A method and system for monitoring specific personnel based on knowledge graph

Abstract

The present invention discloses a method and system for monitoring specific personnel based on a knowledge graph, wherein the method includes, during the real-time monitoring process of the specific personnel, collecting the location data of all the collection moments in each cycle, constructing a knowledge graph for the specific personnel, obtaining each retention period and the corresponding retention position in each cycle, calculating the recent unconventional access coefficient of each retention position in the target cycle, determining the unconventional retention period of the target cycle, obtaining the unconventional retention distribution difference index and the unconventional retention trend progressive index, obtaining the monitoring behavior violation index, and obtaining the monitoring result of the specific personnel based on the knowledge graph and the monitoring behavior violation index. The embodiment of the present invention provides a method and system for monitoring specific personnel based on a knowledge graph, analyzes the behavior pattern of the monitored specific personnel, and improves the real-time monitoring of the specific personnel while ensuring the accuracy of the monitoring of the specific personnel.

Description

Technical Field

[0001] The present invention relates to the technical field of personnel monitoring, and in particular to a method and system for monitoring specific personnel based on a knowledge graph. Background Art

[0002] In public places or sensitive areas, monitoring specific personnel can timely detect and prevent potential threats. With the advancement of technology and the increase in social security needs, monitoring specific personnel has become an important means of maintaining public security and corporate management. Specific personnel can be security guards, cleaners, street vendors and construction workers in public areas, as well as marked personnel designated by users, etc.

[0003] By using monitoring technology to deeply analyze and predict the behavior of specific people in real time, managers can intervene and deal with problems at the early stage to prevent them from getting worse. However, existing monitoring technology cannot simultaneously take into account the accuracy and real-time nature of monitoring specific people. Summary of the invention

[0004] The present invention provides a method and system for monitoring specific personnel based on a knowledge graph, so as to solve the technical problem that the existing monitoring technology is difficult to simultaneously take into account the accuracy and real-time monitoring of specific personnel, and improve the real-time monitoring of specific personnel.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method and system for monitoring specific personnel based on a knowledge graph.

[0006] In the process of real-time monitoring of a specific person, the location data of all collection moments in each cycle are collected, and a knowledge graph of the specific person is constructed based on all the location data;

[0007] Based on the difference and continuity of the position data changes at adjacent collection moments in each cycle, each retention period and corresponding retention position in each cycle are obtained;

[0008] Compare the number of occurrences of each of the retention positions in the selected target period with the number of occurrences of each of the retention positions in all the periods, and calculate the recent irregular access coefficient of each of the retention positions in the target period based on the comparison result; obtain the irregular retention period of the target period according to the length of each of the retention periods in the target period and the recent irregular access coefficient of the corresponding retention position;

[0009] Compare the unconventional detention period in the last cycle with the unconventional detention periods in all the cycles, and obtain an unconventional detention distribution difference index based on the distribution difference in the comparison result; obtain an unconventional detention trend asymptotic index based on the gradual change of the length of the unconventional detention period in all the cycles and the gradual change of the number of occurrences;

[0010] Obtaining a monitoring behavior violation index according to the unconventional detention distribution difference index and the unconventional detention trend progressive index, wherein the monitoring behavior violation index is configured to reflect the degree of abnormal behavior of a specific person through a linear relationship between the unconventional detention distribution difference index and the unconventional detention trend progressive index;

[0011] The monitoring result of the specific person is obtained according to the monitoring behavior violation index and the knowledge graph.

[0012] As one of the preferred solutions, the method of obtaining each retention period and the corresponding retention position in each cycle based on the difference and continuity of the position data changes at adjacent collection moments in each cycle includes:

[0013] Obtaining the personnel detention time based on the difference in the change amplitude of the position data of the adjacent collection moments in each cycle;

[0014] Merging the consecutive detention moments of the personnel in each cycle to obtain the detention time period for the personnel in each cycle, and screening all the detention time periods for the personnel in each cycle to obtain the detention time period for each cycle;

[0015] The center position of the position data corresponding to the detention moments of all the personnel in each detention period is used as the detention position of each detention period.

