Revenue sensing alarm method and device based on atlas

Abstract

The invention provides an income perception alarm method and device based on a graph, and relates to the field of artificial intelligence such as knowledge graphs, big data processing and intelligent alarm. The method comprises the steps that in response to determining that a generation triggering condition is met, a service relation graph is generated for advertisement service data obtained from different data sources in the latest W days, and W is a positive integer; screening out a target entity serving as a monitoring object from the advertising service entities in the service relation graph; and aiming at any target entity, monitoring the income data of the target entity acquired in real time, and performing alarm operation when determining that an alarm triggering condition is met according to the real-time alarm strategy corresponding to the target entity.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a graph-based income perception alarm method and device in the fields of knowledge graphs, big data processing, and intelligent alarms. Background Art

[0002] With the continuous upgrading of the advertising business, the business architecture has become increasingly complex, and the knowledge system has become increasingly large, covering diverse advertising styles, rich traffic ecology and complex advertising product scenarios. Summary of the invention

[0003] The present disclosure provides a graph-based income-aware alarm method and device.

[0004] A graph-based income-aware alarm method, comprising:

[0005] In response to determining that the generation trigger condition is met, generating a business relationship graph for the advertising business data acquired from different data sources in the last W days, where W is a positive integer;

[0006] Filtering target entities as monitoring objects from the advertising business entities in the business relationship graph;

[0007] For any target entity, the income data of the target entity acquired in real time is monitored respectively, and an alarm operation is performed when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0008] An income perception alarm device based on a graph comprises: a graph construction module, an object determination module and a monitoring alarm module;

[0009] The graph construction module is used to generate a business relationship graph for the advertising business data acquired from different data sources in the last W days in response to determining that the generation trigger condition is met, where W is a positive integer;

[0010] The object determination module is used to filter out target entities as monitoring objects from the advertising business entities in the business relationship graph;

[0011] The monitoring alarm module is used to monitor the income data of any target entity obtained in real time, and perform an alarm operation when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0012] An electronic device, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0016] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0017] A computer program product comprises a computer program / instruction, wherein the computer program / instruction implements the method described above when executed by a processor.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0020] Figure 1 It is a flow chart of an embodiment of the graph-based income perception alarm method of the present disclosure;

[0021] Figure 2 A schematic diagram of the implementation process of the graph-based income perception alarm method disclosed in the present invention;

[0022] Figure 3 It is a schematic diagram of the composition structure of an embodiment 300 of the income sensing alarm device based on the graph described in the present disclosure;

[0023] Figure 4 A schematic block diagram of an electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0024] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0025] In addition, it should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0026] Figure 1 Flow chart of an embodiment of the graph-based income perception alarm method of the present disclosure. Figure 1 As shown, the following specific implementation methods are included.

[0027] In step 101, in response to determining that a generation trigger condition is met, a business relationship graph is generated for advertising business data acquired from different data sources in the last W days, where W is a positive integer.

[0028] In step 102, target entities to be monitored are screened out from the advertising business entities in the business relationship graph.

[0029] In step 103, for any target entity, the income data of the target entity acquired in real time is monitored respectively, and an alarm operation is performed when it is determined that the alarm triggering condition is met according to the real-time alarm strategy corresponding to the target entity.

[0030] By adopting the scheme described in the above method embodiment, a business relationship map can be constructed according to the advertising business data obtained in the last W days, and then the target entity as the monitoring object can be screened out from the advertising business entities in the business relationship map, and then the target entity can be monitored according to the real-time income data of the target entity and the corresponding real-time alarm strategy to see whether it meets the alarm triggering conditions, and an alarm operation can be performed when the alarm triggering conditions are met, that is, for online advertisements, its income situation can be monitored, and an alarm can be issued in time when abnormal problems are found, so that the abnormal problems can be handled in time, and the business relationship map can be continuously updated, and accordingly, the determined target entity can also be updated, that is, the monitoring object can be dynamically determined, so that it can quickly adapt to the dynamic changes of the business or system, thereby improving the accuracy, effectiveness and timeliness of monitoring.

[0031] Meeting the generation trigger condition may refer to reaching the start time of each day, such as midnight. In addition, the specific value of W may be determined according to actual needs, such as 1 day. In other words, the business relationship graph may be reconstructed each day, that is, the business relationship graph may be updated each day, and the business relationship graph may be constructed with the help of advertising business data obtained from different data sources in the most recent day (previous day / yesterday).

