CN107040863B - Real-time service recommendation method and system - Google Patents
Real-time service recommendation method and system Download PDFInfo
- Publication number
- CN107040863B CN107040863B CN201510459098.1A CN201510459098A CN107040863B CN 107040863 B CN107040863 B CN 107040863B CN 201510459098 A CN201510459098 A CN 201510459098A CN 107040863 B CN107040863 B CN 107040863B
- Authority
- CN
- China
- Prior art keywords
- event
- real
- user
- time
- rule
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/12—Messaging; Mailboxes; Announcements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/60—Subscription-based services using application servers or record carriers, e.g. SIM application toolkits
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W8/00—Network data management
- H04W8/18—Processing of user or subscriber data, e.g. subscribed services, user preferences or user profiles; Transfer of user or subscriber data
- H04W8/20—Transfer of user or subscriber data
- H04W8/205—Transfer to or from user equipment or user record carrier
Abstract
The invention discloses a real-time service recommendation method, which comprises the following steps: acquiring and monitoring network behavior information of a mobile user in real time, and acquiring event source data in the network behavior information; analyzing the event source data in real time according to event rules, and sending real-time results to event users; and taking the real-time result as a data interface, performing contact judgment by combining the historical behavior of the mobile user, controlling a contact channel between the user and the mobile user according to the event rule, and performing real-time service recommendation activity. The invention also discloses a real-time service recommendation system.
Description
Technical Field
The invention relates to the technical field of mobile data processing, in particular to a real-time service recommendation method and system.
Background
With the increase of the scale of a user group and the increase of behaviors of accessing the internet by users, the existing mobile user data is not a set of only one or two types of basic data, but forms a big data environment aggregated by various data such as call tickets (conversation, flow, service and the like), signaling (position signaling, behavior signaling and the like), service platforms (music, games, videos, positions and the like), third party partners and the like, but is limited by the processing capacity of a mobile terminal, the privacy of the user and the flow of the user, and the prior art cannot rely on the terminal to acquire and transmit complete user information, which brings great pressure to the processing of mobile data. In addition, the mobile internet provides more and more service contents for users, and the difficulty of obtaining the most timely and most relevant information from mass data by users only depending on personal devices is also increasing, so that currently, many communication operator companies and internet companies have started research and exploration of mobile service recommendation technologies.
In the prior art, mobile service recommendation is mainly completed by sending short messages and making calls, user behaviors can be analyzed for realizing accurate recommendation, but the mobile service recommendation is limited by a message acquisition and recommendation channel and generally only can be analyzed by collecting short messages and call bill records by an operator background system. Because the information quantity carried by the short message and the ticket is limited, effective analysis can be carried out only through data accumulation in a longer time period (the ticket record is usually updated according to the day), the data statistics of the conventional recommendation scheme is longer in time consumption and lagged in information, and the recommended timeliness requirement cannot be met. In order to ensure the timeliness of recommendation, the prior art also presents a real-time recommendation scheme based on the current behavior and/or state of a user, but the existing scheme only analyzes short messages and ticket records with smaller time granularity actually, and has low degree of engagement with the actual preference and the like of the user due to too little data; meanwhile, the matchable information is too little, the scene is simple and fixed, and the setting cannot be performed universally, so that the scene needs to be configured and reset independently in each activity, and the workload is large and the cost is high.
Disclosure of Invention
In order to solve the existing technical problems, embodiments of the present invention are expected to provide a real-time service recommendation method and system, which can enhance timeliness of service recommendation, have a higher degree of fitting with actual preferences of users, and have a wider adaptability.
The technical scheme of the embodiment of the invention is realized as follows:
in one aspect of the embodiments of the present invention, a real-time service recommendation method is provided, including:
acquiring and monitoring network behavior information of a mobile user in real time, and acquiring event source data in the network behavior information;
analyzing the event source data in real time according to event rules, and sending real-time results to event users;
and taking the real-time result as a data interface, performing contact judgment by combining the historical behavior of the mobile user, controlling a contact channel between the user and the mobile user according to the event rule, and performing real-time service recommendation activity.
Preferably, the method further comprises: and extracting event elements according to the service scene, and pre-configuring the event rules.
Preferably, after the performing the real-time service recommendation activity, the method further includes: and automatically promoting the next recommendation or contact action according to the feedback information of the mobile user and the event rule, and performing interactive service recommendation and service activities for multiple times.
