Association credibility evaluation method and device based on activity index

Abstract

The present invention discloses a method and device for evaluating association credibility based on activity index. The method for evaluating association credibility comprises: step S1: selecting and obtaining detailed spatiotemporal distribution information of elements within a fixed time range; step S2: obtaining multiple association activities through a time slice association analysis method based on the detailed spatiotemporal distribution information of elements; step S3: evaluating multiple association activities to obtain an activity index; step S4: evaluating association credibility based on the activity index.

Description

Technical Field

[0001] The present invention relates to a method and device for evaluating association credibility, and in particular, to a method and device for evaluating association credibility based on an activity index. Background Art

[0002] With the development of Internet technology, the problem of entity ID association is a problem of connecting the underlying ID of data, especially providing technical support for connecting the real identity and virtual identity in public security business scenarios. Existing technologies generally use the number of associations to reflect the credibility of the association. For example, the more times, the more credible the association. However, in actual use, it is found that although the above-mentioned credibility judgment based on the number of associations is direct and easy to calculate, the credibility obtained is not accurate. The actual situation is that the credibility of the association is highly correlated with the activity within a time range. For example: a person’s current mobile phone number has been active frequently in the past month, but the total number of associations is not as good as the previous account. In this case, the active and frequent association relationship in the recent period should be more credible.

[0003] Therefore, there is an urgent need to develop an association credibility assessment method and device based on activity indicators that overcome the above-mentioned defects. Summary of the invention

[0004] In view of the above problems, the present invention provides a method for evaluating the association credibility based on activity index, which includes:

[0005] Step S1: Select and obtain detailed information on the spatiotemporal distribution of elements within a fixed time range;

[0006] Step S2: obtaining multiple correlation activities through a time slice correlation analysis method based on the detailed information of the spatiotemporal distribution of the elements;

[0007] Step S3: Evaluate the multiple associated activities to obtain an activity index;

[0008] Step S4: Evaluate the association credibility based on the activity index.

[0009] In the above-mentioned association credibility evaluation method, the detailed information of the spatiotemporal distribution of the elements includes: element ID, associated element ID, total number of associated days, total number of associated times and collection time.

[0010] In the above-mentioned association credibility evaluation method, step S2 includes:

[0011] Step S21: calculating the overall association activity of each association element according to the total association days and the total association times;

[0012] Step S22: Calculate the time slice association activity of each association element according to the time slice and the total number of associations.

[0013] In the above-mentioned association credibility evaluation method, step S3 includes:

[0014] Step S31: obtaining a plurality of current activities based on the overall associated activity and the time slice associated activity;

[0015] Step S32: setting a plurality of scoring intervals corresponding to a plurality of the current activity levels;

[0016] Step S33: Mapping the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element.

[0017] In the above-mentioned association credibility evaluation method, the overall association activity is obtained by the following formula:

[0018]

[0019] The time slice associated activity includes: recent day associated activity, recent week associated activity, recent half month associated activity and recent month associated activity.

[0020] The present invention also provides an association credibility evaluation device based on activity index, which includes:

[0021] An information acquisition unit selects and obtains detailed information on the spatiotemporal distribution of elements within a fixed time range;

[0022] An activity calculation unit, which obtains a plurality of associated activities by a time slice association analysis method based on the detailed information of the spatiotemporal distribution of the elements;

[0023] An activity evaluation unit, which evaluates a plurality of the associated activities to obtain an activity index;

[0024] The association credibility evaluation unit evaluates the association credibility based on the activity index.

[0025] In the above-mentioned association credibility evaluation device, the detailed information of the spatiotemporal distribution of elements includes: element ID, associated element ID, total number of associated days, total number of associated times and collection time.

[0026] The above-mentioned association credibility evaluation device, wherein the activity calculation unit includes:

[0027] The overall association activity calculation module calculates the overall association activity of each association element according to the total number of association days and the total number of association times;

[0028] The time slice association activity calculation module calculates the time slice association activity of each association element according to the time slice and the total number of associations.

