of the non-accessed user, and the historical behavior information of the shareable user is analyzed to provide the non-accessed user with hot spot indicators of the current time period to meet the needs of real-time and accurate perception of the non-accessed user.

[0028] In some embodiments of the present application, analyzing historical behavior information of a historical period to determine regularity indicators and target time intervals of regularity indicators includes: Analyze the historical behavior information of each sharable user in the historical period, determine the historical number of visits of each preset indicator of each sharable user in the historical period, and obtain the historical time interval between adjacent historical visit numbers of each preset indicator; Analyze the historical time intervals between multiple adjacent historical access times of each preset indicator, and determine whether there is a time period relationship between the multiple historical time intervals corresponding to the preset indicator based on the analysis results; If there is a time period relationship and the occurrence ratio of the same preset indicator with a time period relationship is greater than the preset ratio threshold, the corresponding preset indicator is set as a regular indicator; Obtain the historical time intervals of the same regularity indicator for all sharable users, and assign corresponding weight coefficients to the historical time intervals of the same regularity indicator according to the similarity between the sharable users and the non-accessed users; A target time interval of the corresponding regular indicator is generated based on the historical time intervals of the same regular indicator of all sharable users and the corresponding weight coefficients.

[0029] In this embodiment, the time period relationship refers to the regularity of the time frequency of the number of visits. For example, if the historical time intervals between adjacent historical number of visits are all 10 hours, then there is regularity.

[0030] In this embodiment, the occurrence ratio = the number of times the same preset indicator with a time period relationship appears in the sharable users / the total number of sharable users. For example, if the same preset indicator with a time period relationship appears in 12 sharable users and the total number of sharable users is 20, the occurrence ratio is 3 / 5, and the preset ratio threshold is 1 / 2.

[0031] In this embodiment, by analyzing the historical number of visits in the historical behavior information, it is determined whether there is time periodicity, and the regular indicators and irregular indicators of each sharable user are determined. The historical time intervals of the sharable users with the same regular indicators are corrected, that is, the weight coefficients of the historical time intervals are set according to the degree of similarity, so as to calculate the historical time intervals of the same regular indicator and obtain the target time interval of the regular indicator. For example, five sharable users all have historical time intervals of the same regular indicator, and the historical time intervals are 2, 4, 5, 4, and 3 respectively. The weight coefficients of the sharable users are 0.3, 0.1, 0.2, 0.3, and 0.1 respectively, then the target time interval is 0.3*2+0.1*4+0.2*5+0.3*4+0.1*3=3.5.

[0032] In this embodiment, by determining regular indicators and irregular indicators, the must-click indicators and hot spot indicators of non-visiting users in the current period are accurately obtained, thereby improving the accuracy of perceiving the needs of non-visiting users.

[0033] In some embodiments of the present application, the irregularity indicator and the target application evaluation value of the irregularity indicator include: If there is no time period relationship, set the corresponding preset indicator as an irregular indicator; Consider the historical period as multiple single-day periods, divide each single-day period into multiple single-day periods based on the preset time interval, and construct the single-day period matrix T; ; in, is the jth single-day period in the i-th single-day period, i=1,2,…m, j=1,2,…n, m is the total number of single-day periods in the historical period; Obtain the historical number of visits to the irregular indicator of each shareable user in each single day period and the historical length of stay corresponding to the historical number of visits, and generate a first application evaluation value of the corresponding irregular indicator in each single day period based on the historical number of visits and the historical length of stay corresponding to the historical number of visits; The weight coefficient of the shareable user is set according to the similarity between the shareable user and the non-accessed user, and the second application evaluation value of the corresponding irregular indicator in each single day period is generated according to the first application evaluation value of each irregular indicator of all shareable users in each single day period and the weight coefficient of the corresponding shareable user; The second application evaluation values ​​of the irregularity indicator in multiple single-day periods in the same preset time interval are averaged to obtain the target application evaluation value of the corresponding irregularity indicator in each preset time interval.

[0034] In this embodiment, the preset time interval refers to a time period set in advance, for example, every two hours is a preset time interval, which can be set according to the actual access situation. By dividing the historical period into multiple single-day periods, and dividing the single-day period into multiple single-day periods, the target application evaluation value of the irregular indicator of each preset time interval is obtained by calculating the first application evaluation value and the second application evaluation value. The hot spot indicator of the irregular indicator in each preset time interval is determined according to the target application evaluation value, the accuracy of the hot spot indicator is improved, and accurate hot spot push is performed for non-visited customers to meet the user's perception needs.