Preferably, the parsing process includes event standardization, event parsing and event distribution; the event standardization is used for carrying out standardized conversion on differentiated event formats and converting the differentiated event formats into a unified event information structure; the event analysis is used for analyzing standardized event information, calling the event rule through an event identification model to judge event characteristics, determining whether to trigger an event or not and generating a real-time result; the event distribution sends the real-time results to the event user.
Preferably, the controlling the contact channel of the user with the mobile user comprises: and performing disturbance filtering according to the historical behaviors, and controlling the user to select a contact channel which has not been contacted with the mobile user.
In another aspect of the embodiments of the present invention, a real-time service recommendation system is also provided, including:
the event monitoring module is used for acquiring and monitoring network behavior information of the mobile user in real time and acquiring event source data in the network behavior information;
the quasi-real-time processing engine is used for analyzing and processing the event source data in real time according to the event rule and sending a real-time result to an event user;
and the channel contact control module is used for taking the real-time result as a data interface, performing contact judgment by combining the historical behaviors of the mobile user, controlling the contact channel between the user and the mobile user according to the event rule and performing real-time service recommendation activities.
Preferably, the system further comprises: and the rule management module is used for extracting event elements according to the service scene and pre-configuring the event rules.
Preferably, the system further comprises: and the feedback optimization module is used for automatically promoting the next recommendation or contact action according to the feedback information of the mobile user and the event rule and carrying out interactive service recommendation and service activities for many times.
Preferably, the quasi real-time processing engine comprises an event standardization module, an event analysis module and an event distribution module; the event standardization module is used for carrying out standardization conversion on differentiated event formats and converting the differentiated event formats into a unified event information structure; the event analysis module is used for analyzing the standardized event information, calling the event rule through an event identification model to judge the event characteristics, determining whether to trigger an event or not and generating a real-time result; and the event distribution module is used for sending the real-time result to an event user.
Preferably, the channel contact control module further comprises: and the disturbance-free module is used for carrying out disturbance filtering according to the historical behaviors and controlling the user to select a contact channel which is not contacted with the mobile user.
The real-time service recommendation method and system provided by the embodiment of the invention are expected to comprehensively collect all network behavior information of the user, perform more accurate analysis on the user by utilizing big data of the mobile network, have higher fitness between an analysis result and actual preference of the user and the like, and can perform more accurate recommendation. Meanwhile, the event processing mechanism of the embodiment of the invention effectively changes the close-coupled relation between the real-time event source and the event user, greatly improves the applicability of the event-triggered real-time recommendation service scene, enhances the timeliness of recommendation, widens the adaptability and can greatly improve the user experience.
Drawings
FIG. 1 is a flow chart illustrating a method for real-time service recommendation according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a flow chart of a system for performing real-time service recommendation according to a preferred embodiment of the present invention;
FIG. 3 is a diagram illustrating an exemplary scenario of capturing an event triggering a user surfing behavior;
fig. 4 is a schematic block diagram of a real-time service recommendation system according to another embodiment of the present invention.
Detailed Description
The existing mobile service recommendation scheme has limited channels contacted with users, mainly depends on sending short messages and making calls to inform the users of the latest service activities, generally only depends on the short messages and call bill records to analyze the user behaviors, and because the information quantity borne by the short messages and the call bills is limited, the recommendation scene formed is too simple and fixed, the practical effect is not good, and the problems of channel resource waste, user disturbance and the like are easily caused. If the analysis is performed based on data of a longer time period, the rule establishment and feedback period is too long, and the timeliness and the flexibility are poor, so that quick and efficient real-time recommendation is difficult to realize.
In the embodiment of the invention, all user behavior data are collected and monitored based on a big data environment, the user use scene is analyzed and restored in real time, the user requirements are deeply insights, and a quasi-real-time recommendation triggering and management mechanism can be provided. As shown in fig. 1, in an embodiment of the present invention, a real-time service recommendation method includes the following steps:
s1: acquiring and monitoring network behavior information of a mobile user in real time, and acquiring event source data in the network behavior information;
s2: analyzing the event source data in real time according to event rules, and sending real-time results to event users;
s3: and taking the real-time result as a data interface, and controlling a contact channel between the user and the mobile user according to the event rule by combining the historical behavior of the mobile user to perform real-time service recommendation activities.