[0032] The specific data sources included in different data sources may be determined according to actual needs, such as billing flow, exposure log, service end database, notice log, etc. The acquired advertising service data may include various information related to the delivered advertisements.

[0033] There is no restriction on how to construct the business relationship map. For example, for the acquired advertising business data, a predetermined type of advertising business entity and attribute information of the advertising business entity can be extracted therefrom, and the attribute information may include revenue attributes (various revenue indicators) and business attributes (such as application type, access type), etc., and then a business relationship map can be constructed based on the extracted advertising business entity and the attribute information.

[0034] In practical applications, the business relationship graph may be a knowledge graph in the form of a high-performance NOSQL graph database (Neo4j), where NOSQL refers to a non-relational database (Not Only Structured Query Language). In addition, the business relationship graph may include nodes and edges, each node may correspond to an advertising business entity, and edges between nodes may be constructed based on the attribute information.

[0035] By adopting the above processing method, the advertising business data obtained from different data sources can be automatically sorted out to obtain the required business relationship map by extracting and organizing information, thus laying a good foundation for subsequent processing. Moreover, the business relationship map can be updated each time the generation trigger conditions are met, thereby improving the accuracy of the business relationship map, and further improving the accuracy of subsequent processing based on the business relationship map.

[0036] According to the business relationship map, the target entities to be monitored can be screened out from the advertising business entities.

[0037] In some embodiments of the present disclosure, the advertising business entities may include: advertising business entities of M entity types, where M is a positive integer greater than 1. Accordingly, filtering out target entities as monitoring objects from the advertising business entities may include: generating entity lists corresponding to the M entity types respectively, the advertising business entities in the same entity list all belong to the corresponding entity type, and determining the advertising business entities in the entity list as target entities.

[0038] In some embodiments of the present disclosure, the M entity types of advertising business entities may include: media identification (appsid) entity, advertising slot identification (cmatch) entity, and advertiser identification (dspid) entity, etc.

[0039] For the above three entity types, corresponding entity lists can be constructed based on the business relationship graph. Taking the media identification entity as an example, the media identification entities in the business relationship graph can be screened to generate an entity list composed of the screened media identification entities. The generation method of other entity lists is the same. After obtaining the three entity lists, the advertising business entities in the three entity lists can be determined as the required target entities, and real-time monitoring and alarm can be performed for the target entities, thereby realizing the effective screening of advertising business entities in the business relationship graph, thereby improving the pertinence and processing efficiency of subsequent monitoring and alarms.

[0040] In some embodiments of the present disclosure, an entity list corresponding to M entity types can be generated in the following manner: the average income of each media identification entity in the business relationship graph in the last N days is obtained respectively, where N is a positive integer greater than 1, and each media identification entity is sorted in order of average income from high to low, and the media identification entities in the top L positions (TOP-L) after sorting are used to form a first entity list, where L is a positive integer greater than 1, and L is less than the number of media identification entities in the business relationship graph; the advertising slot identification entities that have generated income in the last P days are filtered out from each advertising slot identification entity in the business relationship graph, where P is a positive integer, and the second entity list is formed using the filtered advertising slot identification entities; the advertiser identification entities that have generated income in the last P days are filtered out from each advertiser identification entity in the business relationship graph, and the third entity list is formed using the filtered advertiser identification entities.

[0041] The specific values ​​of N, L and P can be determined according to actual needs. For example, the value of N can be 7, the value of L can be 100, and the value of P can be 1 or 7.

[0042] There are usually a large number of media identification entities. If real-time monitoring and alarm are performed for all of them, it will consume more resources and the real-time performance may not be guaranteed. Therefore, the media identification entities can be screened based on the average income. For example, the TOP-100 media identification entities with the most significant income contribution are selected to form the first entity list. Assuming that the value of N is 7, then for any media identification entity, the income of the media identification entity in the past 7 days (total income in one day) can be obtained separately, and then the average of the income of these 7 days can be calculated to obtain the average income of the media identification entity.