[0016] As one of the preferred solutions, the method of obtaining the personnel detention time based on the difference in the change amplitude of the position data of the adjacent collection moments in each cycle includes:

[0017] The distance between each collection moment and the position data of the corresponding adjacent next collection moment is recorded as the personnel position movement distance at each collection moment, the average value of all the personnel position movement distances in each cycle is recorded as the personnel movement judgment factor of each cycle, and the collection moment in each cycle where the personnel position movement distance is less than the personnel movement judgment factor is recorded as the personnel retention moment.

[0018] As one of the preferred solutions, the calculation of the recent unconventional access coefficient of each of the retention locations within the target period based on the comparison result includes:

[0019] The total number of occurrences of each of the retention positions in all the periods is counted, the number of occurrences of each of the retention positions in the target period is recorded as the recent number of occurrences of each of the retention positions in the target period, and the ratio of the recent number of occurrences of each of the retention positions in the target period to the total number of occurrences is recorded as the recent irregular access coefficient of each of the retention positions in the target period.

[0020] As one of the preferred solutions, the obtaining of the unconventional retention period of the target cycle according to the length of each retention period in the target cycle and the recent unconventional access coefficient of the corresponding retention position includes:

[0021] Recording the recent irregular access coefficient of the retention position corresponding to each retention period in the target period as the retention irregular coefficient of each retention period in the target period;

[0022] According to the linear relationship between the length of each detention period in the target cycle and the detention irregularity coefficient, the detention abnormality index of each detention period in the target cycle is calculated, and each detention abnormality index is processed to obtain the irregular detention period of the target cycle.

[0023] As one of the preferred solutions, the method of obtaining the unconventional retention distribution difference index based on the distribution difference in the comparison results includes:

[0024] For each of the collection moments, when each of the collection moments is within any of the irregular retention periods, the value of the retention anomaly factor of each of the collection moments is set to 1; otherwise, the value of the retention anomaly factor of each of the collection moments is set to 0;

[0025] Arrange the retention anomaly factors of all the acquisition moments in each of the cycles in ascending order according to the time sequence of acquisition, as the retention anomaly sequence of each of the cycles;

[0026] The average value of the distance between the last retention anomaly sequence of the cycle and the retention anomaly sequences of all the cycles is recorded as the irregular retention distribution difference index.

[0027] As one of the preferred solutions, the step of obtaining the progressive index of the unconventional retention trend based on the gradual change of the length of the unconventional retention period and the gradual change of the number of occurrences in all the cycles includes:

[0028] Obtaining an unconventional detention duration sequence and an unconventional detention frequency sequence based on the length and occurrence frequency of the unconventional detention time periods in all the cycles;

[0029] Performing a first-order difference on the unconventional detention time sequence to obtain an unconventional detention time difference sequence, and recording the absolute value of the sum of all elements in the unconventional detention time difference sequence as an unconventional detention time trend factor; performing a first-order difference on the unconventional detention frequency sequence to obtain an unconventional detention frequency difference sequence, and recording the absolute value of the sum of all elements in the unconventional detention frequency difference sequence as an unconventional detention frequency trend factor;

[0030] According to the linear relationship between the unconventional detention duration trend factor and the unconventional detention frequency trend factor, the unconventional detention trend asymptotic index is calculated.

[0031] As one of the preferred solutions, the obtaining of the unconventional detention duration sequence and the unconventional detention frequency sequence based on the length and the number of occurrences of the unconventional detention time periods in all the cycles includes:

[0032] The sum of the lengths of all the unconventional detention periods in each cycle is taken as the total unconventional detention duration of each cycle, and all the unconventional detention total durations are arranged in ascending order according to the time sequence of acquisition to obtain an unconventional detention duration sequence;

[0033] The number of occurrences of the unconventional detention period in each cycle is taken as the unconventional detention frequency of each cycle, and all the unconventional detention frequencies are arranged in ascending order according to the time sequence of acquisition to obtain an unconventional detention frequency sequence.