[0035] In some embodiments of the present application, the calculation formula for the first application evaluation value is: ; Where Y1 is the first application evaluation value, z is the historical visit count of the irregular indicator in the corresponding single day time period, Ls is the historical stay time of the sth historical visit count of the irregular indicator in the corresponding single day time period, y1 is the first application conversion coefficient, and y2 is the second application conversion coefficient; The calculation formula of the second application evaluation value is: Wherein, Y2 is the second application evaluation value, a is the total number of sharable users, Y1c is the first application evaluation value of the irregularity index of the c-th sharable user in the corresponding single day period, and qc is the weight coefficient of the c-th sharable user; The calculation formula of the target application evaluation value is: ; Among them, Y0 is the target application evaluation value, and Y2i is the second application evaluation value of the irregular indicator in the i-th single day period corresponding to the preset time interval.

[0036] In some embodiments of the present application, the regularity indicator, the target time interval of the regularity indicator, the irregularity indicator, and the target application evaluation value of the irregularity indicator are analyzed, and the hotspot index of the non-visited users in the current period is obtained according to the analysis results, including: Obtain the previous adjacent historical access time node of each regular indicator, and predict the next access time node of the corresponding regular indicator based on the previous adjacent historical access time node and the target time interval of the corresponding regular indicator; Determine whether the next access time node of each regular indicator is in the current time period. If so, set the corresponding regular indicator as the hot indicator of non-visited users in the current time period, and set the label of the corresponding hot indicator as the period label; Determine the preset time interval in which the current time period is located, and obtain the target application evaluation value of each irregular indicator in the corresponding preset time interval; Presetting a preset application evaluation value threshold; If the target application evaluation value in the corresponding preset time interval is greater than the preset application evaluation value threshold, the corresponding irregular indicator is set as the hotspot indicator of non-visited users in the current period, and the label of the corresponding hotspot indicator is set as the hotspot label.

[0037] In this embodiment, by predicting the next access time node of the regular indicator, it is determined whether the next access time node is in the current time period, and the hot index of the periodic label of the non-visited user in the current time period is obtained. By analyzing the target application evaluation value of the irregular indicator, the high heat index of the sharable user in the current time period is determined, and the hot index of the hot label of the non-visited user in the current time period is obtained to meet the perception needs of the non-visited user in the current time period.

[0038] In some embodiments of the present application, determining the associated alarm indicator based on the association relationship between the real-time alarm indicator and the hotspot indicator includes: Obtain the real-time alarm indicators for the current period and extract the real-time alarm content parameters of each real-time alarm indicator; Obtain the first characteristic parameter of the hotspot indicator with a set period label; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the first characteristic parameter of each hotspot indicator with a set periodic label to obtain a first correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the first correlation coefficient is greater than a preset first correlation coefficient threshold, a first correlation identifier of the real-time alarm indicator and the corresponding hotspot indicator is generated, and a first correlation coefficient difference of the first correlation identifier is calculated; Generate a first comprehensive correlation coefficient between the corresponding real-time alarm indicator and the hotspot indicator with a set periodic label according to the number of first correlation identifiers and the corresponding first correlation coefficient difference; Obtaining a second characteristic parameter of a hotspot indicator having a hotspot label set therein; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the second characteristic parameter of each hotspot indicator set with a hotspot label to obtain a second correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the second correlation coefficient is greater than the preset second correlation coefficient threshold, a second correlation identifier of the real-time alarm indicator and the corresponding hotspot indicator is generated, and a second correlation coefficient difference of the second correlation identifier is calculated; Generating a second comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hotspot indicator set with the hotspot label according to the number of the second correlation identifiers and the corresponding second correlation coefficient difference; Generate a comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hot spot indicator according to the first comprehensive correlation coefficient and the second comprehensive correlation coefficient; Pre-set comprehensive correlation coefficient threshold; A real-time alarm indicator whose comprehensive correlation coefficient is greater than a comprehensive correlation coefficient threshold is set as a correlation alarm indicator, and a label of the correlation alarm indicator is set as an alarm label.