In a preferred embodiment of the present invention, the rules and conditions describing the event trigger generation are recorded and managed by using an event rule base, so preferably, the method further comprises: s0: and extracting event elements according to the service scene, and pre-configuring the event rules. Wherein the event is defined as: an event is a particular state of the environment, and when the environment changes to this state, a corresponding event is triggered. Generally, it can also be understood that a change in a certain state of a real-world object or an occurrence of a certain condition, and there are three common conditions:
1) external changes that can trigger the start of a certain process (e.g.: service transaction application submission);
2) things change to a certain state (e.g.: the number of user arrivals exceeds a threshold);
3) the occurrence of a particular event (such as: the user moves to a particular area).
In the embodiment of the invention, the event source data is original data, records all information possibly related to the event in a big data environment, needs to be further analyzed and processed by using an event rule, identifies whether the user triggers the event and which events are triggered, and distributes the identified specific event and user condition to the user of the specific event. The event rule mainly refers to conditions for generating events, the event generation conditions are configured through the event rule management module, and the configured event rule further comprises the relationship among the conditions and subsequent actions (including but not limited to user limitation, contact channel control, service recommendation, feedback optimization and the like); the event rules are divided into real-time trigger rules and cumulative trigger rules.
In the embodiment of the present invention, the event rule operator generally includes:
| operational character | Means of |
| OR | Or operation, function having one or more possible trigger events |
| XOR | XOR operation, one and only one rule being satisfied at a time |
| AND | With operation, all rules are satisfied simultaneously to trigger an event |
An exemplary event rule base is shown in table 1:
| event rule base | 1002 |
| Monitoring source | User GPRS detailed list |
| Movement of | Filtering data, capturing |
| Condition | |
| 1 | Capture "taobao.com" |
| Conditional join relationships | AND |
| Condition 2 | Capture "iphone 6" |
| Triggering event | Medium-high end switch |
| Follow-up action | Marketing mobile contract machine |
TABLE 1
In a preferred embodiment of the present invention, the above method is implemented using a flow scheme as shown in fig. 2. Specifically, during the initialization (i.e., step S0), the event rules (such as trigger conditions and the like) are pre-configured by the rule management module: the event is defined through analysis processes such as hierarchical discussion, index screening and the like, information such as an event source, a trigger event period, event service meanings, event users, an event transmission mode and the like is determined, and scene presetting is completed. The rule management module completes the configuration of event rules including condition parameters in an event rule base, a human-computer interface of the management module can be used for business personnel and technical personnel to formulate a recommended marketing event rule, and after the marketing rule formulation is completed, the management module can be used for inquiring and analyzing; the event marketing rule can also be managed by automatically generating rule conditions and parameters through an analysis tool and application software and importing the generated rule conditions and parameters into a system in a data interface mode.
In step S1, the event source data acquired after the filtering and matching of the collected user network behavior information includes the following aspects: network location information (location area, cell, geographical grid of locations, etc.), user behavior information (user number, terminal device, network access identification, source IP, target IP, access URL, etc.), user access content (type identification, TOP5 keywords, submit keywords, etc.). The embodiment of the invention does not only rely on the background system of the operator for analyzing the mobile user behavior, but also collects short messages and ticket records, comprehensively collects all network behavior information of the user by monitoring modules respectively arranged in different software and hardware systems such as access equipment, a base station, intermediate exchange equipment, a content provider, an application system and the like, and more accurately analyzes the user by utilizing big data of a mobile network. Preferably, the monitoring is acquired in real time through a monitoring agent, and the acquisition mode adopts data light splitting and mirroring technologies. The obtaining mode adopts a data packet recombination and reduction technology, IP transmission data packets are recombined, five-layer protocols are analyzed and stripped to transmit control information, and user application layer access content data are obtained.
In step S2, the process of parsing the event source data includes three steps: event standardization, event parsing, and event distribution. In order to deal with a big data environment and guarantee the recommended timeliness, the data flow analysis processing is preferably carried out on the event processing in a cloud platform mode, the processing capacity of a single PC server per second reaches 6-10 GB, various real-time perception rules are supported, the user access content is acquired in real time, and the user preference information is enhanced. The behavior information of the user is processed at a high speed through a big data processing technology, the use scene of the user can be timely, efficiently and accurately restored, a foundation is laid for deep insight of the user, and therefore a real-time trigger management mechanism can be provided.