[0034] As one of the preferred solutions, the monitoring result of a specific person is obtained according to the monitoring behavior violation index and the knowledge graph, including:

[0035] The monitoring behavior violation index is processed to obtain the number of autoregressive terms of the autoregressive moving average model, and the autoregressive moving average model is used in combination with the knowledge graph to predict the trajectory of a specific person to obtain the monitoring result of the specific person.

[0036] Yet another embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, a method for monitoring a specific person based on a knowledge graph as described above is implemented.

[0037] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0038] (1) In the process of real-time monitoring of specific personnel, the location data of all collection moments in each cycle are collected, and a knowledge graph of the specific personnel is constructed based on all the location data. The difference and continuity of the location data changes of the specific personnel at adjacent collection moments are first analyzed to obtain each retention period and the corresponding retention position in each cycle. Subsequently, only each retention period and the corresponding retention position are analyzed, which improves the real-time monitoring of specific personnel.

[0039] (2) Based on the difference between the number of occurrences of each retention location in the selected target period and the number of occurrences of each retention location in all the said periods, the recent irregular access coefficient of each retention location in the target period is calculated to measure the possibility of abnormal behavior of a specific person in each retention location in the target period. The possibility of abnormal behavior of a specific person in each retention location in the target period is evaluated by comparing the behavior pattern of the specific person in the target period with that in other periods, rather than just based on the number of occurrences of each retention location in the target period, thereby avoiding misjudging the regular behavior of a specific person as abnormal behavior; and then combining the length of each retention period in the target period to obtain the irregular retention period of the target period, thereby measuring the possibility of abnormal behavior of a specific person in each retention period in the target period, thereby improving the reliability of determining the irregular retention period;

[0040] (3) Then, the differences and gradual changes in the abnormal behavior patterns of specific personnel in the recent period and in the past are analyzed. The distribution difference index of the abnormal detention period in the last cycle and the abnormal detention period in all cycles is obtained. The gradual change of the length of the abnormal detention period in all cycles and the gradual change of the number of occurrences are used to obtain the abnormal detention trend progressive index. The monitoring behavior violation index is obtained based on the abnormal detention distribution difference index and the abnormal detention trend progressive index to measure the abnormal degree of the recent behavior pattern of specific personnel;

[0041] (4) The monitoring behavior violation index is processed to obtain the number of autoregressive terms of the autoregressive moving average model. The autoregressive moving average model is used in combination with the knowledge graph to predict the trajectory of a specific person and obtain the monitoring result of the specific person. The number of autoregressive terms of the autoregressive moving average model is determined based on the abnormality of the recent behavior pattern of the specific person. This ensures the accuracy of monitoring the specific person and further improves the real-time monitoring of the specific person. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of a process of monitoring a specific person based on a knowledge graph in one embodiment of the present invention;

[0043] Figure 2is a schematic diagram of a knowledge graph of a specific person in one embodiment of the present invention;

[0044] Figure 3 It is a schematic diagram of a period of time during which a person is detained for screening in one embodiment of the present invention;

[0045] Figure 4 It is a schematic diagram of obtaining a monitoring behavior violation index in one embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. 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.

[0047] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0048] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used in this article are only for illustrative purposes, 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 the present invention. The term "and / or" used in this article includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0049] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0050] An embodiment of the present invention provides a method for monitoring a specific person based on a knowledge graph. For details, see Figure 1 , Figure 1 Shown is a schematic diagram of the process of monitoring specific personnel based on a knowledge graph in one embodiment of the present invention.

[0051] Step S1: In the process of real-time monitoring of a specific person, the location data of all collection moments in each cycle are collected, and a knowledge graph of the specific person is constructed based on all the location data.

[0052] Location data can provide real-time location information, making the tracking process almost instantaneous and helping to respond quickly to various situations. Therefore, before tracking a specific person, it is necessary to install a positioning device on the specific person's device. During the real-time monitoring of the specific person, the location data of the specific person is collected every t minutes.