[0039] In this embodiment, the first characteristic parameter and the second characteristic parameter refer to high-frequency search parameters or key parameters of a hotspot index with a periodic tag and a hotspot index with a hotspot tag, respectively.

[0040] In this embodiment, the number of first association identifiers refers to the number of hot spot indicators with periodic labels that have first association identifiers with corresponding real-time alarm indicators, and the number of second association identifiers refers to the number of hot spot indicators with hot spot labels that have second association identifiers with corresponding real-time alarm indicators.

[0041] In this embodiment, by respectively calculating the first comprehensive correlation coefficient between the real-time alarm indicator and the hot spot indicator with a periodic label and the second comprehensive correlation coefficient between the real-time alarm indicator and the hot spot indicator with a hot spot label, the comprehensive correlation coefficient between the real-time alarm indicator and the hot spot indicator is obtained, thereby improving the accuracy of the judgment of the correlation between the real-time alarm indicator and the hot spot indicator, screening out related alarm indicators and pushing them, reducing information interference from other unrelated alarm indicators, and improving user satisfaction with indicator push.

[0042] In some embodiments of the present application, the calculation formula of the comprehensive correlation coefficient is: ; Among them, G is the comprehensive correlation coefficient, g1 is the weight coefficient of the first comprehensive correlation coefficient, w1 is the number of first correlation identifiers, and w2 is the total number of hot spot indicators with periodic labels. is the first correlation coefficient difference of the vth first correlation identifier, d1v is the weight coefficient of the vth first correlation identifier, g2 is the weight coefficient of the second comprehensive correlation coefficient, w3 is the number of second correlation identifiers, and w4 is the total number of hotspot indicators with hotspot labels. is the second correlation coefficient difference of the sth second correlation identifier, and d2s is the weight coefficient of the sth second correlation identifier.

[0043] In this embodiment, is the first comprehensive correlation coefficient, is the second comprehensive correlation coefficient.

[0044] In some embodiments of the present application, generating an indicator push sequence of non-visited users in the current period based on hot spot indicators and associated alarm indicators includes: The hot indicators with periodic labels are set to construct a regular indicator set, and the push order of each hot indicator in the regular indicator set is set according to the target time interval, and the regular indicator push sequence B1, B1 (b11, b12, ..., b1 j1 ), where b1 i1 is the i1th hot indicator in the regular indicator push sequence, j1 is the number of hot indicators in the regular indicator push sequence, i1=1,2,…,j1; The hot indicators with hot labels are set to construct an irregular indicator set, and the push order of each hot indicator in the irregular indicator set is set according to the target application evaluation value, and the irregular indicator push sequence B2, B2 (b21, b22, ..., b2 j2 ), where b1 i2 is the i2th hot indicator in the irregular indicator push sequence, j2 is the number of hot indicators in the irregular indicator push sequence, i2=1,2,…,j2; The associated alarm indicators with alarm labels are set to construct an alarm indicator set, and the push order of each associated alarm indicator in the alarm indicator set is set according to the comprehensive correlation coefficient to obtain the alarm indicator push sequence B3, B3 (b31, b32, ..., b3 j3 ), where b3 i3 is the i3th associated alarm indicator in the alarm indicator push sequence, j3 is the number of associated alarm indicators in the alarm indicator push sequence, i3=1,2,…,j3; Based on the combination of indicators with the same serial number in the regular indicator push sequence, hot indicator push sequence and alarm indicator push sequence, multiple groups of indicator push sets are obtained. The multiple groups of indicator push sets are (b11, b21, b31, b12, b22, b32, ..., b1 j1 、b2 j2 、b3 j3 ); Setting the order of indicator pushes in the next set of indicator pushes based on the indicator access status of non-visited users to the indicators in each set of indicator pushes, wherein the indicator access status includes the indicator access order, the number of indicator accesses, and the indicator stay time in each set of indicator pushes; Generate an indicator push sequence for non-visited users in the current period according to the indicator push order of multiple groups of indicator push sets.

[0045] In this embodiment, the smaller the target time interval, the higher the push order of the corresponding hot spot indicator in the regular indicator set; the greater the target application evaluation value, the higher the push order of the corresponding hot spot indicator in the hot spot indicator set; the greater the comprehensive correlation coefficient, the higher the push order of the corresponding correlated alarm indicator in the alarm indicator set.