[0043] Taking into account that the number of ad slot identification entities and advertiser identification entities is relatively small compared to media identification entities, the screening conditions can be appropriately relaxed, that is, the ad slot identification entities that have generated revenue in the last P days can be screened out from the ad slot identification entities in the business relationship graph, and the screened ad slot identification entities can be used to form a second entity list. In addition, the advertiser identification entities that have generated revenue in the last P days can be screened out from the advertiser identification entities in the business relationship graph, and the screened advertiser identification entities can be used to form a third entity list.

[0044] For each target entity in the three entity lists, the following processing can be performed respectively: the income data of the target entity obtained in real time is monitored, and an alarm operation is performed when the alarm triggering condition is determined to be met according to the real-time alarm strategy corresponding to the target entity. Among them, the income data of the target entity can be obtained from a data source such as a billing flow.

[0045] In some embodiments of the present disclosure, for any target entity, the target level to which the target entity belongs can be first determined, the entity type corresponding to the target entity corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective real-time alarm strategies, and then the real-time alarm strategy corresponding to the target level can be determined as the real-time alarm strategy corresponding to the target entity, and an alarm operation can be performed when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0046] In addition, in some embodiments of the present disclosure, for any target entity, the method of determining the target level to which the target entity belongs may include: obtaining the average income of the target entity in the last Q days, where Q is a positive integer, determining the income range to which the average income belongs, and determining the level corresponding to the income range as the target level, where different levels correspond to their own income ranges, and there is no overlap between the income ranges corresponding to any two levels. The specific value of Q may be determined according to actual needs.

[0047] Assuming that the target entity a is a media identification entity, and assuming that the value of Q is 7, then the average income of the target entity a in the past 7 days can be obtained, such as 300,000. Assuming that the entity type of the media identification entity is divided into 4 levels in advance, namely: Level 1: 0-100,000, Level 2: 110,000-250,000, Level 3: 260,000-500,000, Level 4: 510,000-1 million (or above 500,000), and corresponding real-time alarm strategies are set for different levels. Since 300,000 is within the income range of 260,000-500,000, it can be determined that the target level corresponding to the target entity a is level 3, and then the real-time alarm strategy corresponding to level 3 can be determined as the real-time alarm strategy corresponding to the target entity a, and the income data of the target entity a obtained in real time can be monitored. Once it is determined that the alarm trigger condition is met according to the real-time alarm strategy corresponding to the target entity a, the alarm operation can be performed. For example, if it is determined that the abnormal problem defined in the real-time alarm strategy has occurred, it can be determined that the alarm trigger condition is met.

[0048] The above takes the entity type of media identification entity as an example. For other entity types, multiple levels can also be divided respectively. The specific division method can be determined according to actual needs, and corresponding real-time alarm strategies can be set for different levels.

[0049] For the target entities in the three entity lists, the above-mentioned monitoring and alarm methods can be used respectively. It can be seen that the scheme described in the present disclosure proposes a multi-dimensional real-time hierarchical alarm method. Taking into account the uneven distribution and significant differences of entity income in different dimensions, the scheme described in the present disclosure uses the insight of the business relationship map to hierarchize the income levels of media identification entities, advertising space identification entities, and advertiser identification entities. Through the hierarchical mechanism, differentiated configuration of alarm strategies is achieved, that is, each level of each dimension can adopt an alarm strategy suitable for its income characteristics, so that various abnormal problems can be discovered in time, such as income falling to zero or a sudden drop, etc., and false alarms can be avoided as much as possible, improving the accuracy of the alarm. In addition, the newly added income dimension can be dynamically perceived, and potential "income decline" problems can be discovered and dealt with in time to ensure the continuous and stable operation of the business.

[0050] In some embodiments of the present disclosure, the real-time alarm strategy may include: alarm rules and alarm thresholds. For any target entity, in response to determining that the alarm threshold corresponding to the target entity is individually set, the set alarm threshold can be used to replace the alarm threshold in the real-time alarm strategy corresponding to the target entity to obtain an updated real-time alarm strategy, and an alarm operation can be performed when it is determined that the alarm triggering conditions are met according to the updated real-time alarm strategy.

[0051] The alarm rules may include some specific alarm judgment conditions, etc., and the alarm threshold may be a comparison threshold corresponding to the alarm judgment condition. The specific content of the alarm rules and the specific value of the alarm threshold may be determined according to actual needs.