[0053] It should be noted that the length of each cycle and the collection interval t are both preset values. The positioning devices include GPS positioning devices, Beidou satellite positioning devices and Bluetooth positioning devices. In this embodiment, the length of each cycle is 24 hours, and the collection interval t is 5 minutes. The Beidou satellite positioning device is selected to collect the location information of specific personnel. The length of each cycle, the value of the collection interval and the selection of the positioning device are other implementation methods that the implementer can choose at his own discretion, and this application does not impose any special restrictions on this.

[0054] It should be further explained that the knowledge graph is essentially a semantic network that reveals the relationships between entities. It describes knowledge resources and their carriers through visualization technology, and mines, analyzes, constructs, draws and displays knowledge and their interrelationships.

[0055] In order to deeply analyze and predict the behavior of a specific person and then identify the abnormal behavior pattern of the specific person, a knowledge graph of the specific person is constructed based on all location data. In one embodiment of the present invention, the schematic diagram of the knowledge graph of the specific person is as follows: Figure 2 shown.

[0056] Step S2: Based on the difference and continuity of the position data changes at adjacent collection moments in each cycle, each retention period and the corresponding retention position in each cycle are obtained.

[0057] Generally, people's activities in a specific area follow certain rules and patterns. If they stay in a certain area for a long time, it may indicate that the area is of special importance to the monitored person. Therefore, when monitoring a specific person, special attention should be paid to analyzing the area where the specific person stays for a long time.

[0058] Specifically, the distance between the position data of each collection moment and the next adjacent collection moment is recorded as the personnel position movement distance of each collection moment, and the personnel position movement distance represents the distance moved by a specific person in a certain time period. The average value of the position movement distances of all personnel in each cycle is recorded as the personnel movement judgment factor of each cycle, and the personnel movement judgment factor represents the average distance moved by a specific person in each time period. The collection moment in each cycle when the position movement distance of each person is less than the personnel movement judgment factor is recorded as the personnel detention moment.

[0059] It should be noted that when the distance moved by a specific person in a certain time period is less than the average distance moved in each time period, it means that the time he stays in the time period is greater than the average time, and the attention paid to the behavior of the specific person in the time period should be increased.

[0060] Considering that people stay in certain areas for too long, which may span multiple time periods, in order to analyze the stay of specific people in the same area as a whole, the consecutive stay moments of people in each cycle are merged to obtain the stay and screening period of people in each cycle.

[0061] It should be noted that the starting time of each personnel detention waiting for screening period is the first personnel detention time that occurs continuously each time, and the ending time of each personnel detention waiting for screening period is the last personnel detention time that occurs continuously each time.

[0062] In one embodiment of the present invention, a schematic diagram of the period of time during which personnel are detained for screening is shown in FIG. Figure 3 shown. Figure 3 In the figure, each scale represents a collection time, and the ones marked with dots are the personnel detention time.

[0063] Further, all the personnel detention waiting time periods in each cycle are screened to obtain the detention time periods of each cycle. The center position of the position data corresponding to the detention moments of all personnel in each detention time period is used as the detention position of each detention time period.

[0064] As an embodiment of the present application, the method of screening all the time periods in each cycle where personnel are stranded and waiting for screening includes: eliminating the time periods in which personnel are stranded and waiting for screening, which contain a number of personnel stranded moments that is less than a preset stranded threshold.

[0065] It should be noted that the preset retention threshold is a manually preset value. In this application, the preset retention threshold is 3. As for the value of the preset retention threshold, as other implementation methods, the implementer can choose it at will, and this application does not impose any special restrictions on this.

[0066] Step S3: Calculate the recent irregular access coefficient of each residence position in the target period based on the difference between the number of occurrences of each residence position in the selected target period and the number of occurrences of each residence position in all the said periods, and obtain the irregular residence period of the target period according to the length of each residence period in the target period and the recent irregular access coefficient of the corresponding residence position.

[0067] Considering that a place where a person stays for a long time may be an area that the person frequently visits, such as a workplace, a store, etc., or it may be an area that the person occasionally visits, when monitoring a specific person, less attention should be paid to the regular areas that the person frequently visits, and more attention should be paid to the areas that the person occasionally visits recently but stays for a long time.