[0046] In this embodiment, the indicator push order in the next adjacent indicator push set is set by the indicator access order, indicator access times and indicator stay time in each group of indicator push sets, so as to obtain the indicator push sequence for the current time period, thereby improving the satisfaction of non-visited users with the hot indicator push.

[0047] In some embodiments of the present application, before obtaining the historical behavior information of each sharable user in the historical period, the method further includes: Obtain the initial sharing permission level and credibility of users who have not accessed the site; If the initial sharing permission level is greater than the preset permission level threshold or the credibility is less than the preset credibility threshold, a verification instruction of the feature information is sent to the corresponding non-accessed user; If the verification result of the verification command is normal, the sharing permission level of the non-accessed user will not be modified; If the verification structure of the verification quality assurance is abnormal, the sharing permission level of the non-accessed user is set to the sharing permission level threshold.

[0048] In this embodiment, the sharing permission level threshold refers to the lowest level of sharing permission. Feature information includes but is not limited to face, identity verification and other information. When the initial sharing permission level is too high or the credibility is low, non-accessed users are verified to ensure the security of key information on fuel supply in companies and power plants and historical access data of accessed customers, thereby reducing the probability of leakage of company information and user privacy.

[0049] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and replacements can be made without departing from the technical principles of the present application. These improvements and replacements should also be regarded as the scope of protection of the present application.

Claims

1. A hotspot indicator push method based on user behavior analysis is characterized by: include: Obtain basic information of users who have not accessed the site, determine the sharing permission level of the users who have not accessed the site and the users who can share the site based on the basic information, and obtain historical behavior information of each user who can share the site in a historical period, wherein the historical behavior information includes the historical number of visits and historical stay duration of preset indicators; Analyze historical behavior information over historical periods to determine regular indicators, target time intervals for regular indicators, irregular indicators, and target application evaluation values ​​for irregular indicators; Analyze the regularity indicator, the target time interval of the regularity indicator, the irregularity indicator, and the target application evaluation value of the irregularity indicator, and obtain the hotspot indicator of the non-visited users in the current period based on the analysis results; Obtain the real-time alarm indicators for the current period, determine the associated alarm indicators based on the correlation between the real-time alarm indicators and the hot indicators, and generate an indicator push sequence for non-visited users in the current period based on the hot indicators and the associated alarm indicators.

2. The method for pushing hotspot indicators based on user behavior analysis according to claim 1, characterized in that: Determine the sharing permission level of users who have not accessed the service and the users who can share the service based on basic information, including: Obtaining basic information of the non-visited user, including the non-visited user's department importance, role level, length of employment, and work performance; Generate the initial sharing permission level of the unaccessed user based on the department importance and role level of the unaccessed user, and generate the credibility of the unaccessed user based on the length of time the unaccessed user has been employed and work performance. Generate a sharing permission level for users who have not accessed the site based on the initial sharing permission level and credibility; Perform similarity analysis based on the basic information of non-visited users and the basic information of visited users to obtain the similarity level; The shareable users of the non-accessed user are obtained according to the similarity between the non-accessed user and the accessed user and the sharing permission level of the non-accessed user.

3. The method for pushing hotspot indicators based on user behavior analysis according to claim 2, characterized in that: Analyze historical behavior information over historical periods to determine regularity indicators and target time intervals for regularity indicators, including: Analyze the historical behavior information of each sharable user in the historical period, determine the historical number of visits of each preset indicator of each sharable user in the historical period, and obtain the historical time interval between adjacent historical visits of each preset indicator; Analyze the historical time intervals between multiple adjacent historical access times of each preset indicator, and determine whether there is a time period relationship between the multiple historical time intervals corresponding to the preset indicator based on the analysis results; If there is a time period relationship and the occurrence ratio of the same preset indicator with a time period relationship is greater than the preset ratio threshold, the corresponding preset indicator is set as a regular indicator; Obtain the historical time intervals of the same regularity indicator for all sharable users, and assign corresponding weight coefficients to the historical time intervals of the same regularity indicator according to the similarity between the sharable users and the non-accessed users; A target time interval of the corresponding regular indicator is generated based on the historical time intervals of the same regular indicator of all sharable users and the corresponding weight coefficients.