[0029] In the above-mentioned association credibility evaluation device, the activity evaluation unit includes:

[0030] A current activity obtaining module, which obtains a plurality of current activities based on the overall associated activity and the time slice associated activity;

[0031] A scoring interval setting module, which sets a plurality of scoring intervals corresponding to a plurality of current activity levels;

[0032] The activity index obtaining module maps the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element.

[0033] In the above-mentioned association credibility evaluation device, the overall association activity is obtained by the following formula:

[0034]

[0035] The time slice associated activity includes: recent day associated activity, recent week associated activity, recent half month associated activity and recent month associated activity.

[0036] In summary, compared with the prior art, the present invention has the following benefits: by introducing a time decay function to characterize the correlation between credibility and time in the association relationship, which we define as the calculation of the freshness index, the credibility of the association relationship can be better evaluated when switching pairs. For example, in a scenario where an account is used quite frequently in the past month, a more accurate evaluation result can be obtained.

[0037] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 is a flow chart of the method for evaluating the association credibility of the present invention;

[0040] Figure 2 for Figure 1 The step-by-step flow chart of step S2;

[0041] Figure 3 for Figure 1 A step-by-step flow chart of step S3;

[0042] Figure 4 It is a structural schematic diagram of the association credibility evaluation device of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The words “include,” “including,” “have,” “contain,” etc. used in this document are open-ended terms, meaning including but not limited to.

[0045] The term "plurality" as used herein includes "two" and "more than two".

[0046] Please refer to Figure 1 , Figure 1 FIG. 1 is a flow chart of the method for evaluating the association credibility of the present invention. Figure 1 As shown, the association credibility evaluation method of the present invention includes:

[0047] Step S1: Select and obtain detailed information on the spatiotemporal distribution of elements within a fixed time range;

[0048] Step S2: obtaining multiple correlation activities through a time slice correlation analysis method based on the detailed information of the spatiotemporal distribution of the elements;

[0049] Step S3: Evaluate the multiple associated activities to obtain an activity index;

[0050] Step S4: Evaluate the association credibility based on the activity index.

[0051] Specifically, association relationship: association relationship refers to the relationship between an entity and its belongings, such as the entity's ID card, mobile phone number, virtual account, license plate number, etc.; freshness: the freshness of the association relationship within a certain time range, which to a certain extent also reflects the credibility of the association. The present invention is based on the fact that the credibility of the association relationship is related to the freshness of time, and the credibility changes with the change of time. Therefore, the association relationship is characterized by constructing a time decay function to calculate the freshness of the association, such as the number change scenario.

[0052] The detailed information of the spatiotemporal distribution of the elements includes: element ID, associated element ID, total associated days, total associated times and collection time, wherein the total associated days and total associated times are the total number of days and the cumulative number of times recorded from the beginning of the association to the current time.

[0053] Specifically, for the query given to the query element and all other elements of the generated association data, these association data are given the total number of associations and the total number of association days, some data only give the total number of association days; the association data also gives the earliest collection time and the latest collection time, some data only give the latest collection time. The detailed information of the spatiotemporal distribution of elements of the present invention is shown in the following table:

[0054]

[0055] Among them, a general summary statistics is made for the total number of days. The data in the table represents the number of statistics given for each day. A1 is the feature ID, and B1-B4 are all related feature IDs. The ID can be any ID, such as an ID card, license plate, mobile phone number, etc.

[0056] For further information, please refer to Figure 2 , Figure 2 for Figure 1 The step-by-step flow chart of step S2 in FIG. Figure 2 As shown,

[0057] The step S2 includes:

[0058] Step S21: calculating the overall association activity of each association element according to the total association days and the total association times;

[0059] Step S22: Calculate the time slice association activity of each association element according to the time slice and the total number of associations.

[0060] This can be understood as a measure of the activity level of the association between elements over a period of time. The current data will record detailed information on the temporal and spatial distribution of elements. These detailed information are statistics on the number of associations on a daily basis. Based on the statistics on the number of associations on a day, we can complete time-slice-based association analysis. We call this index based on time-slice association analysis an activity index.