[0068] Specifically, the total number of occurrences of each retention position in all cycles is counted, the number of occurrences of each retention position in the selected target cycle is recorded as the recent number of occurrences of each retention position in the target cycle, and the ratio of the recent number of occurrences of each retention position in the target cycle to the total number of occurrences is recorded as the recent irregular access coefficient of each retention position in the target cycle.

[0069] It should be noted that the more recent appearances of each retention location within the target period, but the fewer total appearances, the more likely it is that the retention location is an area where the monitored specific person has occasionally visited recently but stayed for a long time, the more likely the specific person is to exhibit abnormal behavior in the area, and the larger the recent irregular visit coefficient value.

[0070] Further, the recent irregular access coefficient of the retention position corresponding to each retention period in the target cycle is recorded as the retention irregular coefficient of each retention period in the target cycle. According to the linear relationship between the length of each retention period in the target cycle and the retention irregular coefficient, the retention abnormality index of each retention period in the target cycle is calculated, and each retention abnormality index is processed to obtain the irregular retention period of the target cycle.

[0071] It should be noted that the linear relationship means that the change of the dependent variable can be described by the linear function of the independent variable. The longer the length of a certain detention period is and the larger the detention irregularity coefficient is, the more likely it is that the place visited by the specific person during the detention period is an area that he or she has visited occasionally recently and stayed for a long time. The more likely it is that the specific person will exhibit abnormal behavior during the detention period, and the larger the detention anomaly index value of the detention period is.

[0072] As an embodiment of the present application, the length of each detention period plus the detention irregularity coefficient is used as the detention abnormality index of each detention period, each detention abnormality index is normalized to obtain each detention abnormality normalized index, and the detention period whose detention abnormality normalized index is greater than the preset first threshold is recorded as an irregular detention period of the target period.

[0073] It should be noted that the preset first threshold is a manually preset value. In this embodiment, the preset first threshold is 0.5. As for the value of the preset first threshold, as other implementation methods, the implementer can choose it at will, and this application does not impose any special restrictions on this.

[0074] Step S4: Obtain an unconventional detention distribution difference index based on the distribution difference between the unconventional detention periods in the last cycle and the unconventional detention periods in all cycles, obtain an unconventional detention trend progression index based on the gradual variability of the length of the unconventional detention periods in all cycles and the gradual variability of the number of occurrences, and obtain a monitoring behavior violation index based on the unconventional detention distribution difference index and the unconventional detention trend progression index.

[0075] When tracking a specific person, since different people have different behavior patterns, in order to evaluate the degree of abnormality of the specific person's behavior, it is necessary to first analyze the differences between their recent and past abnormal behavior patterns.

[0076] Specifically, for each collection moment, when each collection moment is within any unconventional retention period, the value of the retention anomaly factor of each collection moment is set to 1, otherwise, the value of the retention anomaly factor of each collection moment is set to 0. The retention anomaly factors of all collection moments in each cycle are arranged in ascending order according to the time sequence of acquisition, as the retention anomaly sequence of each cycle. The average value of the distance between the retention anomaly sequence of the last cycle and the retention anomaly sequence of all cycles is recorded as the unconventional retention distribution difference index.

[0077] It should be noted that the retention anomaly sequence of each cycle represents the abnormal behavior pattern of a specific person in each cycle. The greater the distance between the retention anomaly sequence of the last cycle and the retention anomaly sequence of all cycles, the more different the abnormal behavior pattern of the last cycle is from that of the remaining cycles. The greater the value of the unconventional retention distribution difference index, the more likely it is that the specific monitored person has exhibited abnormal behavior.

[0078] Furthermore, the gradual changes in abnormal behavior patterns of specific monitored persons are analyzed.

[0079] Specifically, the sum of the lengths of all unconventional detention periods in each cycle is taken as the total unconventional detention time of each cycle, all unconventional detention time totals are arranged in ascending order according to the chronological order of acquisition to obtain an unconventional detention time sequence, a first-order difference is made on the unconventional detention time sequence to obtain an unconventional detention time difference sequence, and the absolute value of the sum of all elements in the unconventional detention time difference sequence is recorded as the unconventional detention time trend factor.

