CN115983902A - Information pushing method and system based on user real-time event - Google Patents
Information pushing method and system based on user real-time event Download PDFInfo
- Publication number
- CN115983902A CN115983902A CN202310036987.1A CN202310036987A CN115983902A CN 115983902 A CN115983902 A CN 115983902A CN 202310036987 A CN202310036987 A CN 202310036987A CN 115983902 A CN115983902 A CN 115983902A
- Authority
- CN
- China
- Prior art keywords
- event
- user
- attribute
- real
- behavior type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an information pushing method and system based on a user real-time event, which comprises the steps of constructing an event model corresponding to the real-time event data based on the acquired real-time event data, analyzing the event model, and determining an event attribute and a user behavior type corresponding to the real-time event data; the event attribute and the user behavior type are subjected to uniqueness matching with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule in the event library, preference information of the user is determined according to the user behavior type and the user history information, and state information of the user is determined according to the event attribute and the user history information; and determining the scene where the user is located based on the preference information and the state information and by combining the real-time event data, and pushing information according to the trigger label condition corresponding to the scene where the user is located. The method disclosed by the invention can be used for realizing accurate cognition and matching of the customer demands and improving the marketing success rate.
Description
Technical Field
The present disclosure relates to information push technologies, and in particular, to an information push method and system based on a user real-time event.
Background
At present, enterprises generally face a plurality of marketing pain problems of serious product homogenization, increasingly competitive marketing and the like, and a traditional marketing strategy usually sets promotion activities or product recommendations according to analysis of historical data of customers and mining of customer characteristics. However, most of the demands of customers at the present stage are random, various and time-efficient, the traditional marketing strategy no longer adapts to the development requirements of the business at the present stage, and the development of real-time event marketing around the demands of the customers becomes a future development trend.
The existing traditional marketing mode has the following defects:
1. at the present stage, most of the demands of customers are timeliness, and the window period for capturing the marketing opportunities is too short to push proper marketing activities to proper customers in real time;
2. the collaboration between channels is lacked in the client behaviors and events, and the client events in different channels cannot be related.
3. The traditional marketing mode is difficult to multiplex the configured customer groups and activities, so that past marketing experiences cannot be precipitated into a customer group model and a strategy system.
4. Marketing flow configuration is tedious, degree of automation is low, development cycle is long, when a plurality of marketing activities have relevance, a large amount of manual configuration needs to be spent, and therefore management cost is high and marketing efficiency is low.
The information disclosed in this background section is only for enhancement of understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The embodiment of the disclosure provides an information pushing method and system based on a user real-time event, which can at least solve part of problems in the prior art, realize real-time mastering and analysis of a marketing event, and push a marketing campaign for the event in a targeted manner, so that accurate cognition and matching of customer demands are achieved, and the marketing success rate is improved.
In a first aspect of an embodiment of the present disclosure,
the information pushing method based on the real-time events of the users is provided, and comprises the following steps:
based on the acquired real-time event data, constructing an event model corresponding to the real-time event data, analyzing the event model, and determining event attributes and user behavior types corresponding to the real-time event data;
uniquely matching the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information, wherein the event library is used for receiving the real-time event data and uniquely matching the real-time event data with the preset rule;
and determining the scene where the user is located based on the preference information and the state information by combining the real-time event data, and pushing information according to a trigger label condition corresponding to the scene where the user is located, wherein the trigger label condition is used for pushing corresponding label content under the condition that the trigger rule is met.
In an alternative embodiment of the method according to the invention,
the event attribute comprises a first attribute and a second attribute, wherein the first attribute is used for indicating the attribute with the relevance to the user tag larger than the relevance threshold value, and the second attribute is used for indicating the attribute with the relevance to the user tag smaller than the relevance threshold value; the user behavior type comprises a first behavior type and a second behavior type, wherein the first behavior type is used for indicating a behavior type matched with the user label, and the second behavior type is used for indicating an accidental behavior type;
the uniquely matching the event attribute and the user behavior type with the preset rule in the event library comprises the following steps:
respectively allocating corresponding weight values to the event attribute and the user behavior type, wherein if the event attribute is a first attribute, a first weight value is allocated to the event attribute, and if the event attribute is a second attribute, a second weight value is allocated to the event attribute, and the first weight value is greater than the second weight value;
if the user behavior type is the first behavior type, a third weight value is distributed to the user behavior type, and if the user behavior type is the second behavior type, a fourth weight value is distributed to the user behavior type, wherein the third weight value is larger than the fourth weight value;
determining a rule characteristic value corresponding to the real-time event data according to the weight value, the event attribute and the user behavior type;
and calculating a plurality of spatial distances between the event database and preset rules based on the rule characteristic values, and taking the preset rule with the minimum distance in the spatial distances as a rule which is uniquely matched with the real-time event data.
In an alternative embodiment of the method according to the invention,
the method for distributing the corresponding weight values to the event attributes and the user behavior types comprises the following steps:
distributing corresponding evaluation indexes for the event attributes and the user behavior types respectively, and constructing corresponding evaluation index matrixes;
based on the evaluation index matrixes, weighting through a subjective and objective mixed strategy, and respectively assigning corresponding evaluation weight vectors to each evaluation index matrix;
and determining a weight balance point based on a preset target revenue function through the evaluation weight vector and corresponding evaluation indexes allocated to the event attribute and the user behavior type, and allocating corresponding weight values to the event attribute and the user behavior type according to the determined weight balance point and the evaluation weight vector.