[0052] Taking the above-mentioned target entity a as an example, its corresponding target level is level 3. Assume that the alarm threshold in the real-time alarm strategy corresponding to level 3 is threshold A. In addition, assume that the target entity a is facing a decline in income, and this change is in line with business expectations, then the alarm threshold corresponding to the target entity a can be temporarily adjusted, such as adjusted to threshold B. This adjustment can take effect immediately, thereby reducing unnecessary alarm interference, that is, reducing the problem of a surge in alarm volume due to expected income fluctuations for one or some target entities, thereby effectively improving the overall performance of the alarm system.

[0053] In addition, the validity period of the adjustment can be set, and it can automatically restore to threshold A after the validity period expires. Or, if the validity period of the adjustment is not set, it can automatically restore to threshold A after a predetermined period of time, which is very flexible and convenient.

[0054] In some embodiments of the present disclosure, for any target entity, an alarm operation may be performed in response to determining that an alarm triggering condition is met according to a general real-time alarm policy, where the general real-time alarm policy is a general policy set for M entity types.

[0055] That is, in addition to the above-mentioned multi-dimensional real-time hierarchical alarm method, a general real-time alarm method can also be used to monitor and alarm the target entities in the first entity list, the second entity list, and the third entity list, thereby further enriching the alarm method and improving the comprehensiveness of the alarm.

[0056] In addition, in some embodiments of the present disclosure, a fourth entity list can be formed by utilizing media identification entities in the business relationship graph that are not included in the first entity list, and the following processing can be performed respectively for any advertising business entity included in the fourth entity list and the second entity list: the advertising business entity is taken as an entity to be processed, and in response to determining the end of a day, a target level to which the entity to be processed belongs is determined, the entity type corresponding to the entity to be processed corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective daily inspection and alarm strategies; in response to determining that the alarm trigger conditions are met based on the income data of the entity to be processed during the day and the daily inspection and alarm strategy corresponding to the target level, an alarm operation is performed.

[0057] For media identification entities in the business relationship graph, a fourth entity list may be formed using media identification entities other than those in the first entity list, and the following processing may be performed for any media identification entity in the fourth entity list: When a day ends, the target level to which the media identification entity belongs is determined, and then if the alarm triggering conditions are determined to be met based on the revenue data of the media identification entity during the day and the daily inspection and alarm strategy corresponding to the target level, an alarm operation may be performed. The daily inspection and alarm strategy may also include alarm rules and alarm thresholds. For example, the total revenue of the media identification entity during the day may be compared with the total revenue of the previous day. If the decline is greater than the alarm threshold, it may be determined that the alarm triggering conditions are met. Alternatively, the total revenue of the media identification entity during the day may be compared with the average revenue of the previous consecutive days. If the decline is greater than the alarm threshold, it may be determined that the alarm triggering conditions are met, etc.

[0058] Accordingly, taking the fourth entity list as an example, it is necessary to save the revenue data of each media identification entity on each day, such as a two-week data retention mechanism.

[0059] According to the previous processing method, the selected TOP-L media identification entities will be monitored and alarmed in real time, but there is no corresponding processing method for the media identification entities that do not belong to TOP-L. For this reason, the scheme described in the present invention proposes that a daily patrol alarm method can also be used for the media identification entities that do not belong to TOP-L to achieve full coverage of the media identification entities, thereby avoiding the omission of alarms for abnormal problems as much as possible.

[0060] In addition, for each advertising business entity in the second entity list (which also belongs to the target entity), daily inspection and alarm can be performed in a manner similar to that of each media identification entity in the fourth entity list, thereby further enriching the alarm method and achieving a comprehensive and dynamic grasp of the overall status of the system.

[0061] Combined with the above introduction, Figure 2 Schematic diagram of the implementation process of the graph-based income perception alarm method described in the present disclosure. Figure 2 As shown in the figure, assuming that the current day is October 23, 2024, when it reaches midnight on that day, a business relationship graph can be generated based on the advertising business data obtained from different data sources on October 22, 2024, which may include advertising business entities of various entity types, etc. The advertising business entities of various entity types may include media identification entities, advertising slot identification entities, and advertiser identification entities. Figure 2As shown, the first entity list, the second entity list, the third entity list and the fourth entity list can be generated according to the business relationship graph, wherein the first entity list is an entity list consisting of the selected TOP-L media identification entities, the second entity list is an entity list consisting of the selected advertising position identification entities, the third entity list is an entity list consisting of the selected advertiser identification entities, and the fourth entity list is an entity list consisting of the media identification entities that do not belong to the TOP-L. Figure 2 As shown, monitoring and alarming can then be performed based on the generated entity lists, wherein, for each advertising business entity (target entity) in the first entity list, the second entity list and the third entity list, a multi-dimensional real-time hierarchical alarm method can be used for monitoring and alarming, and, for each advertising business entity in the first entity list, the second entity list and the third entity list, a general real-time alarm method can also be used for monitoring and alarming, and in addition, for each advertising business entity in the fourth entity list and the second entity list, a daily patrol alarm method can be used for monitoring and alarming.