Further, event standardization refers to that an event source system carries out standardized conversion on differentiated event formats and converts the differentiated event formats into a unified event information structure for subsequent processing. The standardized processing comprises four links of denoising, cleaning, conversion and verification. The data in the event source is original data, and in order to obtain readable and understandable effective information, data signaling correlation preprocessing is firstly performed, and meanwhile, identification, enhancement and standardized conversion of multiple key data such as user information, terminal information, network information, position information, service behavior information and content information are performed. Preferably, the event monitoring module finishes the acquisition of the event source to acquire information for filtering and matching, and realizes the perception of the data of the event source. The invention can also access data acquisition and unified configuration management of filtering rules in real time, then filters the real-time access data according to the rules, realizes cleaning and noise reduction of the data, and finally classifies and summarizes the real-time access data.
In one example of the present invention, the standardized information format for the mobile phone to access the internet event is:
the event analysis is to analyze the standardized event information and call an event rule through an event identification model to judge the event characteristics. The event analysis receives an external rule interface to realize the step-by-step analysis of rule-event-object: the perception rules are resolved into event objects, and the event objects are resolved into perception objects, namely telecommunication users. In addition, the recognition model library of the event is one of the cores of event analysis, a professional team is required to continuously track the definition and update the transition of the business event, and multiple times of cyclic analysis of simulation, collection, analysis, induction and verification are required to realize the sorting of the recognition model of the business event. The multidimensional event tracking and identifying engine monitors through the combination of a plurality of trigger mechanisms, can accurately and efficiently identify user scenes, and the specific trigger mechanisms include but are not limited to:
content-specific perceptual triggering mechanisms: real-time triggering based on content resources;
location-specific triggering mechanism: real-time triggering based on location switching;
specific network access triggering mechanism: real-time triggering based on a specific network access mode;
the specific network busy and idle index monitoring triggering mechanism comprises the following steps: triggering in real time based on busy and idle indexes of a base station, a hot spot and a link;
user-specific triggering mechanism: the internet surfing action of the designated user is triggered;
flow direction, flow rate and other trigger mechanisms: triggering indexes such as flow direction, peak flow velocity and the like in real time;
online time, residence time, service access usage time trigger mechanism, etc.: and the real-time triggering of the dimensions such as the online time, the regional stay time, the service access time and the like of the user is supported.
Preferably, the multidimensional event tracking and identifying engine identifies the analyzed event by using at least one sensing module, the sensing module is arranged in a distributed deployed system, the sensing acquired information is filtered and matched with the event monitoring rule, and the rule event is triggered when the rule condition of some event is met.
Event distribution is the sending of event instances generated by event parsing to the users of the events. And after the event processing center analyzes the user internet behavior information, a real-time user label view containing the user service semantics can be generated. The real-time user tag view is a group of data group sequences containing a user mobile phone number, a service tag, a timestamp and a geographic position, and the sequence attributes of the data group sequences can include terminal type numbers, access applications and other attributes. The real-time user tag view effectively expresses the business activity trend and tendency of the user within a certain time period, and can be used as a data interface of a real-time marketing activity.
In step S3, after the event is triggered, the event distribution execution link is entered, and the user of the event contacts the user through channel control, so as to complete the specific service recommendation activity. The user of the event includes various external systems including a marketing management platform, different from the prior art which only adopts short messages and telephones for recommendation, the user of the event in the embodiment of the invention includes various external systems with different identities, respective communication platforms are utilized to call an event interface (a real-time result of the event analysis processing, such as a real-time user label view, is used as a data interface) to acquire the user and the event which need to be contacted, and then the user can be contacted by means of channels such as short messages, outbound calls, WAP Push, clients, Internet and the like. For example, the business marketing event interface supported by the current marketing management platform includes:
the WLAN potential user marketing realizes the analysis and identification of the WLAN potential user and provides support for the WLAN service marketing;
the 4G terminal marketing is matched with the customized terminal user machine changing time through analyzing the preference of the user terminal, and provides support for accurate marketing of the user intelligent terminal and flow promotion of the user;
the package matching marketing realizes the adaptation of the user flow packages by constructing a user and flow package matching analysis model, provides support for the upgrade of the flow packages and the superposition of the flow packages of the user, and improves the value of the flow packages and the satisfaction degree of the user on the flow packages;
flow product and content marketing is based on analysis of content preferences of users, the content preferences of the users and products suitable for recommendation of the content preferences are found, and support is provided for accurate marketing.