In an alternative embodiment of the method according to the invention,
the assigning of the corresponding evaluation weight vector to each evaluation index matrix through the weighting of the subjective and objective mixed strategies based on the evaluation index matrix comprises:
wherein, W represents an evaluation weight vector, M and N represent the number of evaluation indexes and the number of evaluation objects respectively, wherein the evaluation objects comprise the event attribute and the user behavior type, a and b represent a first weight proportion and a second weight proportion respectively, W1 and W2 represent a first strategy weighting matrix and a second strategy weighting matrix respectively, xij represents an evaluation index matrix and represents the contribution value of the ith evaluation index in the jth evaluation object.
In an alternative embodiment of the method according to the invention,
the determining the scene where the user is located based on the preference information and the state information by combining the real-time event data, and pushing the information according to the trigger tag condition corresponding to the scene where the user is located comprises:
respectively extracting preference features from the preference information and state features from the state information, and taking the preference features and the state features as user features;
extracting a plurality of event features from the real-time event data, calculating the similarity of any two event features in the event features, and taking the event features with the similarity larger than a preset similarity threshold value as an initial event feature set;
and determining the association degree of the associated event feature set and the scene where the user is located, judging whether the event feature with the highest association degree meets the trigger tag condition, and if so, pushing information, wherein the associated event feature set is used for indicating the event feature which generates intersection with the scene where the user is located in the initial event feature set.
In an alternative embodiment of the method according to the invention,
the trigger label condition comprises a first trigger condition and a second trigger condition, the first trigger condition is used for indicating that the preference information and the label corresponding to the state information are matched with a label prestored in the event library to carry out real-time trigger, the second trigger condition is used for indicating that the preference information and the label corresponding to the state information are not matched with the label prestored in the event library to carry out delay trigger,
the pushing information according to the trigger tag condition corresponding to the scene where the user is located comprises:
if the trigger label condition is a first trigger condition, carrying out information push in real time;
and if the trigger label condition is a second trigger condition, delaying to push information.
In a second aspect of an embodiment of the present disclosure,
providing a user real-time event based information push system applying the user real-time event based information push method of any one of the preceding claims, the system comprising:
the data source module is used for acquiring original real-time event data through the acquisition device and pushing the original real-time event data to a first rabbitmq queue to provide a basic data source for the event library module;
the event library module is used for processing the real-time event data according to the basic attribute and the extended attribute through the flink, checking whether the attribute is consistent with a preset rule or not, and pushing the attribute to a second rabbitmq queue;
the strategy library module is used for receiving corresponding queue data through the second rabbitmq queue designated vsost, and pushing real-time information through the channel module based on the event model after the event and attribute screening and in combination with the anti-disturbance pushing limit;
and the channel module is used for establishing connection with each channel and event acquisition capacity and providing an event data source for the data source module.
In a third aspect of the embodiments of the present disclosure,
there is provided an information push system including:
the first unit is used for constructing an event model corresponding to the real-time event data based on the acquired real-time event data, analyzing the event model and determining an event attribute and a user behavior type corresponding to the real-time event data;
the second unit is used for performing uniqueness matching on the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information, wherein the event library is used for receiving the real-time event data and performing uniqueness matching on the real-time event data with the preset rule;
and a third unit, configured to determine, based on the preference information and the state information, a scene where the user is located in combination with the real-time event data, and perform information push according to a trigger tag condition corresponding to the scene where the user is located, where the trigger tag condition is used to indicate that, under the condition that a trigger rule is satisfied, push corresponding tag content.
In a fourth aspect of an embodiment of the present disclosure,
provided is an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the aforementioned method.
In a fifth aspect of an embodiment of the present disclosure,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as set out above.
The method disclosed by the invention determines the event attribute and the user behavior type corresponding to the real-time event data based on the acquired real-time event data, and uniquely matches the event attribute and the user behavior type with the preset rule in the event library, so that the accuracy of information push content can be improved;
corresponding weight values are distributed according to the event attributes and the user behavior types, and the advantages of two weight assignment methods, namely subjective weight and objective weight, are integrated, so that a proper balance point can be found between the two methods, and the objective accuracy of weight assignment and the priority of important index evaluation are improved.
The computation of more than ten thousand levels of data is completed within a few seconds in a dynamic configuration mode, labels can be computed in real time based on trillion levels of historical data, real-time data processing and real-time label computation are completed, the requirements of users are rapidly identified, and targeted marketing service is provided.
The system disclosed by the invention realizes real-time calculation and pushing by using flink and rabbitmq, the coupling between systems is reduced by the whole scheme, and the intermediate storage is reduced to realize real-time event marketing.
The channel module adapts to the difference of various channel systems, establishes the connection with each channel and the event acquisition capability, registers the service and control route of each application system through a uniform central node, avoids point-to-point tight coupling connection between the systems, and realizes efficient development and implementation and message processing.
The strategy library module supports the establishment of a unified strategy system, can construct strategies according to different application scenes, marketing targets and strategy groups, and precipitates the successful marketing experience in the past to form reusable strategy assets.