[0062] Furthermore, in some embodiments of the present disclosure, when it is determined that the alarm trigger conditions are met and an alarm operation is performed, the following processing method may be adopted: in response to determining that the alarm trigger conditions are met, and determining that an alarm operation has not been performed on the target problem before or determining that the time since the last alarm operation on the target problem is greater than or equal to a predetermined time, an alarm operation is performed, and the target problem is the abnormal problem corresponding to this alarm; in response to determining that the alarm trigger conditions are met, and determining that the time since the last alarm operation on the target problem is less than a predetermined time, filtering out this alarm.

[0063] The above processing method is mainly aimed at alarms generated under the multi-dimensional real-time layered alarm method and the general real-time alarm method.

[0064] For the target entity, when it is determined that the alarm triggering conditions are met, the alarm operation can be performed directly, or, preferably, it can be determined first whether the following conditions are met: no alarm or operation has been performed on the target problem before; it is determined that the time since the last alarm operation on the target problem is greater than or equal to the predetermined time, and the specific value of the predetermined time can be determined according to actual needs. If so, an alarm operation can be performed, such as an alarm can be made to the on-duty personnel through sound and / or text, etc. In addition, alarm record information corresponding to this alarm can also be generated, which may include information such as the corresponding target problem and alarm time, and the alarm record information can be stored. In this way, when the same alarm is triggered again, if it is determined according to the alarm record information that the time since the last alarm operation on the target problem is less than the predetermined time, the alarm can be filtered out, thereby reducing the amount of unnecessary alarms and bringing a more intelligent and efficient on-duty experience to the on-duty personnel.

[0065] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are expressed as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the described order of actions, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0066] The above is an introduction to the method embodiment. The following is a further explanation of the scheme disclosed in the present invention through an apparatus embodiment.

[0067] Figure 3 FIG. 3 is a schematic diagram of the composition structure of an embodiment 300 of the income sensing alarm device based on the graph described in the present disclosure. Figure 3 As shown, it includes: a graph construction module 301, an object determination module 302 and a monitoring and alarm module 303.

[0068] The graph construction module 301 is used to generate a business relationship graph for the advertising business data obtained from different data sources in the last W days in response to determining that the generation trigger condition is met, where W is a positive integer.

[0069] The object determination module 302 is used to filter out target entities as monitoring objects from the advertising business entities in the business relationship graph.

[0070] The monitoring and alarm module 303 is used to monitor the income data of any target entity obtained in real time, and perform an alarm operation when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0071] In some embodiments of the present disclosure, the advertising business entities may include: advertising business entities of M entity types, where M is a positive integer greater than 1. Accordingly, the object determination module 302 may filter out target entities as monitoring objects from various advertising business entities, which may include: generating entity lists corresponding to the M entity types respectively, where all advertising business entities in the same entity list belong to the corresponding entity type, and determining the advertising business entities in the entity list as target entities.

[0072] In some embodiments of the present disclosure, the M types of advertising business entities may include: a media identification entity, an advertisement slot identification entity, and an advertiser identification entity, etc.

[0073] In some embodiments of the present disclosure, the object determination module 302 may generate an entity list corresponding to M entity types in the following manner: respectively obtain the average income of each media identification entity in the business relationship graph in the last N days, where N is a positive integer greater than 1, and sort the media identification entities in order of average income from high to low, and use the media identification entities in the first L positions after sorting to form a first entity list, where L is a positive integer greater than 1, and L is less than the number of media identification entities in the business relationship graph; filter out the advertising slot identification entities that have generated income in the last P days from each advertising slot identification entity in the business relationship graph, where P is a positive integer, and use the filtered advertising slot identification entities to form a second entity list; filter out the advertiser identification entities that have generated income in the last P days from each advertiser identification entity in the business relationship graph, and use the filtered advertiser identification entities to form a third entity list.

