Search

CN-121985031-A - Message pushing method and system based on mobile internet user behavior analysis

CN121985031ACN 121985031 ACN121985031 ACN 121985031ACN-121985031-A

Abstract

The invention relates to a message pushing method and a system based on mobile internet user behavior analysis. The SDK is embedded in the terminal equipment, when a user enters a specific area, the SDK collects scene information and user behavior information, the server judges whether to trigger a preset geofence or not based on the geographic position information of the terminal equipment and judges whether the user enters the specific area, when the user enters the specific area, the SDK screens and forms a message set to be pushed based on the scene information and the user behavior information, and the highest-scoring message in the message set to be pushed is sent to the terminal equipment to form visual processing through the SDK.

Inventors

  • CHEN QIONG
  • LIU CHANZI

Assignees

  • 东莞理工学院

Dates

Publication Date
20260505
Application Date
20251215

Claims (10)

  1. 1. A message pushing system based on mobile internet user behavior analysis, the system comprising: an SDK (100) embedded in the terminal device (300), the SDK (100) collecting scene information and user behavior information when a user enters a specific area, and collecting a score of the user on a message; The server (200) judges whether to trigger a preset geofence or not based on the geographic position information of the terminal equipment (300) and judges whether a user enters a specific area, screens and forms a message set to be pushed based on the scene information and the user behavior information when the user enters the specific area, and sends the highest-scoring message in the message set to be pushed to the terminal equipment (300) so as to form visual processing through the SDK (100).
  2. 2. The system according to claim 1, wherein the step of the server (200) determining whether the user enters a specific area comprises: comparing the geographic position of the terminal equipment (300) with a preset geofence, and calculating the spherical straight line distance between the geographic position of the terminal equipment (300) and the center point of the geofence based on a HAVERSINE spherical formula; Comparing the spherical straight line distance with a preset fence radius, and judging the behavior state of a user aiming at a specific area based on a three-level state judgment rule; the behavioral state includes entering, being in, and exiting the particular region.
  3. 3. The system according to claim 1 or 2, wherein the three-level status determination rule by which the server (200) determines whether the user enters a specific area comprises: Setting a step threshold delta d, setting the radius of the fence as R, and setting the spherical straight line distance between the geographic position and the center point of the geographic fence as d; When d is less than or equal to R-delta d, determining that the current geographic position is in a core area inside the geofence, and triggering service in a specific area; Entering a fence area, wherein when R-delta d is less than or equal to d and less than or equal to R+delta d, the current position is judged to be in a geofence boundary transition area, the server (200) starts continuous positioning sampling, if the continuous sampling results meet d and less than or equal to R+delta d, the entering of the fence area is confirmed, and an event is triggered to enter the area; Leaving the fenced area when d > R+Δd, determining that the current location is outside the geofence, and triggering an area leave event if the previous determination was in or entering the fenced area.
  4. 4. A system according to any one of claims 1-3, wherein the step of the server (200) screening and composing a set of messages to be pushed based on the context information and user behavior information comprises: When the terminal equipment (300) enters a preset geofence area, the SDK (100) collects multi-dimensional information of the user and the terminal equipment (300) and sends the multi-dimensional information to the server (200); the server (200) calculates the scores of the messages based on the user portrait, the scene information, the user behavior information and the filtering algorithm, and selects the information interested by the user based on a preset threshold value to form a message set to be pushed interested by the user.
  5. 5. The system according to any one of claims 1-4, wherein the step of the SDK (100) collecting user scores for messages comprises: Giving different weights to different behaviors of the user, and carrying out weighted calculation on the weights based on the user behavior information; Wherein, the score=message opening behavior weight×a+message browsing duration weight×b+user active interaction behavior weight×c, and A, B and C are weight coefficients.
  6. 6. The system according to any one of claims 1 to 5, wherein the server (200) further performs overall evaluation of the recommended effect of the information from a plurality of dimensions; wherein the plurality of dimensions of the overall evaluation include a click rate dimension, an average scoring dimension of messages, and an interaction rate dimension; The higher the click rate, average score of messages and interaction rate, the better the recommendation effect.
  7. 7. The system according to any one of claims 1 to 6, wherein the communication connection manner between the SDK (100) and the server (200) includes: -the SDK (100) generates a unique identifier CID identifying the terminal device (300); the SDK (100) establishes a bidirectional channel with the server (200) for sending the scene information and the user behavior information collected by the SDK and receiving the pushed personalized message.
  8. 8. A message pushing method based on mobile internet user behavior analysis, the method comprising: When a user enters a specific area, the SDK (100) collects scene information and user behavior information and sends the scene information and the user behavior information to the server (200); The server (200) judges whether to trigger a preset geofence or not based on the geographic position information of the terminal equipment (300) and judges whether a user enters a specific area, when the user enters the specific area, the server (200) screens and forms a message set to be pushed based on the scene information and the user behavior information, and the server (200) sends the message with the highest score in the message set to be pushed to the terminal equipment (300) so as to form visual processing through the SDK (100).
  9. 9. The method of claim 8, wherein the step of determining whether the user enters a particular area comprises: comparing the geographic position of the terminal equipment (300) with a preset geofence, and calculating the spherical straight line distance between the geographic position of the terminal equipment (300) and the center point of the geofence based on a HAVERSINE spherical formula; Comparing the spherical straight line distance with a preset fence radius, and judging the behavior state of a user aiming at a specific area based on a three-level state judgment rule; the behavioral state includes entering, being in, and exiting the particular region.
  10. 10. The method according to claim 8 or 9, wherein the three-level state decision rule for deciding whether the user enters a specific area comprises: Setting a step threshold delta d, setting the radius of the fence as R, and setting the spherical straight line distance between the geographic position and the center point of the geographic fence as d; When d is less than or equal to R-delta d, determining that the current geographic position is in a core area inside the geofence, and triggering service in a specific area; Entering a fence area, namely judging that the current position is in a geofence boundary transition area when R-delta d is less than or equal to R+delta d, starting continuous positioning sampling by a server (200), and if the continuous sampling results all meet d less than or equal to R+delta d, confirming to enter the fence area, and triggering an area entering event; Leaving the fenced area when d > R+Δd, determining that the current location is outside the geofence, and triggering an area leave event if the previous determination was in or entering the fenced area.