[0080] It should be noted that when the total duration of unconventional detention in adjacent periods shows an increasing or decreasing trend, the absolute value of the sum of all elements in the unconventional detention duration difference sequence is larger, and the value of the unconventional detention duration trend factor is larger; and when the total duration of unconventional detention in adjacent periods sometimes increases and sometimes decreases, the absolute value of the sum of all elements in the unconventional detention duration difference sequence is smaller, and the value of the unconventional detention duration trend factor is smaller.

[0081] The number of times an unconventional detention period occurs in each cycle is taken as the unconventional detention frequency of each cycle. All unconventional detention frequencies are arranged in ascending order according to the chronological order of acquisition to obtain an unconventional detention frequency sequence. The unconventional detention frequency sequence is subjected to a first-order difference to obtain an unconventional detention frequency difference sequence. The absolute value of the sum of all elements in the unconventional detention frequency difference sequence is recorded as the unconventional detention frequency trend factor.

[0082] It should be noted that when the unconventional detention frequency of adjacent periods shows an increasing or decreasing trend, the absolute value of the sum of all elements in the unconventional detention frequency difference sequence is larger, and the unconventional detention frequency trend factor value is larger; when the unconventional detention frequency of adjacent periods sometimes increases and sometimes decreases, the absolute value of the sum of all elements in the unconventional detention frequency difference sequence is smaller, and the unconventional detention frequency trend factor value is smaller.

[0083] Furthermore, based on the linear relationship between the unconventional detention duration trend factor and the unconventional detention frequency trend factor, the unconventional detention trend asymptotic index is calculated, and based on the linear relationship between the unconventional detention distribution difference index and the unconventional detention trend asymptotic index, the monitoring behavior violation index is calculated.

[0084] It should be noted that the larger the unconventional detention distribution difference index is, the greater the difference between the recent and past abnormal behavior patterns of the monitored specific personnel is, and the more likely the monitored specific personnel are to have abnormal behavior; at the same time, the larger the unconventional detention duration trend factor and the unconventional detention frequency trend factor are, the stronger the gradual change of the abnormal behavior pattern of the monitored specific personnel is, the larger the value of the unconventional detention trend progressive index is, the more frequent the abnormal behavior of the monitored specific personnel is, and the larger the value of the monitoring behavior violation index is.

[0085] As an embodiment of the present application, an unconventional detention duration trend factor is added to an unconventional detention frequency trend factor as an unconventional detention trend progressive index, the product of a preset first coefficient and an unconventional detention distribution difference index is used as a first judgment factor, the product of a preset second coefficient and an unconventional detention trend progressive index is used as a second judgment factor, and the first judgment factor plus the second judgment factor is used as a monitoring behavior violation index.

[0086] It should be noted that the preset first coefficient and the preset second coefficient are both artificially preset values. In this embodiment, the preset first coefficient is 0.8, and the preset second coefficient is 0.2. For the values ​​of the preset first coefficient and the preset second coefficient, as other implementation methods, the implementer can choose them at will, and this application does not impose any special restrictions on this.

[0087] In one embodiment of the present invention, the schematic diagram of obtaining the monitoring behavior violation index is as follows: Figure 4 shown.

[0088] Step S5: Based on the knowledge graph and the monitoring behavior violation index, the monitoring results of the specific person are obtained.

[0089] It should be noted that the autoregressive moving average model predicts future values ​​by combining the autoregressive characteristics, differential stationarity and moving average characteristics of historical data. By using the autoregressive moving average model to conduct in-depth analysis and prediction of the behavior of specific personnel based on the knowledge graph, it is possible to identify the abnormal behavior patterns of specific personnel and then monitor the specific personnel. However, the traditional autoregressive moving average model has a uniform setting of the number of autoregressive items, which does not take into account the inherent behavior patterns of specific personnel, making it difficult to simultaneously take into account the accuracy and real-time monitoring of specific personnel.

[0090] In order to ensure the accuracy of monitoring of specific personnel and improve the real-time monitoring of specific personnel, the monitoring behavior violation index is processed to obtain the autoregressive term number of the autoregressive moving average model.