[0074] For any target entity, the monitoring alarm module 303 can monitor the income data of the target entity obtained in real time, and can perform an alarm operation when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0075] In some embodiments of the present disclosure, for any target entity, the monitoring alarm module 303 can first determine the target level to which the target entity belongs, the entity type corresponding to the target entity corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective real-time alarm strategies, and then the real-time alarm strategy corresponding to the target level can be determined as the real-time alarm strategy corresponding to the target entity, and an alarm operation can be performed when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

[0076] In some embodiments of the present disclosure, for any target entity, the monitoring alarm module 303 may determine the target level to which the target entity belongs, including: obtaining the average income of the target entity in the last Q days, where Q is a positive integer, determining the income range to which the average income belongs, and determining the level corresponding to the income range as the target level, different levels corresponding to their own income ranges, and there is no overlap between the income ranges corresponding to any two levels.

[0077] In some embodiments of the present disclosure, the real-time alarm strategy may include: alarm rules and alarm thresholds. For any target entity, the monitoring alarm module 303, in response to determining that the alarm threshold corresponding to the target entity has been individually set, can use the set alarm threshold to replace the alarm threshold in the real-time alarm strategy corresponding to the target entity, to obtain an updated real-time alarm strategy, and can perform an alarm operation when it is determined that the alarm trigger conditions are met according to the updated real-time alarm strategy.

[0078] In addition, in some embodiments of the present disclosure, for any target entity, the monitoring alarm module 303 may perform an alarm operation in response to determining that the alarm triggering conditions are met according to a general real-time alarm strategy, where the general real-time alarm strategy is a general strategy set for M entity types.

[0079] Furthermore, in some embodiments of the present disclosure, the object determination module 302 may also utilize media identification entities in the business relationship graph that are not located in the first entity list to form a fourth entity list, and the monitoring alarm module 303 may perform the following processing for any advertising business entity located in the fourth entity list and the second entity list: taking the advertising business entity as an entity to be processed, in response to determining the end of a day, determining the target level to which the entity to be processed belongs, the entity type corresponding to the entity to be processed corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to their own daily inspection and alarm strategies; in response to determining that the alarm trigger conditions are met based on the income data of the entity to be processed during the day and the daily inspection and alarm strategy corresponding to the target level, performing an alarm operation.

[0080] In some embodiments of the present disclosure, when the monitoring alarm module 303 determines that the alarm trigger conditions are met and performs an alarm operation, it may adopt the following processing method: in response to determining that the alarm trigger conditions are met, and determining that an alarm operation has not been performed on the target problem before or determining that the time since the last alarm operation on the target problem is greater than or equal to a predetermined time, perform an alarm operation, and the target problem is the abnormal problem corresponding to this alarm; in response to determining that the alarm trigger conditions are met, and determining that the time since the last alarm operation on the target problem is less than a predetermined time, filter out this alarm.

[0081] Figure 3 The specific working process of the illustrated device embodiment can refer to the relevant description in the aforementioned method embodiment and will not be described in detail.

[0082] The scheme disclosed in the present invention can be applied to the field of artificial intelligence, especially to the fields of knowledge graph, big data processing and intelligent alarm. Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It has both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing. Artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, as well as machine learning / deep learning, big data processing technology, knowledge graph technology, and other major directions.

[0083] In addition, the various data in the embodiments described in this disclosure are not for a specific user and cannot reflect the personal information of a specific user. In the technical solution of this disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0084] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0085] Figure 4 A schematic block diagram of an electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0086] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 to a random access memory (RAM) 403. In RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, ROM 402 and RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0087] Multiple components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0088] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI, Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP, Digital Signal Processing), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the methods described in the present disclosure. For example, in some embodiments, the methods described in the present disclosure may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the methods described in the present disclosure may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to execute the method described in the present disclosure in any other appropriate manner (for example, by means of firmware).