4. The method for pushing hotspot indicators based on user behavior analysis according to claim 3, characterized in that: The irregular indicators and target application evaluation values ​​of the irregular indicators include: If there is no time period relationship, set the corresponding preset indicator as an irregular indicator; Consider the historical period as multiple single-day periods, divide each single-day period into multiple single-day periods based on the preset time interval, and construct the single-day period matrix T; ； in, is the jth single-day period in the i-th single-day period, i=1,2,…m, j=1,2,…n, m is the total number of single-day periods in the historical period; Obtain the historical number of visits to the irregular indicator of each shareable user in each single day period and the historical length of stay corresponding to the historical number of visits, and generate a first application evaluation value of the corresponding irregular indicator in each single day period based on the historical number of visits and the historical length of stay corresponding to the historical number of visits; The weight coefficient of the shareable user is set according to the similarity between the shareable user and the non-accessed user, and the second application evaluation value of the corresponding irregular indicator in each single day period is generated according to the first application evaluation value of each irregular indicator of all shareable users in each single day period and the weight coefficient of the corresponding shareable user; The second application evaluation values ​​of the irregularity indicator in multiple single-day periods in the same preset time interval are averaged to obtain the target application evaluation value of the corresponding irregularity indicator in each preset time interval.

5. The method for pushing hotspot indicators based on user behavior analysis according to claim 4, characterized in that: The calculation formula of the first application evaluation value is: ； Where Y1 is the first application evaluation value, z is the historical visit count of the irregular indicator in the corresponding single day time period, Ls is the historical stay time of the sth historical visit count of the irregular indicator in the corresponding single day time period, y1 is the first application conversion coefficient, and y2 is the second application conversion coefficient; The calculation formula of the second application evaluation value is: Wherein, Y2 is the second application evaluation value, a is the total number of sharable users, Y1c is the first application evaluation value of the irregularity index of the c-th sharable user in the corresponding single day period, and qc is the weight coefficient of the c-th sharable user; The calculation formula of the target application evaluation value is: ; Among them, Y0 is the target application evaluation value, and Y2i is the second application evaluation value of the irregular indicator in the i-th single day period corresponding to the preset time interval.

6. The method for pushing hotspot indicators based on user behavior analysis according to claim 5, characterized in that: Analyze the regularity indicator, the target time interval of the regularity indicator, the irregularity indicator, and the target application evaluation value of the irregularity indicator. Based on the analysis results, obtain the hotspot indicators of non-visited users in the current period, including: Obtain the previous adjacent historical access time node of each regular indicator, and predict the next access time node of the corresponding regular indicator based on the previous adjacent historical access time node and the target time interval of the corresponding regular indicator; Determine whether the next access time node of each regular indicator is in the current time period. If so, set the corresponding regular indicator as the hot indicator of non-visited users in the current time period, and set the label of the corresponding hot indicator as the period label; Determine the preset time interval in which the current time period is located, and obtain the target application evaluation value of each irregular indicator in the corresponding preset time interval; Presetting a preset application evaluation value threshold; If the target application evaluation value in the corresponding preset time interval is greater than the preset application evaluation value threshold, the corresponding irregular indicator is set as the hotspot indicator of non-visited users in the current period, and the label of the corresponding hotspot indicator is set as the hotspot label.

7. The method for pushing hotspot indicators based on user behavior analysis according to claim 6, characterized in that: Determine the associated alarm indicators based on the correlation between real-time alarm indicators and hot spot indicators, including: Obtain the real-time alarm indicators for the current period and extract the real-time alarm content parameters of each real-time alarm indicator; Obtain the first characteristic parameter of the hotspot indicator with a set period label; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the first characteristic parameter of each hotspot indicator with a set periodic label to obtain a first correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the first correlation coefficient is greater than a preset first correlation coefficient threshold, a first correlation identifier of the real-time alarm indicator and the corresponding hotspot indicator is generated, and a first correlation coefficient difference of the first correlation identifier is calculated; Generate a first comprehensive correlation coefficient between the corresponding real-time alarm indicator and the hotspot indicator with a set periodic label according to the number of first correlation identifiers and the corresponding first correlation coefficient difference; Obtaining a second characteristic parameter of a hotspot indicator having a hotspot label set therein; Performing correlation analysis on the real-time alarm content parameter of each real-time alarm indicator and the second characteristic parameter of each hotspot indicator set with a hotspot label to obtain a second correlation coefficient between the real-time alarm indicator and the corresponding hotspot indicator; If the second correlation coefficient is greater than the preset second correlation coefficient threshold, a second correlation identifier of the real-time alarm indicator and the corresponding hot spot indicator is generated, and a second correlation coefficient difference of the second correlation identifier is calculated; Generating a second comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hotspot indicator set with the hotspot label according to the number of the second correlation identifiers and the corresponding second correlation coefficient difference; Generate a comprehensive correlation coefficient corresponding to the real-time alarm indicator and the hot spot indicator according to the first comprehensive correlation coefficient and the second comprehensive correlation coefficient; Pre-set comprehensive correlation coefficient threshold; A real-time alarm indicator whose comprehensive correlation coefficient is greater than a comprehensive correlation coefficient threshold is set as a correlation alarm indicator, and a label of the correlation alarm indicator is set as an alarm label.