The atom strategy and the flow canvas support more complex comprehensive marketing scenes, and the marketing automation flow can be quickly brought on line through a dragging and pulling mode, so that the efficiency of marketing activities is improved. And all current marketing activities are displayed in a calendar executing mode, and starting and stopping operations of the marketing activities on the calendar executing mode are supported, so that the marketing process is clearer, simpler and more convenient.
Drawings
Fig. 1 is a schematic flow chart of an information pushing method based on a user real-time event according to an embodiment of the present disclosure.
Fig. 2 is a schematic structural diagram of an information push system based on a user real-time event according to an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present disclosure and in the drawings described above, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in other sequences than those illustrated or described herein.
It should be understood that, in various embodiments of the present disclosure, the sequence number of each process does not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present disclosure.
It should be understood that in the present disclosure, "including" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that in the present disclosure, "plurality" means two or more. "and/or" is merely an association relationship describing an associated object, meaning that there may be three relationships, for example, and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprising a, B and C", "comprising a, B, C" means that all three of a, B, C are comprised, "comprising a, B or C" means comprising one of three of a, B, C, "comprising a, B and/or C" means comprising any 1 or any 2 or 3 of three of a, B, C.
It should be understood that in this disclosure, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, from which B can be determined. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.
As used herein, the term "if" may be interpreted as "at ... …" or "in response to a determination" or "in response to a detection" depending on the context.
The technical solution of the present disclosure is explained in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a schematic flow diagram of an information pushing method based on a user real-time event according to an embodiment of the present disclosure, and as shown in fig. 1, the method includes:
s101, based on the acquired real-time event data, constructing an event model corresponding to the real-time event data, analyzing the event model, and determining event attributes and user behavior types corresponding to the real-time event data;
illustratively, the real-time event data may include transaction, client, product, etc. data related to marketing, for example, transaction occurred at the current time node by the target client, the client's ID, and product data bought and sold by the client; in addition, the real-time event data may include behaviors of the user in the operated application system, such as purchasing commodities, browsing activities, browsing commodities, sharing activities, and the like, and different application systems have different behaviors.
And the event model corresponding to the real-time event data is used for mathematically abstracting the real-time event data, extracting the characteristics corresponding to the real-time event data, carrying out rule matching on the extracted characteristics and an event library, and simultaneously providing a basic data source for the event library. The event model may include an existing neural network model, and the comparison of the embodiments of the present disclosure is not repeated.
The event attribute corresponding to the real-time event data may include data such as an event ID, an event name, an event occurrence time, an event classification ID, an event source, and the like, and in addition, an extension attribute corresponding to each event may be created in a customized manner according to the current event attribute, for example, attribute information of a specific event is added, and an event occurs in a specific scene.
The user behavior types may include a behavior type matching the user representation, which may include, for example, the user representation being a single young woman, and a casual behavior type matching the user representation, which may include buying a tide play, traveling a country ticket, etc.; the type of casual activity may include the purchase of maternal and infant products such as milk powder and infant products.
S102, uniquely matching the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule in the event library, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information;
illustratively, the event attribute comprises a first attribute and a second attribute, wherein the first attribute is used for indicating the attribute with the relevance of the user tag larger than the relevance threshold value, and the second attribute is used for indicating the attribute with the relevance of the user tag smaller than the relevance threshold value; the user behavior type comprises a first behavior type and a second behavior type, wherein the first behavior type is used for indicating a behavior type matched with the user label, and the second behavior type is used for indicating a contingency behavior type;
the relevance between the event attribute and the user tag can be determined by calculating the spatial distance between the vector corresponding to the event attribute and the vector corresponding to the user tag, and determining whether the event attribute is the first attribute or the second attribute by comparing the spatial distance with a preset relevance threshold value.
For example, the preference information of the user may be determined according to the behavior type of the user and the historical information of the user, specifically, if the tag corresponding to the user is a pregnant woman, the behavior type of the user is a first behavior type, that is, a behavior type matched with the user tag, and the historical purchase record of the user in the past 30 days is combined, for example, the user purchases a maternal-infant product in the past for more than 500 yuan, then the preference information of the user may be determined as a preferred maternal-infant product;
for example, the status information of the user may be determined according to an event attribute and user history information, specifically, if the event attribute is an attribute that the association with the user tag is greater than an association threshold, or if the tag corresponding to the user is a pregnant woman, the event attribute is that a crib is purchased, and in combination with the historical purchase record of the user in the last 30 days, for example, that the user purchases a maternal-infant product for more than 500 yuan in the past, the status information of the user may be determined as a user with a higher consumption intention.
In an alternative embodiment of the method according to the invention,
the uniquely matching the event attribute and the user behavior type with a preset rule in an event library comprises the following steps:
respectively allocating corresponding weight values to the event attribute and the user behavior type, wherein if the event attribute is a first attribute, a first weight value is allocated to the event attribute, and if the event attribute is a second attribute, a second weight value is allocated to the event attribute, and the first weight value is greater than the second weight value;
if the user behavior type is a first behavior type, a third weight value is distributed to the user behavior type, and if the user behavior type is a second behavior type, a fourth weight value is distributed to the user behavior type, wherein the third weight value is larger than the fourth weight value;
determining a rule characteristic value corresponding to the real-time event data according to the weight value, the event attribute and the user behavior type;
and calculating a plurality of spatial distances between the event database and preset rules based on the rule characteristic values, and taking the preset rule with the minimum distance in the spatial distances as a rule which is uniquely matched with the real-time event data.