[0089] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0090] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0091] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM, Electronically Programmable Read-Only Memory), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM, Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0092] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0093] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by digital data communication (e.g., a communication network) in any form or medium. Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0094] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0095] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0096] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A graph-based income perception alarm method, comprising: In response to determining that the generation trigger condition is met, generating a business relationship graph for the advertising business data acquired from different data sources in the last W days, where W is a positive integer; Filtering target entities as monitoring objects from the advertising business entities in the business relationship graph; For any target entity, the income data of the target entity acquired in real time is monitored respectively, and an alarm operation is performed when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

2. The method according to claim 1, wherein: The advertising business entities include: advertising business entities of M entity types, where M is a positive integer greater than 1; The screening out of target entities as monitoring objects includes: generating entity lists corresponding to the M entity types respectively, the advertising business entities in the same entity list all belong to the corresponding entity type, and determining the advertising business entities in the entity list as the target entities.

3. The method according to claim 2, wherein: The M types of entity types of advertising business entities include: a media identification entity, an advertisement position identification entity, and an advertiser identification entity.

4. The method according to claim 3, wherein: The generating of entity lists corresponding to the M entity types respectively includes: The average income of each media identification entity in the business relationship graph in the last N days is obtained respectively, where N is a positive integer greater than 1, and the media identification entities are sorted in descending order according to the average income, and the first L media identification entities after sorting are used to form a first entity list, where L is a positive integer greater than 1, and L is less than the number of media identification entities in the business relationship graph; Filter out the advertising slot identification entities that have generated revenue in the last P days from the advertising slot identification entities in the business relationship graph, where P is a positive integer, and use the filtered advertising slot identification entities to form a second entity list; Advertiser identification entities that have generated revenue in the last P days are screened out from the advertiser identification entities in the business relationship graph, and the screened advertiser identification entities are used to form a third entity list.

5. The method according to claim 2, wherein: The performing of the alarm operation when the alarm triggering condition is determined to be met according to the real-time alarm strategy corresponding to the target entity includes: Determine a target level to which the target entity belongs, the entity type corresponding to the target entity corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective real-time alarm strategies; The real-time alarm policy corresponding to the target level is determined as the real-time alarm policy corresponding to the target entity, and an alarm operation is performed when it is determined according to the real-time alarm policy corresponding to the target entity that an alarm triggering condition is met.

6. The method according to claim 5, wherein: Determining the target level to which the target entity belongs includes: Get the average income of the target entity in the last Q days, where Q is a positive integer; Determine the income range to which the average income belongs, and determine the level corresponding to the income range as the target level. Different levels correspond to their own income ranges, and there is no overlap between the income ranges corresponding to any two levels.

7. The method according to claim 5, wherein: The real-time alarm strategy includes: alarm rules and alarm thresholds; The method also includes: in response to determining that the alarm threshold corresponding to the target entity has been set separately, replacing the alarm threshold in the real-time alarm policy corresponding to the target entity with the set alarm threshold to obtain an updated real-time alarm policy, and performing an alarm operation when it is determined that the alarm triggering conditions are met according to the updated real-time alarm policy.

8. The method according to claim 2, further comprising: For any target entity, in response to determining that the alarm triggering condition is met according to the universal real-time alarm strategy, an alarm operation is performed, wherein the universal real-time alarm strategy is a universal strategy set for the M entity types.

9. The method according to claim 4, further comprising: Using media identification entities in the business relationship graph that are not in the first entity list to form a fourth entity list; For any advertising service entity in the fourth entity list and the second entity list, the following processing is performed respectively: The advertising business entity is taken as an entity to be processed, and in response to determining that a day has ended, a target level to which the entity to be processed belongs is determined, the entity type corresponding to the entity to be processed corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective daily inspection alarm strategies; In response to determining that an alarm triggering condition is met based on the income data of the entity to be processed during the day and the daily inspection alarm strategy corresponding to the target level, an alarm operation is performed.

10. The method according to any one of claims 1 to 9, wherein: The alarm operation comprises: In response to determining that the alarm triggering condition is met, and determining that an alarm operation has not been performed for the target problem before or determining that the time length since the last alarm operation for the target problem is greater than or equal to a predetermined time length, an alarm operation is performed, and the target problem is the abnormal problem corresponding to this alarm; In response to determining that the alarm triggering condition is met and determining that the time since the last alarm operation for the target problem is less than the predetermined time, the current alarm is filtered out.