Since many marketing events are in contact with the user, the strategic control and enforcement of channel contact is accomplished by the channel contact control module in the preferred embodiment of the invention. The module controls what channels are contacted by the user at what time periods and can do multiple times of marketing and contact based on the user's feedback, the event rules including channel contact rules. In the preferred embodiment of the present invention, channel control is performed in combination with the historical behaviors of the mobile user, for example, to avoid disturbing the user, the historical behaviors of the user contacting the channel are obtained to perform contact judgment, and the channel which has not been contacted with the user is adopted to perform service recommendation: for example, for a user who makes many calls for a long distance, a service with a long-distance preference is suitable for the user, but the user does not obtain the response of the user before sending the short message recommendation, and at the moment, the user is contacted with the service through a channel control selection telephone recommendation mode. Further, the event processing center triggers a real-time marketing event, generates a real-time user label and pushes the real-time user label to the marketing management platform, if the user has special channel preference setting, the user is additionally provided with a channel contact rule and a row channel matching rule of the user, and the quasi-real-time analysis result is pushed with data through a matched execution channel.
The technical solution of the present invention is further explained in an exemplary application scenario, in which a "medium-high terminal switching" event is triggered by capturing a user internet keyword, so as to perform marketing recommendation of a contract machine. The event rule base for the example scenario is shown in the table of the 1002 base above and will not be repeated here. An example trace capture process for user surfing behavior is shown in FIG. 3: capturing a URL address string from the internet surfing behavior of a user, and identifying and analyzing two keywords connected with the business semantics of the medium-high terminal switch machine by using a word segmentation technology, namely "taobao.com" and "iphone 6". And after the event processing center analyzes the user internet behavior information, a real-time user label view containing the user service semantics can be generated. The real-time user label view is used for supporting event users including a marketing management platform to use, has certain effective user characteristic preference, can reflect purchasing willingness and tendency of the users in a certain time period, is used for the marketing management platform and other service platforms, and is used for recommending the type of the mobile phone by contacting the users at proper time by high-end and medium-end machine marketing personnel in the scene.
Meanwhile, according to the feedback information of the user, the next marketing or contact action can be automatically promoted according to the event rule, and multiple interactive marketing and service activities are carried out. The event rules followed by the complex feedback optimization processes can be set by the marketing management platform and the event processing center at the beginning, and after the configuration is completed, the system can automatically acquire, analyze and process the behavior information of the user, generate a real-time user information view and distribute the user information view to the marketing management platform for use.
Business personnel can directly use the system, set business event definition and deploy business activities, and when new marketing requirements appear, the corresponding new events can be flexibly configured and rapidly deployed. After the event rules can be flexibly set, the batch event marketing can be set as a quasi-real-time event marketing activity, and the rule combination can also be formed in order aiming at complex event factors.
As shown in fig. 4, in another embodiment of the present invention, a real-time service recommendation system 1 is further provided, which includes:
the event monitoring module 101 is configured to collect and monitor network behavior information of a mobile user in real time, and acquire event source data in the network behavior information;
the quasi-real-time processing engine 102 is used for analyzing and processing the event source data in real time according to the event rule and sending a real-time result to an event user;
and the channel contact control module 103 is configured to use the real-time result as a data interface, perform contact judgment by combining with the historical behavior of the mobile user, control a contact channel between the user and the mobile user according to the event rule, and perform a real-time service recommendation activity.
The system of the preferred embodiment of the present invention may further comprise: the rule management module is used for extracting event elements according to a service scene and pre-configuring the event rules; and the feedback optimization module is used for automatically promoting the next recommendation or contact action according to the feedback information of the mobile user and the event rule and carrying out interactive service recommendation and service activities for many times. Event rules followed by the feedback optimization process can also be pre-configured in the rules, and after the configuration is completed, the system can automatically acquire, analyze and process behavior information of the user, generate a real-time user information view and distribute the user information view to a marketing management platform and the like for use.
Further, the rule management module in the system comprises: the configuration module is used for configuring event rules comprising conditions and parameters through a human-computer interface; and/or the data mining module is used for automatically generating conditions and parameters of the event rule through the analysis tool and the application software, and importing and managing the conditions and the parameters in a data interface mode.
The quasi real-time processing engine comprises an event standardization module, an event analysis module and an event distribution module; the event standardization module is used for carrying out standardization conversion on differentiated event formats and converting the differentiated event formats into a unified event information structure; the event analysis module is used for analyzing the standardized event information, calling the event rule through an event identification model to judge the event characteristics, determining whether to trigger an event or not and generating a real-time result; and the event distribution module is used for sending the real-time result to an event user.