Illustratively, the event attribute and the user behavior type play different roles in information pushing, wherein the event attribute is used for indicating the occurrence of an event, and the user behavior is used for indicating the behavior type related to the user portrait, and different weight values are allocated to the event attribute and the user behavior type, so that the emphasis of the event attribute and the behavior type in the last information pushing can be highlighted, and the accuracy and the pertinence of the information pushing are improved.
The first attribute is used for indicating the attribute with the relevance to the user tag larger than the relevance threshold, and the second attribute is used for indicating the attribute with the relevance to the user tag smaller than the relevance threshold; attributes greater than the association threshold, for which the dominance should be strengthened to make the final result more accurate, indicate a strong association with a given user label; correspondingly, the second weight value is smaller than the first weight value, the proportion of the second attribute in subsequent calculation is reduced, the calculation pressure is balanced, and the pushing result is more targeted. The first weight value and the second weight value vary according to the size of the first attribute value and the second attribute value.
Correspondingly, the third weight value and the fourth weight value are also the same, and are not described herein again.
Determining a rule characteristic value corresponding to the real-time event data according to the weight value, the event attribute and the user behavior type;
calculating a plurality of spatial distances to preset rules in the event library based on the rule characteristic values, and taking the preset rule with the minimum distance in the plurality of spatial distances as a rule which is uniquely matched with the real-time event data;
exemplarily, the rule characteristic value of the embodiment of the present disclosure may be represented as G = [ Q ] i ,S j ,Y j ]
Wherein j belongs to [1,2,3,4], j belongs to [1,2], qi represents the ith weight value, sj represents the jth event attribute, and Yj represents the jth user behavior type.
The method for calculating the plurality of spatial distances to the preset rules in the event library based on the rule characteristic values comprises the following steps:
the spatial distance is calculated according to the following formula:
wherein Gxm represents a preset rule Gx in the mth event library.
The preset rule with the minimum distance in the plurality of spatial distances is used as the rule which is uniquely matched with the real-time event data, the prediction rule with the minimum distance in the spatial distances represents the rule which is closest to the real-time event data, the rule which is most matched with the real-time event data can be selected from the preset rules in the event library, and the accuracy of information pushing is improved.
In an alternative embodiment of the method according to the invention,
the method for assigning the corresponding weight values to the event attributes and the user behavior types comprises the following steps:
distributing corresponding evaluation indexes for the event attributes and the user behavior types respectively, and constructing corresponding evaluation index matrixes;
respectively endowing each evaluation index matrix with a corresponding evaluation weight vector through a mixed weight endowing strategy based on the evaluation index matrix;
and determining a weight balance point based on a preset target revenue function through the evaluation weight vector and corresponding evaluation indexes allocated to the event attribute and the user behavior type, and allocating corresponding weight values to the event attribute and the user behavior type according to the determined weight balance point and the evaluation weight vector.
Corresponding evaluation indexes are distributed aiming at different evaluation objects, corresponding weight values are determined by integrating multiple factors, uncertainty of various factors can be quantified, the weight values are aggregated, and an optimal solution is guaranteed.
For example, the evaluation index matrix may be represented as:
wherein Xij represents an evaluation index matrix and represents a contribution value of the ith evaluation index in the jth evaluation object, wherein the contribution value can be represented as the proportion of a single evaluation index in all the evaluation indexes of a certain evaluation object; indicating the accuracy of the evaluation index on the evaluation object.
In an alternative embodiment of the method according to the invention,
the assigning, based on the evaluation index matrices and by a hybrid weighting strategy, corresponding evaluation weight vectors to each evaluation index matrix respectively includes:
wherein, W represents an evaluation weight vector, M and N represent the number of evaluation indexes and the number of evaluation objects respectively, wherein the evaluation objects comprise the event attribute and the user behavior type, a and b represent a first weight proportion and a second weight proportion respectively, W1 and W2 represent a first strategy weighting matrix and a second strategy weighting matrix respectively, xij represents an evaluation index matrix and represents the contribution value of the ith evaluation index in the jth evaluation object.
The step of determining a weight balance point based on a preset target revenue function by the evaluation weight vector and allocating corresponding evaluation indexes to the event attribute and the user behavior type, and allocating corresponding weight values to the event attribute and the user behavior type according to the determined weight balance point and the evaluation weight vector comprises the following steps:
the method for determining the weight balance point may be as follows:
wherein, the weight balance point is the minimum value of minL, f i (W) represents a target revenue function, xij represents an evaluation index matrix, Z represents the number of loop iterations, u represents a resolution coefficient, and Δ k represents closeness.
Assigning corresponding weight values may be as follows:
Q＝minL*W
and setting a combined weight value according to the weight balance point and the evaluation weight vector.
S103, determining the scene where the user is located based on the preference information and the state information and in combination with the real-time event data, and pushing information according to a trigger tag condition corresponding to the scene where the user is located, wherein the trigger tag condition is used for pushing corresponding tag content under the condition that a trigger rule is met.