[0061] The current technology uses two sources of information to describe activity. One is the total number of times and total daily data statistics in the current association table; the other is based on the detailed information of the spatial and temporal distribution of elements, such as the detailed record of the number of associations counted by day in a month. The relevant indicators are calculated by using the time slice association analysis method.

[0062] Calculate the overall association activity based on the total number of days and total times of the association table:

[0063] Based on the detailed information of the spatiotemporal distribution of elements, the time slice association activity is calculated by counting the number of associations every day for one month:

[0064]

[0065] Further, please refer to Figure 3 , Figure 3 for Figure 1 The step-by-step flow chart of step S3 in FIG. Figure 3 As shown, the step S3 includes:

[0066] Step S31: obtaining a plurality of current activities based on the overall associated activity and the time slice associated activity;

[0067] Step S32: setting a plurality of scoring intervals corresponding to a plurality of the current activity levels;

[0068] Step S33: Mapping the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element.

[0069] Specifically, the overall activity calculation: we calculate the current activity of multiple related Range1, Range2, Range3, and Range4 according to the indicator to see whether the average daily frequency increases over time. The specific calculation is as follows:

[0070] Range1 = Index2-Index3, Range1 indicates whether the number of active times in the most recent day is greater than the average number of active times in the most recent week;

[0071] Range2 = Index3 - Index4, Range2 indicates whether the weekly activity is greater than the half-monthly activity;

[0072] Range3 = Index4-Index5, Range3 indicates whether the half-monthly activity is greater than the monthly activity

[0073] Range4=Index5-Index1, Range2 indicates whether the monthly active number is greater than the total active number.

[0074] Among them, in this embodiment, the best effect is when the current Range1>Range2>Range3>Range4, which is a better implementation method, so we need to evaluate the specific activity value according to the number of times the time slice is associated. We use a percentage system. The specific process is as follows:

[0075] 1) Set a basic score, here we assume 90 points, which can be adjusted according to business content during implementation;

[0076] 2) Set a minimum score, here we assume 50 points, which can be set according to actual conditions in implementation to reduce the score range;

[0077] 3) Set a certain score range for each Range. Note that the cumulative score of the four Ranges cannot exceed 100 and cannot be lower than a minimum score. Based on the above assumptions of a basic score of 90 and a minimum score of 50, we can determine the score range of each Range to be [-10, 2.5]. The specific algorithm is:

[0078] 4) Based on the actual value of Range, it is mapped to the score range [-10, 2.5]. If the Range result is a positive number and greater than 2.5, then 2.5 points are added to the overall basic score. If it is less than 2.5 points, the corresponding Range value is added. If the Range result is a negative number and less than -10, then -10 is added to the overall basic score. If it is less than -10 points, the corresponding Range value is added. The piecewise function is as follows:

[0079]

[0080] 5) Calculate the above steps for each Range respectively, and the result is the final activity Index, as shown below:

[0081] Index=f(f(f(f(basic score,Range4),Range3),Range2),Range1).

[0082] 6) Based on the incoming data, the value of each indicator is calculated according to the algorithm of each indicator and an overall score value is given. In this embodiment, the results are given to 2 decimal places. The activity indicators of the specific data output are shown in the following table:

[0083]

[0084] The output format can be an object that wraps a single piece of content, and finally a List collection is given, or it can be expressed as a Json structure.

[0085] Furthermore, according to the table, the activity index of A1->B1 is 100, the activity index of A1->B2 is 90, the freshness index of A1->B3 is 80, and the activity index of A1->B4 is 60. Therefore, based on the above results, it can be evaluated that the association credibility of A1->B1 is the highest, and the evaluation result is output.

[0086] Please refer to Figure 4 , Figure 4 FIG. 1 is a schematic diagram of the structure of the associated credibility evaluation device of the present invention. Figure 4 As shown, the association credibility evaluation device based on activity index of the present invention includes:

[0087] An information acquisition unit 11 selects and obtains detailed information on the spatiotemporal distribution of elements within a fixed time range;

[0088] An activity calculation unit 12, based on the spatiotemporal distribution detailed information of the elements, obtains a plurality of associated activities by a time slice association analysis method;

[0089] An activity evaluation unit 13, which evaluates a plurality of the associated activities to obtain an activity index;

[0090] The association credibility evaluation unit 14 evaluates the association credibility based on the activity index.