11. A graph-based income sensing alarm device, comprising: Graph construction module, object determination module and monitoring alarm module; The graph construction module is used to generate a business relationship graph for the advertising business data acquired from different data sources in the last W days in response to determining that the generation trigger condition is met, where W is a positive integer; The object determination module is used to filter out target entities as monitoring objects from the advertising business entities in the business relationship graph; The monitoring alarm module is used to monitor the income data of any target entity obtained in real time, and perform an alarm operation when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

12. The device according to claim 11, wherein The advertising business entities include: advertising business entities of M entity types, where M is a positive integer greater than 1; The object determination module generates entity lists corresponding to the M entity types respectively, and the advertising business entities in the same entity list all belong to the corresponding entity type, and the advertising business entities in the entity list are determined as the target entities.

13. The device according to claim 12, wherein: The M types of entity types of advertising business entities include: a media identification entity, an advertisement position identification entity, and an advertiser identification entity.

14. The device according to claim 13, wherein: The object determination module obtains the average income of each media identification entity in the business relationship graph in the last N days, where N is a positive integer greater than 1, and sorts each media identification entity in order of average income from high to low, and uses the media identification entities in the first L positions after sorting to form a first entity list, where L is a positive integer greater than 1, and L is less than the number of media identification entities in the business relationship graph, and filters out the advertising slot identification entities that have generated income in the last P days from each advertising slot identification entity in the business relationship graph, where P is a positive integer, and uses the filtered advertising slot identification entities to form a second entity list, and filters out the advertiser identification entities that have generated income in the last P days from each advertiser identification entity in the business relationship graph, and uses the filtered advertiser identification entities to form a third entity list.

15. The device according to claim 12, wherein: The monitoring alarm module determines the target level to which the target entity belongs, the entity type corresponding to the target entity corresponds to at least two levels, the target level is one of the at least two levels, and different levels correspond to respective real-time alarm strategies. The real-time alarm strategy corresponding to the target level is determined as the real-time alarm strategy corresponding to the target entity, and an alarm operation is performed when it is determined that the alarm triggering conditions are met according to the real-time alarm strategy corresponding to the target entity.

16. The device according to claim 15, wherein: The monitoring alarm module obtains the average income of the target entity in the last Q days, where Q is a positive integer, determines the income range to which the average income belongs, and determines the level corresponding to the income range as the target level. Different levels correspond to their own income ranges, and there is no overlap between the income ranges corresponding to any two levels.

17. The device according to claim 15, wherein: The real-time alarm strategy includes: alarm rules and alarm thresholds; The monitoring alarm module is further used to, in response to determining that the alarm threshold corresponding to the target entity has been individually set, replace the alarm threshold in the real-time alarm policy corresponding to the target entity with the set alarm threshold, obtain an updated real-time alarm policy, and perform an alarm operation when it is determined that the alarm triggering conditions are met according to the updated real-time alarm policy.

18. The device according to claim 12, wherein: The monitoring alarm module is further used to perform an alarm operation for any target entity in response to determining that an alarm trigger condition is met according to a general real-time alarm strategy, where the general real-time alarm strategy is a general strategy set for the M entity types.

19. The device according to claim 14, wherein: The object determination module is further used to form a fourth entity list using media identification entities in the business relationship graph that are not in the first entity list; The monitoring alarm module is further used to perform the following processing on any advertising business entity located in the fourth entity list and the second entity list: taking the advertising business entity as an entity to be processed, in response to determining the end of a day, determining the target level to which the entity to be processed belongs, the entity type corresponding to the entity to be processed corresponds to at least two levels, the target level is one of the at least two levels, different levels correspond to respective daily inspection and alarm strategies, and in response to determining that the alarm triggering conditions are met based on the income data of the entity to be processed during the day and the daily inspection and alarm strategy corresponding to the target level, performing an alarm operation.

20. The device according to any one of claims 11 to 19, wherein: The monitoring alarm module is further used to perform an alarm operation in response to determining that the alarm trigger conditions are met and that an alarm operation has not been performed on the target problem before or that the time since the last alarm operation on the target problem is greater than or equal to a predetermined time, wherein the target problem is the abnormal problem corresponding to this alarm, and in response to determining that the alarm trigger conditions are met and that the time since the last alarm operation on the target problem is less than the predetermined time, filter out this alarm.

21. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1 to 10.

23. A computer program product, comprising a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images