Illustratively, the tag is a bridge connecting the user, the scene and the event, the scene where the user is located is determined by combining preference information and state information of the user and the real-time event data, and information push is performed according to a trigger tag condition corresponding to the scene where the user is located.
In an alternative embodiment of the method according to the invention,
the determining the scene where the user is located based on the preference information and the state information in combination with the real-time event data and pushing information according to the trigger tag condition corresponding to the scene where the user is located comprises:
respectively extracting preference features from the preference information and state features from the state information, and taking the preference features and the state features as user features;
extracting a plurality of event features from the real-time event data, calculating the similarity of any two event features in the event features, and taking the event features with the similarity larger than a preset similarity threshold value as an initial event feature set;
and determining the association degree of the associated event feature set and the scene where the user is located, judging whether the event feature with the highest association degree meets the trigger tag condition, and if so, pushing information, wherein the associated event feature set is used for indicating the event feature which generates intersection with the scene where the user is located in the initial event feature set.
For example, the similarity of any two event features in the plurality of event features may be calculated by referring to a pearson correlation coefficient method using a variation of cosine similarity. The increasing speed of the events is often slower than the increasing speed of the users and is smaller than the magnitude order of the users, so that the recommending speed is often faster by considering the similarity between the events from the aspect of the characteristics of the users.
In an alternative embodiment of the method according to the invention,
the trigger label conditions comprise a first trigger condition and a second trigger condition, the first trigger condition is used for indicating that the preference information and the label corresponding to the state information are matched with a pre-stored label in the event library to carry out real-time trigger, the second trigger condition is used for indicating that the preference information and the label corresponding to the state information are not matched with the pre-stored label in the event library to carry out delay trigger,
the pushing information according to the trigger tag condition corresponding to the scene where the user is located comprises:
if the trigger label condition is a first trigger condition, carrying out information push in real time;
and if the trigger label condition is a second trigger condition, delaying to push information.
Illustratively, the trigger tag conditions comprise a first trigger condition and a second trigger condition, wherein the first trigger condition is real-time trigger, real-time calculation can be performed according to real-time behavior data of a user in a real-time trigger mode, the requirements of the user are rapidly identified, and information pushing is performed in real time;
the second trigger condition is a delayed trigger, that is, the preference information and the tag corresponding to the state information are not matched with the pre-stored tag in the event library, for example, the user is a pregnant woman, but the heavy locomotive is searched by accident, and the tag is not matched with the pre-stored tag in the event library, so that the delayed trigger can be considered, the message is prevented from being sent by mistake, and various requirements of the user can be met.
In a second aspect of an embodiment of the present disclosure,
providing a user real-time event based information push system applying the user real-time event based information push method according to any one of the foregoing, fig. 2 is a schematic structural diagram of the user real-time event based information push system according to the embodiment of the present disclosure, as shown in fig. 2, the system includes:
the data source module is used for acquiring original real-time event data through the acquisition device and pushing the original real-time event data to a first rabbitmq queue to provide a basic data source for the event library module;
the event library module is used for processing the real-time event data according to the basic attribute and the extended attribute through the flink, checking whether the attribute is consistent with a preset rule or not, and pushing the attribute to a second rabbitmq queue;
the strategy library module is used for receiving corresponding queue data through the second rabbitmq queue designated vsost, and pushing real-time information through the channel module based on the event model after the event and attribute screening and in combination with the anti-disturbance pushing limit;
and the channel module is used for establishing connection with each channel and event acquisition capacity and providing an event data source for the data source module.
Illustratively, flink is a distributed processing engine oriented to distributed streaming data and batch data, which can provide both streaming (DataStreams) and batch (DataSet) types of functionality.
Flink supports exact-Once, thereby ensuring that each message is consumed only Once. The message passing property exact-Once of Flink is an implementation of the Chandy and Lamport based distributed snapshot paper. The user can customize Checkpoint interval time, a special snapshot flag message (Barrier) will be inserted into all data sources periodically, barrier and other data messages flow together in a directed acyclic graph, but it will not be processed by the user-defined business logic, since the storage of snapshots is asynchronous and incremental operations so that the processing of data messages will not be blocked. If abnormal conditions such as node crash occur, only the distributed snapshot state which is successfully saved last time can be recovered.
In an alternative embodiment of the method according to the invention,
the atomic strategy establishes an event model through event and attribute screening, the event model needs to be configured with triggered events and attribute screening rules corresponding to the events, and targeted information pushing is configured for the event model.
In an alternative embodiment of the method according to the invention,
the process canvas displays information of all current activities in a calendar execution mode through an automatic process in an dragging and pulling mode, and supports the operations of starting and stopping the activities on the calendar execution.
The data source module is responsible for accessing real-time data of transactions, customers, products and the like related to event marketing, and acquiring original event data and pushing the original event data to the rabbitmq through the acquisition device so as to provide a basic data source for the event library module.
In an exemplary manner, the first and second electrodes are,
the event library module defines events and attribute rules (including basic attributes such as event ID, event name, event occurrence time, event classification ID, event source and client ID) around the user, occurrence time, occurrence place, specific content and the like, and establishes expansion attributes corresponding to each event in a self-defined manner. And then abstracting a data model based on the accessed real-time data, and establishing a corresponding relation between the data model and the rule.