Description

Message pushing method and system based on mobile internet user behavior analysis Technical Field The invention relates to the technical field of message pushing, in particular to a message pushing method and system based on mobile internet user behavior analysis. Background At present, mobile internet technology is rapidly developed, the functions of a smart phone are continuously innovated, and the smart phone is completely integrated with the life of people. The cell phone media has become the fifth largest media following newspapers, radio, television, internet. Electronic commerce and travel industry based on mobile internet are also rapidly developing, and online shopping behavior and life style of people are also silently changed. However, the problems of low quality of service, insufficient information, insufficient personalized service, etc. become important factors restricting the development of these industries. Therefore, there is a need for an efficient recommendation method to solve the above-mentioned problems. The location services, also called mobile location services, are the acquisition of location information of a mobile terminal user via a network of a telecommunication mobile operator, such as a GSM network, a CDMA network. The location service integrates mobile communication technology, spatial location technology, geographic information system technology (GIS, geographic Information System), distributed computing technology, embedded database technology, and internet technology, and belongs to a high-tech industry. The business scope of the location service already covers the services of a plurality of industries such as personal location information service, related information inquiry, POI inquiry, path planning, real-time navigation, logistics management, emergency rescue, weather condition inquiry, road condition information inquiry and the like. The location-based service is capable of acquiring geographic location information, such as geographic location coordinates, of a mobile terminal user using software and hardware resources of a telecommunications carrier. With the support of the GIS system platform, the services can bring about corresponding economic growth. The traditional content-based and location-based message recommendation techniques have various drawbacks, such as excessive information load, less useful information, low message pushing accuracy, etc., which results in less than ideal recommendation results. The user mainly relies on signing in, inquiring and other modes to acquire information. The recommendation system based on the scene and the user behavior can make up for the defects of the traditional recommendation system, and simultaneously, the information overload problem faced by the user can be relieved more effectively. Meanwhile, the targeting and pertinence of the pushing service are further improved by combining a collaborative filtering algorithm and a user image, the requirement of the user on personalized information is met, and the message pushing modes of needle fishing and fierce frying in the sea are avoided. Accurate and personalized message pushing is not only beneficial to improving the brand value of information and clients, but also can enhance the user's pleasure. However, invalid information tends to give rise to a strong objection to the user. Thus, in addition to focusing on information distribution speed, information content quality, and user experience of an application, message pushing also requires grasping of appropriate methods and skills. The cloud pushing platform supporting the third party secure access and having good performance and high reliability is constructed, and is a basis for realizing efficient message pushing. In order to ensure the quality of the push service, the performance, reliability and safety problems of the cloud platform need to be comprehensively considered. The Reactor mode is an important idea in server design, and it uses event driving as a core, separates (demultiplex) and schedules (dispatches) service requests of one or more clients, and synchronously and orderly processes multiple requests received simultaneously. CN110415012A discloses a dynamic scene area determining method, device and equipment and a readable storage medium, wherein the method comprises the steps of obtaining position information of a plurality of shops, determining electronic fences corresponding to the shops according to the position information of the shops and preset electronic fence parameters, carrying out fusion processing on the electronic fences corresponding to the shops to obtain a plurality of aggregation areas, obtaining population density data of the aggregation areas, screening the aggregation areas according to the population density data of the aggregation areas, determining the screened aggregation areas as dynamic scene areas, determining terminal equipment in the dynamic scene areas according to the dynamic