To avoid being overly intrusive to the user, the channel contact control module further comprises: and the disturbance-free module is used for carrying out disturbance filtering according to the historical behaviors and controlling the user to select a contact channel which is not contacted with the mobile user. In addition, in order to deal with a big data environment, the system of the embodiment of the invention also processes the data of the user in the mobile communication pipeline at high speed through a big data processing technology. Preferably, the near real-time processing engine further comprises: and the big data processing module is used for analyzing and processing the data stream in a cloud platform mode. The big data processing can support various real-time perception rules, acquire user access content in real time and enhance user preference information.
The real-time service recommendation method and system provided by the embodiment of the invention are expected to comprehensively collect all network behavior information of the user, perform more accurate analysis on the user by utilizing big data of the mobile network, have higher fitness between an analysis result and actual preference of the user and perform more accurate recommendation. Meanwhile, the event processing mechanism of the embodiment of the invention effectively changes the close-coupled relation between the real-time event source and the event user, greatly improves the applicability of the event-triggered real-time recommendation service scene, enhances the timeliness of recommendation, widens the adaptability and can greatly improve the user experience.
The marketing scene planning configuration work and the marketing event capturing processing execution are decoupled through the quasi-real-time event processing engine, and meanwhile, a real-time information view is generated through analysis of the event processing engine to support event users including a marketing management platform to use. The event processing real-time collected user behavior information is processed, a real-time information view is generated after calculation and processing, namely the user characteristic preference with certain effectiveness can reflect the purchase intention and trend of the user in a certain time period, and the formatted real-time information view can be used for a marketing management platform and other business platforms.
In addition, the invention provides a real-time service recommendation scheme based on event processing in a big data environment, which processes data of a user in a mobile communication pipeline at high speed through a big data processing technology, acquires user internet behavior information in real time, transmits the data to an event processing center for analysis after data acquisition is carried out by an event monitoring module, effectively solves the tight coupling relationship between a real-time event source and an event user through an event processing mechanism, and simultaneously, defines and configures a marketing event independently, so that the real-time service recommendation scheme can be flexibly used by service personnel, and the applicability of a real-time marketing service scene triggered by the event is greatly improved.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.
Claims (10)
1. A method for real-time service recommendation, the method comprising:
acquiring and monitoring network behavior information of a mobile user in real time, and acquiring event source data in the network behavior information;
analyzing the event source data in real time according to event rules to identify whether the mobile subscriber triggers an event and the triggered event, and sending a real-time result to a user of the event, wherein the real-time result is a real-time subscriber tag graph containing the service semantics of the mobile subscriber; the event rule comprises conditions for generating events, relations among the conditions and subsequent actions; the follow-up actions comprise user limitation, contact channel control, service recommendation and feedback optimization;
and acquiring the mobile user and the event needing to be contacted by taking the real-time result as a data interface, judging the contact by combining the historical behavior of the mobile user, controlling a contact channel between the user and the mobile user according to the event rule, and performing real-time service recommendation activity.
2. The method of claim 1, further comprising:
and extracting event elements according to the service scene, and pre-configuring the event rules.
3. The method of claim 1, wherein after performing the real-time business recommendation activity, the method further comprises:
and according to the feedback information of the mobile user and the event rule, promoting the next recommendation or contact action, and performing interactive service recommendation and service activities for multiple times.
4. The method of claim 1, wherein the parsing process comprises event normalization, event parsing, and event distribution; wherein the content of the first and second substances,
the event standardization carries out standardized conversion on differentiated event formats and converts the differentiated event formats into a unified event information structure;
the event analysis is used for analyzing standardized event information, calling the event rule through an event identification model to judge event characteristics, determining whether to trigger an event or not and generating a real-time result;
the event distribution sends the real-time results to the event user.
5. The method of claim 1, wherein said controlling the user's contact channel with the mobile user comprises:
and performing disturbance filtering according to the historical behaviors, and controlling the user to select a contact channel which has not been contacted with the mobile user.