8. The method for pushing hotspot indicators based on user behavior analysis according to claim 7, characterized in that: The calculation formula of the comprehensive correlation coefficient is: ； Among them, G is the comprehensive correlation coefficient, g1 is the weight coefficient of the first comprehensive correlation coefficient, w1 is the number of first correlation identifiers, and w2 is the total number of hot spot indicators with periodic labels. is the first correlation coefficient difference of the vth first correlation identifier, d1v is the weight coefficient of the vth first correlation identifier, g2 is the weight coefficient of the second comprehensive correlation coefficient, w3 is the number of second correlation identifiers, and w4 is the total number of hotspot indicators with hotspot labels. is the second correlation coefficient difference of the sth second correlation identifier, and d2s is the weight coefficient of the sth second correlation identifier.

9. The method for pushing hotspot indicators based on user behavior analysis according to claim 8, characterized in that: Generates a push sequence of indicators for non-visited users in the current period based on hot indicators and associated alarm indicators, including: The hot indicators with periodic labels are set to construct a regular indicator set, and the push order of each hot indicator in the regular indicator set is set according to the target time interval, and the regular indicator push sequence B1, B1 (b11, b12, ..., b1 j1 ), where b1 i1 is the i1th hot indicator in the regular indicator push sequence, j1 is the number of hot indicators in the regular indicator push sequence, i1=1,2,…,j1; The hot indicators with hot labels are set to construct an irregular indicator set, and the push order of each hot indicator in the irregular indicator set is set according to the target application evaluation value, and the irregular indicator push sequence B2, B2 (b21, b22, ..., b2 j2 ), where b1 i2 is the i2th hot indicator in the irregular indicator push sequence, j2 is the number of hot indicators in the irregular indicator push sequence, i2=1,2,…,j2; The associated alarm indicators with alarm labels are set to construct an alarm indicator set, and the push order of each associated alarm indicator in the alarm indicator set is set according to the comprehensive correlation coefficient to obtain the alarm indicator push sequence B3, B3 (b31, b32, ..., b3 j3 ), where b3 i3 is the i3th associated alarm indicator in the alarm indicator push sequence, j3 is the number of associated alarm indicators in the alarm indicator push sequence, i3=1,2,…,j3; Based on the combination of indicators with the same serial number in the regular indicator push sequence, hot indicator push sequence and alarm indicator push sequence, multiple groups of indicator push sets are obtained. The multiple groups of indicator push sets are (b11, b21, b31, b12, b22, b32, ..., b1 j1 、b2 j2 、b3 j3 ); Setting the order of indicator pushes in the next set of indicator pushes based on the indicator access status of non-visited users to the indicators in each set of indicator pushes, wherein the indicator access status includes the indicator access order, the number of indicator accesses, and the indicator stay time in each set of indicator pushes; Generate an indicator push sequence for non-visited users in the current period according to the indicator push order of multiple groups of indicator push sets.

10. The method for pushing hotspot indicators based on user behavior analysis according to claim 2, characterized in that: Before obtaining the historical behavior information of each sharable user in the historical period, it also includes: Obtain the initial sharing permission level and credibility of users who have not accessed the site; If the initial sharing permission level is greater than the preset permission level threshold or the credibility is less than the preset credibility threshold, a verification instruction of the feature information is sent to the corresponding non-accessed user; If the verification result of the verification command is normal, the sharing permission level of the non-accessed user will not be modified; If the verification structure of the verification quality assurance is abnormal, the sharing permission level of the non-accessed user is set to the sharing permission level threshold.

Images