[0091] The detailed information of the spatiotemporal distribution of elements includes: element ID, associated element ID, total associated days, total associated times and collection time.

[0092] Furthermore, the activity calculation unit 12 includes:

[0093] The overall association activity calculation module 121 calculates the overall association activity of each association element according to the total association days and the total association times;

[0094] The time slice association activity calculation module 122 calculates the time slice association activity of each association element according to the time slice and the total number of associations.

[0095] Furthermore, the activity evaluation unit 13 includes:

[0096] A current activity obtaining module 131, which obtains a plurality of current activities based on the overall associated activity and the time slice associated activity;

[0097] A scoring interval setting module 132 is configured to set a plurality of scoring intervals corresponding to the plurality of current activity levels;

[0098] The activity index obtaining module 133 maps the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element.

[0099] Furthermore, the overall association activity is obtained by the following formula:

[0100]

[0101] The time slice associated activity includes: recent day associated activity, recent week associated activity, recent half month associated activity and recent month associated activity.

[0102] In summary, the present invention adopts the correlation between the credibility of the association relationship and the activity within a time range to design a method based on whether the activity in recent period of time is increasing to reflect whether the association relationship is credible, and the obtained association credibility evaluation result is more accurate.

[0103] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating association credibility based on activity index, characterized in that: Used to resolve entity ID associations, including: Step S1: Select and obtain detailed information on the spatiotemporal distribution of elements within a fixed time range; Step S2: obtaining multiple correlation activities through a time slice correlation analysis method based on the detailed information of the spatiotemporal distribution of the elements; Step S3: Evaluate the multiple associated activities to obtain an activity index; Step S4: Evaluate the association credibility based on the activity index; The detailed information of the temporal and spatial distribution of the elements includes: element ID, associated element ID, total associated days, total associated times and collection time; Wherein, the step S2 includes: Step S21: calculating the overall association activity of each association element according to the total association days and the total association times; Step S22: Calculate the time slice association activity of each association element according to the time slice and the total number of associations; Wherein, the step S3 includes: Step S31: obtaining a plurality of current activities based on the overall associated activity and the time slice associated activity; Step S32: setting a plurality of scoring intervals corresponding to a plurality of the current activity levels; Step S33: mapping the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element; The overall association activity is obtained by the following formula: ； The time slice associated activity includes: recent day associated activity, recent week associated activity, recent half month associated activity and recent month associated activity.

2. A device for evaluating the association credibility based on activity index, characterized in that: Used to resolve entity ID associations, including: An information acquisition unit selects and obtains detailed information on the spatiotemporal distribution of elements within a fixed time range; An activity calculation unit, which obtains a plurality of associated activities by a time slice association analysis method based on the detailed information of the spatiotemporal distribution of the elements; An activity evaluation unit, which evaluates a plurality of the associated activities to obtain an activity index; an association credibility evaluation unit, which evaluates association credibility based on the activity index; The detailed information of the temporal and spatial distribution of the elements includes: element ID, associated element ID, total associated days, total associated times and collection time; Wherein, the activity calculation unit includes: The overall association activity calculation module calculates the overall association activity of each association element according to the total number of association days and the total number of association times; A time slice association activity calculation module is used to calculate the time slice association activity of each association element according to the time slice and the total number of associations; Wherein, the activity evaluation unit includes: A current activity obtaining module, which obtains a plurality of current activities based on the overall associated activity and the time slice associated activity; A scoring interval setting module, which sets a plurality of scoring intervals corresponding to a plurality of current activity levels; An activity index obtaining module maps the multiple current activities to the multiple scoring intervals to obtain the activity index of the associated element; The overall association activity is obtained by the following formula: ； The time slice associated activity includes: recent day associated activity, recent week associated activity, recent half month associated activity and recent month associated activity.

Images