6. A real-time service recommendation system, the system comprising:
the event monitoring module is used for acquiring and monitoring network behavior information of the mobile user in real time and acquiring event source data in the network behavior information;
the quasi-real-time processing engine is used for analyzing and processing the event source data in real time according to event rules so as to identify whether the mobile subscriber triggers an event and the triggered event, and sending a real-time result to a user of the event, wherein the real-time result is a real-time subscriber label graph containing the service semantics of the mobile subscriber; the event rule comprises conditions for generating events, relations among the conditions and subsequent actions; the follow-up actions comprise user limitation, contact channel control, service recommendation and feedback optimization;
and the channel contact control module is used for acquiring the mobile user and the event which need to be contacted by taking the real-time result as a data interface, performing contact judgment by combining the historical behavior of the mobile user, controlling the contact channel between the user and the mobile user according to the event rule and performing real-time service recommendation activity.
7. The system of claim 6, further comprising:
and the rule management module is used for extracting event elements according to the service scene and pre-configuring the event rules.
8. The system of claim 6, further comprising:
and the feedback optimization module is used for automatically promoting the next recommendation or contact action according to the feedback information of the mobile user and the event rule and carrying out interactive service recommendation and service activities for many times.
9. The system of claim 6, wherein the near real-time processing engine comprises an event standardization module, an event parsing module, and an event distribution module; wherein the content of the first and second substances,
the event standardization module is used for carrying out standardization conversion on the differentiated event formats and converting the differentiated event formats into a unified event information structure;
the event analysis module is used for analyzing the standardized event information, calling the event rule through an event identification model to judge the event characteristics, determining whether to trigger an event or not and generating a real-time result;
and the event distribution module is used for sending the real-time result to an event user.
10. The system of claim 6, wherein the channel contact control module further comprises:
and the disturbance-free module is used for carrying out disturbance filtering according to the historical behaviors and controlling the user to select a contact channel which is not contacted with the mobile user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510459098.1A CN107040863B (en) | 2015-07-30 | 2015-07-30 | Real-time service recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510459098.1A CN107040863B (en) | 2015-07-30 | 2015-07-30 | Real-time service recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107040863A CN107040863A (en) | 2017-08-11 |
| CN107040863B true CN107040863B (en) | 2021-01-15 |
Family
ID=59532900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510459098.1A Active CN107040863B (en) | 2015-07-30 | 2015-07-30 | Real-time service recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107040863B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688987A (en) * | 2017-08-31 | 2018-02-13 | 平安科技（深圳）有限公司 | Electronic installation, insurance recommendation method and computer-readable recording medium |
| US11245786B2 (en) * | 2017-09-18 | 2022-02-08 | Huawei Technologies Co., Ltd. | Recommendation method and terminal |
| CN107833111A (en) * | 2017-11-28 | 2018-03-23 | 中国银行股份有限公司 | A kind of Products Show method, apparatus and system |
| CN108460629A (en) * | 2018-02-10 | 2018-08-28 | 深圳壹账通智能科技有限公司 | User, which markets, recommends method, apparatus, terminal device and storage medium |
| CN108520448B (en) * | 2018-03-07 | 2022-05-17 | 创新先进技术有限公司 | Event management method and device |
| CN108681925A (en) * | 2018-05-21 | 2018-10-19 | 中国联合网络通信有限公司重庆市分公司 | A kind of scenario reduction method based on mobile Internet |
| CN109379330A (en) * | 2018-09-07 | 2019-02-22 | 阿里巴巴集团控股有限公司 | The implementation method and device that content is launched |
| CN110247972A (en) * | 2019-06-17 | 2019-09-17 | 杭州数金在线云科技有限公司 | A kind of transaction processing system |
| CN110599243B (en) * | 2019-09-03 | 2023-04-18 | 浩鲸云计算科技股份有限公司 | Customer-oriented journey marketing method and system |
| CN110688581A (en) * | 2019-10-30 | 2020-01-14 | 南京领行科技股份有限公司 | Information real-time pushing method and device, computing equipment and medium |
| CN110809282B (en) * | 2019-11-13 | 2023-06-09 | 山东易路方舟信息科技有限公司 | Traffic information identification and feedback method, system, equipment and storage medium |