[0091] It should be noted that when the monitoring behavior violation index is larger, it means that the specific person being monitored is more likely to have abnormal behavior, and the abnormal behavior occurs more frequently. At this time, a larger number of autoregressive items should be set to enhance the accuracy of tracking the specific person; when the monitoring behavior violation index is smaller, it means that the behavior pattern of the specific person being monitored is more regular, and the accurate prediction of the trajectory of the specific person can be achieved based on less historical data. At this time, a smaller number of autoregressive items should be set to improve the real-time tracking of the specific person.

[0092] As an embodiment of the present application, the monitoring behavior violation index is normalized, and the integer value of the product of the monitoring behavior violation index and the preset third coefficient is used as the number of autoregressive terms of the autoregressive moving average model.

[0093] It should be noted that the preset third coefficient is a manually preset value. In this embodiment, the preset third coefficient is 10. As for the value of the preset third coefficient, as other implementation methods, the implementer can select it at will, and this application does not impose any special restrictions on this.

[0094] Furthermore, the autoregressive moving average model is combined with the knowledge graph to predict the trajectory of specific personnel and obtain the monitoring results of the specific personnel.

[0095] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0096] Accordingly, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to perform the steps in a method for monitoring a specific person based on a knowledge graph as in the above embodiment, for example Figure 1 Steps S1 to S5 described in .

[0097] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for monitoring specific personnel based on knowledge graph, It is characterized in that include: In the process of real-time monitoring of a specific person, the location data of all collection moments in each cycle are collected, and a knowledge graph of the specific person is constructed based on all the location data; Based on the difference and continuity of the position data changes at adjacent collection moments in each cycle, each retention period and corresponding retention position in each cycle are obtained; Compare the number of occurrences of each of the retention positions in the selected target period with the number of occurrences of each of the retention positions in all the periods, and calculate the recent irregular access coefficient of each of the retention positions in the target period based on the comparison result; obtain the irregular retention period of the target period according to the length of each of the retention periods in the target period and the recent irregular access coefficient of the corresponding retention position; Compare the unconventional detention period in the last cycle with the unconventional detention periods in all the cycles, and obtain an unconventional detention distribution difference index based on the distribution difference in the comparison result; obtain an unconventional detention trend asymptotic index based on the gradual change of the length of the unconventional detention period in all the cycles and the gradual change of the number of occurrences; Obtaining a monitoring behavior violation index according to the unconventional detention distribution difference index and the unconventional detention trend progressive index, wherein the monitoring behavior violation index is configured to reflect the degree of abnormal behavior of a specific person through a linear relationship between the unconventional detention distribution difference index and the unconventional detention trend progressive index; The monitoring result of the specific person is obtained according to the monitoring behavior violation index and the knowledge graph.

2. According to the method for monitoring specific personnel based on knowledge graph according to claim 1, It is characterized in that The obtaining of each retention period and the corresponding retention position in each cycle based on the difference and continuity of the position data changes at adjacent collection moments in each cycle includes: Obtaining the personnel detention time based on the difference in the change amplitude of the position data of the adjacent collection moments in each cycle; Merging the consecutive detention moments of the personnel in each cycle to obtain the detention time period for the personnel in each cycle, and screening all the detention time periods for the personnel in each cycle to obtain the detention time period for each cycle; The center position of the position data corresponding to the detention moments of all the personnel in each detention period is used as the detention position of each detention period.

3. According to the method for monitoring specific personnel based on knowledge graph according to claim 2, It is characterized in that The obtaining of the personnel detention time based on the difference in the change amplitude of the position data of the adjacent collection moments in each cycle includes: The distance between each collection moment and the position data of the corresponding adjacent next collection moment is recorded as the personnel position movement distance at each collection moment, the average value of all the personnel position movement distances in each cycle is recorded as the personnel movement judgment factor of each cycle, and the collection moment in each cycle where the personnel position movement distance is less than the personnel movement judgment factor is recorded as the personnel retention moment.