The event library uses flink and rabbitmq to realize real-time calculation and pushing of mass data. And the event task can be rapidly configured and submitted by one key, the flink processes the event data according to the basic attribute and the expansion attribute, and pushes the event data to a downstream rabbitmq queue after checking whether the attribute is consistent with the rule. Wherein the message middleware rabbitmq ensures that the data is accurate once by enabling the uniqueness value to interact with the flink's checkpoint. The whole module enables a rabbitmq dead trust queue to be used for checking abnormal data, and a yarn deployment mode is adopted to facilitate user migration. And the resource management is flexible, and the capacity expansion and operation and maintenance are convenient.
The event library module can trigger the change of the automatic maintenance attribute according to the events of various channels, receive event data in real time, process the events, check the event attribute, update the state of the client object and be used for supporting the accurate control of which time point to recommend information to the client in the marketing process.
Optionally, the policy library module supports establishment of a unified policy system, and can construct policies according to different application scenarios, marketing targets and policy groups, and precipitate successful marketing experiences in the past to form reusable policy assets. The marketing strategy library is established, so that the marketing strategy with excellent effect can be changed into a template with parameters to be copied or adjusted quickly.
Marketing strategies are divided into an atomic strategy and a flow canvas according to the complexity of marketing scenes:
the "atomic policy" supports independent marketing scenarios with finer granularity, and establishes an event model (such as a customer behavior model, a customer transaction model, a product state model, etc.) by screening events and attributes, wherein the event model needs to be configured with triggered events and attribute screening rules corresponding to the events, and is configured with targeted marketing asset push for the event model, and the marketing assets include, but are not limited to: dialogs, marketing campaigns, equity, and products.
The flow canvas supports more complex comprehensive marketing scenes, and the marketing automation flow can be quickly brought on line through a dragging and pulling mode, so that the efficiency of marketing activities is improved. The comprehensive marketing scene usually comprises a plurality of atomic strategies, and relevance and state change among the atomic strategies can be comprehensively considered through the process canvas, so that timely and accurate information pushing is provided for clients. And all current marketing activities are displayed in a calendar executing mode, and starting and stopping operations of the marketing activities on the calendar executing mode are supported, so that the marketing process is clearer, simpler and more convenient.
The marketing strategy module supports an anti-disturbance function at the same time, pushing limits of the same user in an atomic strategy, a flow canvas or a global range can be set in a user-defined mode, namely the number of times that the same user can be pushed at most in a period is used for controlling the marketing frequency of real-time event marketing to the same user, user experience is improved, the user is prevented from being disturbed, and the marketing effect is maximized.
The strategy library module appoints the vhost to receive the corresponding queue data through the rabbitmq, and real-time pushing of marketing assets is carried out through the channel module according to an event model after the events and attributes are screened in the atom strategy or the flow canvas and in combination with anti-disturbance pushing limitation.
In an alternative embodiment of the method according to the invention,
channels should include, but are not limited to: intelligent outbound call, application program message, short message, template message and customer relationship management system. The channel module adapts to the difference of various channel systems, establishes connection with each channel and event acquisition capability, and provides an event data source for the data source module. The module registers the service and control route of each application system through a unified central node by the basic framework support of a standardized service communication bus and an integrated service platform, avoids point-to-point tight coupling connection between the systems, and realizes efficient development and implementation and message processing.
In a third aspect of the embodiments of the present disclosure,
there is provided an information push system including:
the first unit is used for constructing an event model corresponding to the real-time event data based on the acquired real-time event data, analyzing the event model and determining an event attribute and a user behavior type corresponding to the real-time event data;
the second unit is used for performing uniqueness matching on the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information, wherein the event library is used for receiving the real-time event data and performing uniqueness matching on the real-time event data with the preset rule;
and a third unit, configured to determine, based on the preference information and the state information, a scene where the user is located in combination with the real-time event data, and perform information push according to a trigger tag condition corresponding to the scene where the user is located, where the trigger tag condition is used to indicate that, under the condition that a trigger rule is satisfied, push corresponding tag content.
In a fourth aspect of an embodiment of the present disclosure,
provided is an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the aforementioned method.
In a fifth aspect of the embodiments of the present disclosure,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as described above.
The present invention may be methods, apparatus, systems and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for carrying out aspects of the invention.
The computer-readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as a punch card or an in-groove protruding structure with instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present invention may be assembler instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processing unit of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It is noted that, unless expressly stated otherwise, all features disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features. Where used, further, preferably, still further and more preferably is a brief introduction to the description of the other embodiment based on the foregoing embodiment, the combination of the contents of the further, preferably, still further or more preferably back strap with the foregoing embodiment being a complete construction of the other embodiment. Several further, preferred, still further or more preferred arrangements of the belt after the same embodiment may be combined in any combination to form a further embodiment.
It will be appreciated by persons skilled in the art that the embodiments of the invention described above and shown in the drawings are given by way of example only and are not limiting of the invention. The objects of the invention have been fully and effectively accomplished. The functional and structural principles of the present invention have been shown and described in the examples, and any variations or modifications of the embodiments of the present invention may be made without departing from the principles.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present disclosure, and not for limiting the same; while the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present disclosure.