| CN111127077A (en) * | 2019-11-29 | 2020-05-08 | 中国建设银行股份有限公司 | Recommendation method and device based on stream computing |
| CN111489178A (en) * | 2020-04-16 | 2020-08-04 | 北京思特奇信息技术股份有限公司 | Method and system for marketing support through event detection |
| CN111818154B (en) * | 2020-07-06 | 2022-11-18 | 桦蓥(上海)信息科技有限责任公司 | Service pushing system and method based on network layer message analysis |
| CN112132612A (en) * | 2020-09-18 | 2020-12-25 | 贝壳技术有限公司 | Task processing method and device, electronic equipment and storage medium |
| CN112435047A (en) * | 2020-10-30 | 2021-03-02 | 四川新网银行股份有限公司 | Marketing outbound data recommendation method based on buried point data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102387461A (en) * | 2011-10-18 | 2012-03-21 | 北京佳信汇通科技有限公司 | Mobile data service recommendation method, apparatus and system thereof |
| CN102547554A (en) * | 2011-12-28 | 2012-07-04 | 华中科技大学 | Mobile service recommendation method based on mobile user behavior |
| CN103501340A (en) * | 2013-10-08 | 2014-01-08 | 北京金山网络科技有限公司 | Message push method, message push device and client |
| CN103793465A (en) * | 2013-12-20 | 2014-05-14 | 武汉理工大学 | Cloud computing based real-time mass user behavior analyzing method and system |
| CN103944800A (en) * | 2013-01-21 | 2014-07-23 | 电信科学技术研究院 | Information interaction method and equipment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120316902A1 (en) * | 2011-05-17 | 2012-12-13 | Amit Kumar | User interface for real time view of web site activity |
| US20130018982A1 (en) * | 2011-07-12 | 2013-01-17 | Salesforce.Com, Inc. | Method and system for providing recommended information from a customer relationship management system |
-
2015
- 2015-07-30 CN CN201510459098.1A patent/CN107040863B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102387461A (en) * | 2011-10-18 | 2012-03-21 | 北京佳信汇通科技有限公司 | Mobile data service recommendation method, apparatus and system thereof |
| CN102547554A (en) * | 2011-12-28 | 2012-07-04 | 华中科技大学 | Mobile service recommendation method based on mobile user behavior |
| CN103944800A (en) * | 2013-01-21 | 2014-07-23 | 电信科学技术研究院 | Information interaction method and equipment |
| CN103501340A (en) * | 2013-10-08 | 2014-01-08 | 北京金山网络科技有限公司 | Message push method, message push device and client |
| CN103793465A (en) * | 2013-12-20 | 2014-05-14 | 武汉理工大学 | Cloud computing based real-time mass user behavior analyzing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107040863A (en) | 2017-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107040863B (en) | Real-time service recommendation method and system | |
| US20080056144A1 (en) | System and method for analyzing and tracking communications network operations | |
| CN107086922B (en) | User behavior identification method and device | |
| CN110519177A (en) | A kind of network flow identification method and relevant device | |
| CN107645562A (en) | Data transmission processing method, device, equipment and system | |
| CN103200262A (en) | Method, device and system for advertisement scheduling based on mobile network | |
| CN101373533A (en) | Real time accurate marketing apparatus based on mobile communication signaling gateway and data processing method | |
| CN101883133A (en) | Accurate influence marketing system based on signalling analysis and method thereof | |
| CN111882367B (en) | Method for monitoring and tracking online advertisements through analysis of user surfing behavior | |
| CN106911523A (en) | The method and system that mobile interchange network users are positioned by LTE indulging in the internet | |
| CN113412607B (en) | Content pushing method and device, mobile terminal and storage medium | |
| US20110106579A1 (en) | System and Method of Management and Reduction of Subscriber Churn in Telecommunications Networks | |
| CN103491110A (en) | Information system based on mobile dynamic data engine | |
| CN106559498A (en) | Air control data collection platform and its collection method | |
| CN105791349A (en) | Data analysis method, device and system, terminal and server | |
| CN102905236B (en) | A kind of junk short message monitoring method, Apparatus and system | |
| CN101420336A (en) | Method for recognizing network telephone flow quantity in network and system thereof | |
| US10439919B2 (en) | Real time event monitoring and analysis system | |
| CN109246282B (en) | Call information storage method and device, storage medium and electronic device | |
| CN108989116B (en) | Big data collection and analysis system and method | |
| CN101827328A (en) | Device and method for monitoring short-message | |
| CN109308298B (en) | Engineering project integrated management system | |
| CN111882368B (en) | On-line advertisement DPI encryption buried point and transparent transmission tracking method | |
| CN104734898B (en) | A kind of business recommended method and system based on social network analysis | |
| CN102591860A (en) | Cloud webpage non-specific object real-time communication and browse behavioral analysis system |
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 |