4. According to the method for monitoring specific personnel based on knowledge graph according to claim 1, It is characterized in that The calculating, based on the comparison result, the recent irregular access coefficient of each of the retention locations within the target period includes: The total number of occurrences of each of the retention positions in all the periods is counted, the number of occurrences of each of the retention positions in the target period is recorded as the recent number of occurrences of each of the retention positions in the target period, and the ratio of the recent number of occurrences of each of the retention positions in the target period to the total number of occurrences is recorded as the recent irregular access coefficient of each of the retention positions in the target period.

5. The method for monitoring specific personnel based on knowledge graph according to claim 1, It is characterized in that The step of obtaining the unconventional retention period of the target cycle according to the length of each retention period in the target cycle and the recent unconventional access coefficient of the corresponding retention position comprises: Recording the recent irregular access coefficient of the retention position corresponding to each retention period in the target period as the retention irregular coefficient of each retention period in the target period; According to the linear relationship between the length of each detention period in the target cycle and the detention irregularity coefficient, the detention abnormality index of each detention period in the target cycle is calculated, and each detention abnormality index is processed to obtain the irregular detention period of the target cycle.

6. The method for monitoring specific personnel based on knowledge graph according to claim 1, It is characterized in that The obtaining of the unconventional retention distribution difference index based on the distribution difference in the comparison results includes: For each of the collection moments, when each of the collection moments is within any of the irregular retention periods, the value of the retention anomaly factor of each of the collection moments is set to 1; otherwise, the value of the retention anomaly factor of each of the collection moments is set to 0; Arrange the retention anomaly factors of all the acquisition moments in each of the cycles in ascending order according to the time sequence of acquisition, as the retention anomaly sequence of each of the cycles; The average value of the distance between the last retention anomaly sequence of the cycle and the retention anomaly sequences of all the cycles is recorded as the irregular retention distribution difference index.

7. The method for monitoring specific personnel based on knowledge graph according to claim 1, It is characterized in that The step of obtaining the progressive index of the unconventional retention trend based on the gradual change of the length of the unconventional retention period and the gradual change of the number of occurrences in all the cycles comprises: Obtaining an unconventional detention duration sequence and an unconventional detention frequency sequence based on the length and occurrence frequency of the unconventional detention time periods in all the cycles; Performing a first-order difference on the unconventional detention time sequence to obtain an unconventional detention time difference sequence, and recording the absolute value of the sum of all elements in the unconventional detention time difference sequence as an unconventional detention time trend factor; performing a first-order difference on the unconventional detention frequency sequence to obtain an unconventional detention frequency difference sequence, and recording the absolute value of the sum of all elements in the unconventional detention frequency difference sequence as an unconventional detention frequency trend factor; According to the linear relationship between the unconventional detention duration trend factor and the unconventional detention frequency trend factor, the unconventional detention trend asymptotic index is calculated.

8. The method for monitoring specific personnel based on knowledge graph according to claim 7, It is characterized in that The obtaining of the unconventional detention duration sequence and the unconventional detention frequency sequence based on the length and the number of occurrences of the unconventional detention time periods in all the cycles includes: The sum of the lengths of all the unconventional detention periods in each cycle is taken as the total unconventional detention duration of each cycle, and all the unconventional detention total durations are arranged in ascending order according to the time sequence of acquisition to obtain an unconventional detention duration sequence; The number of occurrences of the unconventional detention period in each cycle is taken as the unconventional detention frequency of each cycle, and all the unconventional detention frequencies are arranged in ascending order according to the time sequence of acquisition to obtain an unconventional detention frequency sequence.

9. The method for monitoring a specific person based on a knowledge graph according to claim 1, It is characterized in that The monitoring result of a specific person is obtained according to the monitoring behavior violation index and the knowledge graph, including: The monitoring behavior violation index is processed to obtain the number of autoregressive terms of the autoregressive moving average model, and the autoregressive moving average model is used in combination with the knowledge graph to predict the trajectory of a specific person to obtain the monitoring result of the specific person.

10. A system for monitoring specific personnel based on knowledge graph, It is characterized in that The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for monitoring a specific person based on a knowledge graph as described in any one of claims 1 to 9 is implemented.

Images