Claims (10)
1. An information push method based on a user real-time event is characterized by comprising the following steps:
based on the acquired real-time event data, constructing an event model corresponding to the real-time event data, analyzing the event model, and determining an event attribute and a user behavior type corresponding to the real-time event data;
uniquely matching the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information, wherein the event library is used for receiving the real-time event data and uniquely matching the real-time event data with the preset rule;
and determining the scene where the user is located based on the preference information and the state information in combination with the real-time event data, and pushing information according to a trigger tag condition corresponding to the scene where the user is located, wherein the trigger tag condition is used for pushing corresponding tag content under the condition that a trigger rule is met.
2. The method of claim 1, wherein the event attribute comprises a first attribute and a second attribute, wherein the first attribute is used for indicating an attribute with the association with the user tag larger than an association threshold value, and the second attribute is used for indicating an attribute with the association with the user tag smaller than the association threshold value; the user behavior type comprises a first behavior type and a second behavior type, wherein the first behavior type is used for indicating a behavior type matched with the user label, and the second behavior type is used for indicating a contingency behavior type;
the uniquely matching the event attribute and the user behavior type with a preset rule in an event library comprises the following steps:
respectively allocating corresponding weight values to the event attribute and the user behavior type, wherein if the event attribute is a first attribute, a first weight value is allocated to the event attribute, and if the event attribute is a second attribute, a second weight value is allocated to the event attribute, and the first weight value is greater than the second weight value;
if the user behavior type is a first behavior type, a third weight value is distributed to the user behavior type, and if the user behavior type is a second behavior type, a fourth weight value is distributed to the user behavior type, wherein the third weight value is larger than the fourth weight value;
determining a rule characteristic value corresponding to the real-time event data according to the weight value, the event attribute and the user behavior type;
and calculating a plurality of spatial distances between the event database and preset rules based on the rule characteristic values, and taking the preset rule with the minimum distance in the spatial distances as a rule which is uniquely matched with the real-time event data.
3. The method of claim 2, wherein the assigning the event attribute and the user behavior type with corresponding weight values comprises:
distributing corresponding evaluation indexes for the event attributes and the user behavior types respectively, and constructing corresponding evaluation index matrixes;
respectively endowing each evaluation index matrix with a corresponding evaluation weight vector through a mixed weight endowing strategy based on the evaluation index matrix;
and determining a weight balance point based on a preset target revenue function through the evaluation weight vector and corresponding evaluation indexes allocated to the event attribute and the user behavior type, and allocating corresponding weight values to the event attribute and the user behavior type according to the determined weight balance point and the evaluation weight vector.
4. The method according to claim 3, wherein the assigning each evaluation index matrix with a corresponding evaluation weight vector through a hybrid weighting strategy based on the evaluation index matrices comprises:
wherein, W represents an evaluation weight vector, M and N represent the number of evaluation indexes and the number of evaluation objects respectively, wherein the evaluation objects comprise the event attribute and the user behavior type, a and b represent a first weight proportion and a second weight proportion respectively, W1 and W2 represent a first strategy weighting matrix and a second strategy weighting matrix respectively, xij represents an evaluation index matrix and represents the contribution value of the ith evaluation index in the jth evaluation object.
5. The method of claim 1, wherein the determining a scene where the user is located based on the preference information and the state information in combination with the real-time event data, and pushing information according to a trigger tag condition corresponding to the scene where the user is located comprises:
extracting preference features from the preference information and extracting state features from the state information respectively, and taking the preference features and the state features as user features;
extracting a plurality of event features from the real-time event data, calculating the similarity of any two event features in the event features, and taking the event features with the similarity larger than a preset similarity threshold value as an initial event feature set;
and determining the association degree of the associated event feature set and the scene where the user is located, judging whether the event feature with the highest association degree meets the trigger tag condition, and if so, pushing information, wherein the associated event feature set is used for indicating the event feature which generates intersection with the scene where the user is located in the initial event feature set.
6. The method of claim 1, wherein the trigger tag condition comprises a first trigger condition and a second trigger condition, the first trigger condition is used for indicating that the preference information and the tag corresponding to the status information are matched with a pre-stored tag in the event library for real-time triggering, the second trigger condition is used for indicating that the preference information and the tag corresponding to the status information are not matched with the pre-stored tag in the event library for delayed triggering,
the pushing information according to the trigger tag condition corresponding to the scene where the user is located comprises:
if the trigger label condition is a first trigger condition, carrying out information push in real time;
and if the trigger label condition is a second trigger condition, delaying to push information.
7. A user real-time event based information push system applying the user real-time event based information push method according to any one of claims 1 to 6, the system comprising:
the data source module is used for acquiring original real-time event data through the acquisition device and pushing the data to the first rabbitmq queue to provide a basic data source for the event library module;
the event library module is used for processing the real-time event data according to the basic attribute and the extended attribute through the flink, checking whether the attribute is consistent with a preset rule or not, and pushing the attribute to a second rabbitmq queue;
the strategy library module is used for receiving corresponding queue data through the second rabbitmq queue designated vhost, and carrying out real-time information push through the channel module based on the event model after the event and the attribute are screened and in combination with the anti-disturbance push limitation;
and the channel module is used for establishing connection with each channel and event acquisition capacity and providing an event data source for the data source module.
8. An information push system, comprising:
the first unit is used for constructing an event model corresponding to the real-time event data based on the acquired real-time event data, analyzing the event model and determining an event attribute and a user behavior type corresponding to the real-time event data;
the second unit is used for performing uniqueness matching on the event attribute and the user behavior type with a preset rule in an event library, if the event attribute and the user behavior type are matched with the preset rule, determining preference information of the user according to the user behavior type and the user history information, and determining state information of the user according to the event attribute and the user history information, wherein the event library is used for receiving the real-time event data and performing uniqueness matching on the real-time event data with the preset rule;
a third unit, configured to determine a scene where the user is located based on the preference information and the state information in combination with the real-time event data, and perform information push according to a trigger tag condition corresponding to the scene where the user is located, where the trigger tag condition is used to indicate that, in a case where a trigger rule is satisfied, a corresponding tag content is pushed.
9. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the memory-stored instructions to perform the method of any of claims 1 to 6.
10. A computer readable storage medium having computer program instructions stored thereon, which when executed by a processor implement the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310036987.1A CN115983902B (en) | 2023-01-10 | 2023-01-10 | Information pushing method and system based on user real-time event |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310036987.1A CN115983902B (en) | 2023-01-10 | 2023-01-10 | Information pushing method and system based on user real-time event |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115983902A true CN115983902A (en) | 2023-04-18 |
| CN115983902B CN115983902B (en) | 2023-10-20 |
Family
ID=85959548
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310036987.1A Active CN115983902B (en) | 2023-01-10 | 2023-01-10 | Information pushing method and system based on user real-time event |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115983902B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117591320A (en) * | 2024-01-17 | 2024-02-23 | 北京智麟科技有限公司 | Optimized pushing method and system based on multi-channel message |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561603A (en) * | 2020-12-25 | 2021-03-26 | 杭州云徙科技有限公司 | Event label implementation method and system based on real-time user behaviors |
| CN113065894A (en) * | 2021-03-24 | 2021-07-02 | 天天惠民(北京)智能物流科技有限公司 | Data collection method and device based on user portrait and order analysis and storage medium |
| WO2022100518A1 (en) * | 2020-11-12 | 2022-05-19 | 北京沃东天骏信息技术有限公司 | User profile-based object recommendation method and device |
| CN115002200A (en) * | 2022-05-31 | 2022-09-02 | 平安银行股份有限公司 | User portrait based message pushing method, device, equipment and storage medium |
-
2023
- 2023-01-10 CN CN202310036987.1A patent/CN115983902B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022100518A1 (en) * | 2020-11-12 | 2022-05-19 | 北京沃东天骏信息技术有限公司 | User profile-based object recommendation method and device |
| CN112561603A (en) * | 2020-12-25 | 2021-03-26 | 杭州云徙科技有限公司 | Event label implementation method and system based on real-time user behaviors |
| CN113065894A (en) * | 2021-03-24 | 2021-07-02 | 天天惠民(北京)智能物流科技有限公司 | Data collection method and device based on user portrait and order analysis and storage medium |
| CN115002200A (en) * | 2022-05-31 | 2022-09-02 | 平安银行股份有限公司 | User portrait based message pushing method, device, equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117591320A (en) * | 2024-01-17 | 2024-02-23 | 北京智麟科技有限公司 | Optimized pushing method and system based on multi-channel message |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115983902B (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110135901A (en) | A kind of enterprise customer draws a portrait construction method, system, medium and electronic equipment | |
| US20190122268A1 (en) | Assessing probability of winning an in-flight deal for different price points | |
| CN111737546B (en) | Method and device for determining entity service attribute | |
| US11257110B2 (en) | Augmenting missing values in historical or market data for deals | |
| US20170161855A1 (en) | Optimized small screen device to display visual elements in a real property dashboard involving predictive analytics | |
| US11501239B2 (en) | Metric specific machine learning model improvement through metric specific outlier removal | |
| US20200410355A1 (en) | Explainable machine learning based on heterogeneous data | |
| US10902446B2 (en) | Top-down pricing of a complex service deal | |
| CN112925664A (en) | Target user determination method and device, electronic equipment and storage medium | |
| CN115983902B (en) | Information pushing method and system based on user real-time event | |
| US10528965B2 (en) | Bundling application programming interfaces | |
| CN114943279A (en) | Method, device and system for predicting bidding cooperative relationship | |
| CN114282011A (en) | Knowledge graph construction method and device, and graph calculation method and device | |
| US11693579B2 (en) | Value-based replication of streaming data | |
| US20220207606A1 (en) | Prediction of future occurrences of events using adaptively trained artificial-intelligence processes | |
| US20210142406A1 (en) | Vehicle selection platform | |
| CN110795638A (en) | Method and apparatus for outputting information | |
| CN111476511B (en) | Data display method and device for specific risks faced by server | |
| US11379929B2 (en) | Advice engine | |
| CN111951011A (en) | Monitoring system threshold determination method and device | |
| CN111882339A (en) | Prediction model training and response rate prediction method, device, equipment and storage medium | |
| KR102449831B1 (en) | Electronic device for providing information regarding new text, server for identifying new text and operation method thereof | |
| CN113762699A (en) | Joint defense joint control risk processing method and device | |
| CN113011986A (en) | Abstract method and system for insurance calculation | |
| CN118037468A (en) | Recovery risk prediction model training and bill recovery risk assessment method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |