CN110312222B - Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint - Google Patents
Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint Download PDFInfo
- Publication number
- CN110312222B CN110312222B CN201910473832.8A CN201910473832A CN110312222B CN 110312222 B CN110312222 B CN 110312222B CN 201910473832 A CN201910473832 A CN 201910473832A CN 110312222 B CN110312222 B CN 110312222B
- Authority
- CN
- China
- Prior art keywords
- user equipment
- information
- friend
- user
- making
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- 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
- H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
- H04W4/21—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel for social networking applications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/35—Services specially adapted for particular environments, situations or purposes for the management of goods or merchandise
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]
- H04L51/046—Interoperability with other network applications or services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/21—Monitoring or handling of messages
- H04L51/222—Monitoring or handling of messages using geographical location information, e.g. messages transmitted or received in proximity of a certain spot or area
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/52—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail for supporting social networking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/58—Message adaptation for wireless communication
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprints, which comprises the following steps: establishing a WiFi fingerprint database and a shop distribution map of a mall; acquiring user information; acquiring a friend recommendation list; obtaining information of making friends with meals; receiving friend-making request information; the method is based on the wifi fingerprint positioning technology and combines the WeChat small program, and finally, the better friend-making experience is obtained.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a shopping mall friend recommendation method based on WiFi (wireless fidelity) position fingerprints.
Background
With the development of the internet and intelligent terminals, network friend making between people gradually changes from a traditional adding mode based on user account searching to a mode of adding friends by WeChat scanning or adding friends by searching nearby people, and the like, so that the network friend making becomes more diversified and intelligent. The way of adding strangers to friends is more popular among people.
In markets, shopping centers and the like, a group of strangers are simultaneously positioned in a closed shopping floor, the duration is long, the mood of users is pleasant, the friend making demand is improved, the dinning willingness of people in the markets and the shopping centers is obvious, and merchants also give out activity advantages such as group spelling and the like.
The problem that how to enable people to make friends more conveniently in shopping malls and shopping centers is worthy of solving.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a shopping mall friend recommendation method based on WiFi (wireless fidelity) position fingerprints.
The purpose of the invention can be achieved by adopting the following technical scheme:
a shopping mall friend recommendation method based on WiFi location fingerprints is characterized by comprising the following steps:
establishing a WiFi fingerprint database and a shop distribution map of a mall;
acquiring user information;
acquiring a friend recommendation list;
obtaining information of making friends with meals;
and accepting friend-making request information.
Preferably, the mall store distribution map includes:
collecting shop distribution information inside a mall and drawing a shop distribution map of the mall;
and classifying the shops inside the shopping mall.
Preferably, the acquiring the user information includes:
obtaining a wechat user id, a wechat user name, a wechat user head portrait, a city where the wechat user is located, a province where the wechat user is located, a country where the wechat user is located, a user real name, a user gender and a user age through wechat applet authorization;
acquiring MAC address information of user equipment through WIFI signal sending equipment;
and the user information automatically selects the public content according to the user requirement.
Preferably, the obtaining of the friend recommendation list includes:
acquiring a real-time positioning position of user equipment according to a wifi fingerprint positioning technology;
generating a list of present user devices;
establishing a user-interest scoring matrix;
and generating the friend recommendation list by combining pearson correlation coefficients according to the equipment list of the users in the presence and the user-interest and hobby scoring matrix.
Preferably, the information for making friends with meals includes:
and generating the information of making friends with meals every preset time within the time period of the meal spots.
Preferably, the generating of the friend-making information at intervals of preset time within the time period of the meal point includes:
acquiring training data of a friend-making machine learning model;
training a friend-making machine learning model;
and traversing the friend recommendation list, and generating the dining friend-making information according to the friend-making machine learning model.
Preferably, the acquiring training data of the friend-making machine learning model includes:
detecting whether the friend making request information contains a preset communication text or not;
if the sending end user equipment and the receiving end user equipment which contain the preset communication text and the position synchronization time of the sending end user equipment and the receiving end user equipment of the friend making request information in a preset time period exceeds a preset threshold value, marking the identity information and the interest and hobby grading information of the sending end user equipment and the receiving end user equipment of the friend making request information as machine learning positive samples;
and if the preset communication text is contained and the position synchronization time of the sending end user equipment and the receiving end user equipment of the friend making request information does not exceed a preset threshold value in a preset time period, marking the identity information and the interest and hobby grading information of the sending end user equipment and the receiving end user equipment of the friend making request information as machine learning negative samples.
Preferably, the generating of the friend-making information at intervals of preset time within the time period of the meal point includes:
and generating the information for making friends with meals according to the friend recommendation list and a similarity algorithm.
Preferably, the generating of the friend-making information at intervals of preset time within the time period of the meal point includes:
and generating the information for making friends with meals according to the friend recommendation list and the group spelling requirement.
Drawings
Fig. 1 is a flow chart of establishing a WiFi fingerprint database and a mall store profile according to an embodiment of the present invention.
Fig. 2 is a flowchart of a method for recommending mall friends based on WiFi location fingerprints in a server according to an embodiment of the present invention.
Fig. 3 is a flowchart of a mall friend recommendation method based on WiFi location fingerprints in a user equipment according to an embodiment of the present invention.
Detailed Description
The following detailed description of the embodiments of the present invention will be made with reference to the accompanying drawings and examples, so that how to implement the present invention by applying technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented.
The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Herein as "exemplary"
Any embodiment described is not necessarily to be construed as preferred or advantageous over other embodiments.
Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present disclosure.
It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present disclosure.
The workflow of a shopping mall friend recommendation method based on WiFi location fingerprints is described in detail as follows:
the WiFi fingerprint database and the mall store distribution map are established, and it should be noted that the establishing process of the WiFi fingerprint database and the mall store distribution map is not sequential. Based on a pre-established WiFi fingerprint database and a mall store distribution map, a positioning client with a WiFi function in a mall can be positioned, as shown in fig. 1, including the following steps:
and step S101, drawing a map. The method specifically comprises the steps of collecting shop distribution information inside a mall, drawing a shop distribution map of the mall, and storing the shop distribution map in a server;
the stores inside the mall may be classified into the following categories: restaurants, men's clothing stores, women's clothing stores, convenience stores, grocery stores, leisure and entertainment stores, beverage stores, luxury decorations, and the like. The restaurants can be divided into sub-categories such as Sichuan-style hot pot restaurants and Guangdong-style snack restaurants.
The stores in the shopping mall are classified, so that the interest and hobby attributes of the user can be distinguished when the behavior track of the user is analyzed.
Step S201: establishing a WiFi fingerprint database, specifically comprising the steps of determining a plurality of fingerprint points in the shop distribution map of the mall drawn in the step S101, then carrying out fingerprint acquisition in the mall according to the selected fingerprint points, forming a fingerprint database and storing the fingerprint database in the server;
the fingerprint collection is to scan MAC address information of a WiFi signal generator published in the mall and a signal intensity value corresponding to the WiFi signal generator at the position of a fingerprint point by using user equipment with a WiFi function, and collect the MAC address information and the signal intensity value;
the User Equipment (UE) is a device capable of connecting to the WIFI signal, and may be, for example, a mobile phone, a Personal Digital Assistant (PDA), or the like.
Example 1
The method for recommending the shopping mall friends based on the WiFi location fingerprints is used in a server. The shopping mall friend recommendation method based on the WiFi position fingerprint comprises the following steps.
Step S301: and scanning the received signal strength of all the user equipment and positioning in real time. The method specifically comprises the following steps: and at a preset time point, the WiFi signal generator scans the received signal intensity of each user equipment in all the user equipment, and sends the obtained signal intensity corresponding to each user equipment and the MAC address information corresponding to each user equipment to the server.
All the user equipment refers to all the user equipment which can scan the signal intensity and the MAC address information through the WiFi signal generator in the shopping mall.
And the server receives and obtains the position of each user equipment in all the user equipment in the shop distribution map of the mall by utilizing WiFi fingerprint positioning in combination with the WiFi fingerprint database according to the signal strength corresponding to each user equipment in all the user equipment and the MAC address information corresponding to each user equipment in all the user equipment. And the server stores the position, the MAC address information and the time point information of the moment in the shop distribution map of the mall corresponding to each user equipment.
Step S401: and detecting and storing the action track of each user equipment in all the user equipment in real time, and establishing a user-interest and hobby scoring matrix. The method specifically comprises the following steps: the server regards the position of the user equipment in the mall store distribution map, the MAC address information of the user equipment, the time point information of the moment and the like obtained in step S301 as a behavior track of the user, based on the criterion that the continuous stay time of the user equipment in a certain store exceeds a preset time length, for example, the position of the user equipment stays in a certain men ' S clothing store within 15 minutes, so as to infer that the user is interested in men ' S clothing, and the score of the user on men ' S clothing may be 1. Further, if the user equipment goes to a men's clothing store every time the user equipment appears in a shopping mall within one month, and the stay time exceeds a preset threshold value, it is inferred that the user is very interested in men's clothing, and the score of the user on men's clothing can be added with 2 to be 3. For example, at a certain meal point time, the user device stays in a certain Sichuan-style hot pot restaurant for 30 minutes, so that the user is inferred to have finished eating, and is slightly interested in the Sichuan-style hot pot restaurant, the score of the user in the Sichuan-style hot pot restaurant can be 1 point, if the user device stays in the Sichuan-style hot pot restaurant for more than 30 times in a month for more than 3 times, so that the user is inferred to be very interested in the Sichuan-style hot pot restaurant, and the score of the user in the Sichuan-style hot pot restaurant can be increased by 2 points to 3 points. Similarly, if the user device is not located in the restaurant for a period of time, such as 30 minutes, at the meal time, it may be inferred that the user has not eaten.
By the method, the user-interest scoring matrix can be established after each user device in all the user devices is traversed.
The user equipment is one of the user equipments.
The time of the meal is preset and can be 11:30-13:30 at noon or 16:30-22:00 at night.
Step S501: and the server receives and stores the identity information sent by the presence user equipment.
Step S601: the server receives a friend recommendation list acquisition request sent by the presence user equipment, and calculates a friend recommendation list, which specifically comprises the following steps:
and adding the user equipment which is in the current time position in the shopping mall and authorized by the WeChat friend-making applet in the current time present user equipment list to obtain the current time present user equipment list.
And calculating pearson correlation coefficients of the present user equipment and all other user equipment in the present user equipment list at the current moment according to the user-interest and hobby scoring matrix, and arranging the present user equipment list at the current moment in a descending manner of the correlation coefficients from high to low to obtain a friend recommendation list.
The information corresponding to each user equipment in the friend recommendation list may include identity information corresponding to the user equipment, location information corresponding to the user equipment, and interest score information corresponding to the user equipment.
The present user equipment is one user equipment in the present user equipment list at the current moment.
The all user devices are different from the currently present user device list in that each user device in the currently present user device list requires presence inside the mall at the current time location and passes authorization of a WeChat friend making applet, while the all user devices do not have this limitation.
The authorization WeChat friend-making applet proves that the holder of the user equipment has friend recommendation requirements, friend recommendation does not interfere with the user equipment, and the user equipment which is not in the mall at the moment can be excluded from a part of user equipment which is currently in the mall, so that the rapidity of instant friend-making is improved.
All the user equipment is not limited, data are collected after the user authorizes the WeChat friend-making small program, time points are not limited, the method is favorable for collecting preference attributes of the user equipment in a market more widely, and data support is provided for the user-interest preference scoring matrix subsequently.
This distinction can greatly improve the user experience.
Step S701: and the server sends the friend recommendation list to the presence user equipment.
Step S801: the server receives friend making request information sent by first presence user equipment and sends the friend making request information to second presence user equipment; the server stores friend making information, wherein the friend making information is in a form of a triplet and comprises identity information of the first presence user equipment, the friend making request information and identity information of the second presence user equipment.
The friend making request information includes: the identity information of the first presence user equipment, the identity information of the second presence user equipment and the communication information.
The first presence user equipment is one user equipment in the list of user equipment present at the present moment.
Said second presence user equipment is a user equipment in said list of currently present user equipment other than said first presence user equipment.
Step S901: recording and storing training data of the friend-making machine learning model, specifically comprising: extracting communication information in friend making request information in the friend making information according to the friend making information obtained in the step S801, detecting whether communication texts such as 'friend making is good', 'noodle bar' and the like are contained in the communication information, if yes, obtaining position information of a first user equipment in the friend making information and a second user equipment in the friend making information according to the wifi fingerprint positioning method in the step S301, detecting whether the position information of the first user equipment and the second user equipment in the friend making information is synchronous within one month, and judging that the friend making is successful if the time exceeds 30 minutes, and if yes, judging that the friend making is successful: taking the identity information and the interest scoring information of the first user equipment in the friend making information and the identity information and the interest scoring information of the second user equipment in the friend making information as data characteristic items, and storing the 'success of making friends' as data labels into the server, wherein the data can be used as positive sample training data for training a friend making machine learning model; on the contrary, if the position information of the two is kept synchronous within one month for no more than 30 minutes, the friend-making is presumed to fail: and taking the identity information and the interest scoring information of the first user equipment in the friend making information and the identity information and the interest scoring information of the second user equipment in the friend making information as data characteristic items, and storing the failure of making friends as data labels into the server, wherein the data can be used as negative sample training data for training a friend making machine learning model.
And the positive sample training data of the friend making machine learning model and the negative sample training data of the friend making machine learning model are continuously superposed according to the increase of the friend making request information.
And S1001, under a preset condition, training the friend-making machine learning model by adopting a Support Vector Machine (SVM) algorithm according to the positive sample training data and the negative sample training data obtained in the step S901.
The friend-making machine learning model is a binary classifier and can be used for predicting the prediction results of 'recommending friend-making' and 'not recommending friend-making'.
The preset condition may be a preset time point, for example, 12 am every day, or a preset data amount condition, for example, when the increment of the positive sample training data of the training friend-making machine learning model and the increment of the negative sample training data of the training friend-making machine learning model exceeds 1000, the preset condition is met.
Step S1101: in a meal point time period, the server pushes meal friend-making information to the user equipment on site every preset time, and the method specifically comprises the following steps: according to the method of step S401, whether the user equipment on the spot finishes eating is detected, if the user equipment finishes eating, the user equipment does not push the information of making friends with eating, and the step is finished. If the user has not finished eating, sequentially traversing each user equipment in the friend recommendation list according to the friend recommendation list obtained in the step S601, wherein the traversed single user equipment is subjected to the following calculation process: sending identity information and interest scoring information corresponding to the user equipment and identity information and interest scoring information of the user equipment in the presence as feature items into the friend making machine learning model to obtain prediction results of 'recommending friend making' and 'not recommending friend making', if the prediction result is 'recommending friend making', and according to the method of the step S401, detecting that the user equipment does not finish eating, ending traversal, sending the identity information and interest scoring information corresponding to the user equipment and the user equipment in the presence as eating friend making information by the server, wherein the eating friend making information also can comprise the following word face information: "the system detects that you have not eaten, that your interests are similar, and can consider the recommended languages of" eat together ".
The machine learning-based dining friend-making information pushing method for the restaurant time period is combined with a machine learning algorithm and based on the high-demand friend-making time of the rice point, the success rate of friend-making recommendation can be greatly improved, and the user experience is improved.
In another scenario, step S1101 may also be: in a meal point time period, the server pushes meal friend-making information to the user equipment on site every preset time, and the method specifically comprises the following steps: according to the method of step S401, whether the user equipment on the spot finishes eating is detected, if the user equipment finishes eating, the user equipment does not push the information of making friends with eating, and the step is finished. If the user has not finished eating, sequentially traversing each user equipment in the friend recommendation list according to the friend recommendation list obtained in the step S601, wherein the traversed single user equipment is subjected to the following calculation process: calculating the similarity between the interest scoring information corresponding to the user equipment and the interest scoring information of the user equipment in the field, wherein the score of each kind of interest in the interest scoring information is improved or reduced according to a preset weight rule, for example: the scoring weight of the hobbies and interests of the catering is improved, and the scoring weight of other hobbies and interests is reduced. And obtaining the similarity of the interest and hobby scoring information of the user equipment in the presence, and calculating the user equipment with the highest score and without having finished eating. The server sends the identity information and the interest scoring information corresponding to the user equipment, and the identity information and the interest scoring information of the user equipment in the field serving as the information for making friends with dinner to the user equipment and the user equipment in the field, wherein the information for making friends with dinner can also comprise the following literal information: "the system detects that you have not eaten, that your dining interests are similar, and can consider the recommended languages of" eat together ".
The similarity algorithm may be cosine similarity, pearson similarity, or the like.
Due to the fact that the similarity algorithm is combined and the meal-point-based high-demand friend making time is used, the weight of catering hobbies is improved, the success rate of friend making recommendation can be greatly improved, and user experience is improved.
In the third scenario, step S1101 may also be: in a meal point time period, the server pushes meal friend-making information to the user equipment on site every preset time, and the method specifically comprises the following steps: detecting whether the user equipment in the presence has a grouping requirement, if so, sequentially traversing each user equipment in the friend recommendation list according to the friend recommendation list obtained in the step S601 to determine whether the user equipment in the presence has the grouping requirement identical to that of the user equipment in the presence;
the server sends the identity information and the interest scoring information corresponding to the user equipment with the same grouping requirement as the user equipment on the spot, and the identity information and the interest scoring information of the user equipment on the spot are used as the information for making friends with meals to the user equipment with the same grouping requirement as the user equipment on the spot and the user equipment on the spot, and the information for making friends with meals can also comprise the following literal information: "the system detects that you are both looking at a group, can consider having a meal together", etc.
The party is taken as a popular commercial promotion means, more and more consumers and merchants have accepted, and the friend making information is pushed in the restaurant time period based on the party demand, so that the user can meet the demand of reducing the dining cost and the friend making demand of the user based on the high demand friend making time of the rice point due to the combination of the party demand, the success rate of friend making recommendation can be greatly improved, and the user experience is improved.
Example 2
The shopping mall friend recommendation method based on the WiFi location fingerprint is used for user equipment. The shopping mall friend recommendation method based on the WiFi position fingerprint comprises the following steps.
Step S1201: the two-dimensional code of the WeChat friend-making small program entering a market is a pattern which is distributed with black and white at intervals on a plane (usually in a two-dimensional direction) according to a certain rule by using a certain specific geometric figure so as to record data symbol information. I.e. a two-dimensional code is a carrier for information recording.
The information acquisition page is generated through the WeChat applet, so that strangers in the vehicle can make friends without installing the application in advance by a user, the storage space of the mobile terminal is reduced, and inconvenience caused by forgetting to install the mobile terminal is avoided.
Step S1301: and the mall WeChat friend-making applet acquires the authorization of user WeChat and sends the identity information to the server. The authorization content comprises user information and Wi-Fi permission of WeChat;
in the invention, when the mall WeChat friend-making small program is opened, the mall WeChat friend-making small program acquires the authorization of user WeChat, and the authorization content comprises the user information acquisition and the Wi-Fi permission of WeChat.
The authorization content can be selected by the user according to the user requirement, convenience is provided for the user, and safety and user satisfaction are improved.
The mall WeChat friend-making applet can indicate that the user needs to make friends in the mall and is willing to open the mall WeChat friend-making applet, and the method can shield a part of users who do not need to make friends in the mall and improve user experience.
The identity information of the present user equipment includes, but is not limited to, MAC address information of the present user equipment, a micro credit user id, a micro credit user name, a micro credit user head portrait, a city where the micro credit user is located, a province where the micro credit user is located, a country where the micro credit user is located, a real name of the user, a gender of the user, an age of the user, and the like.
The identity information of the on-site user equipment can be selected by the user according to the user requirement, so that convenience is provided for the user, and the safety and the user satisfaction are improved.
Step S1401: and the presence user equipment sends a friend recommendation list acquisition request to the server.
Step S1501: and the on-site user equipment receives the friend recommendation list sent by the server and presents information in the friend recommendation list in the mall WeChat friend making applet.
Step S1601: and the on-site user equipment receives and presents the dining friend-making information sent by the server.
Step S1701: and the first presence user equipment receives friend making request operation of the user to obtain the friend making request information.
The friend making request operation may be: and according to the information in the friend recommendation list or the information for making friends with meals, the user clicks or inputs the communication information and the identity information of second on-site user equipment.
Step S1801: and the first presence user equipment sends the friend making request information to the server.
Step S1901: and the second presence user equipment receives the friend making request information sent by the server.
Claims (7)
1. A shopping mall friend recommendation method based on WiFi location fingerprints is characterized by comprising the following steps:
establishing a WiFi fingerprint database and a shop distribution map of a mall;
acquiring user information;
acquiring a real-time positioning position of user equipment according to a wifi fingerprint positioning technology;
generating a present user equipment list, including adding user equipment, of all the user equipment, of which the current time position exists in the shopping mall and which is authorized by the WeChat friend-making applet, into the present user equipment list at the current time to obtain the present user equipment list;
calculating a friend recommendation list;
receiving friend-making request information;
acquiring training data of a friend-making machine learning model;
training a friend-making machine learning model;
detecting whether the user equipment on site finishes eating, and if the user equipment finishes eating, not pushing the information of making friends with eating;
if the meal is not finished, generating meal dating information, wherein the meal dating information comprises traversing each user equipment in the friend recommendation list according to the friend recommendation list in sequence, and the traversed single user equipment is subjected to the following calculation process: identity information and interest score information corresponding to the single user equipment,
sending the identity information and the interest scoring information of the on-site user equipment as feature items to the friend making machine learning model to obtain prediction results of 'recommending friend making' and 'not recommending friend making', finishing traversal if the prediction result is 'recommending friend making', and detecting that the single user equipment does not finish eating, wherein the server takes the identity information and the interest scoring information corresponding to the single user equipment as eating friend making information;
and sending the dining friend-making information to the single user equipment and the user equipment in the presence.
2. The method of claim 1, wherein the method comprises: what is needed is
The mall store profile comprising:
collecting shop distribution information inside a mall and drawing a shop distribution map of the mall;
and classifying the shops inside the shopping mall.
3. The method of claim 1, wherein the method comprises: what is needed is
The acquiring of the user information comprises:
obtaining a wechat user id, a wechat user name, a wechat user head portrait, a city where the wechat user is located, a province where the wechat user is located, a country where the wechat user is located, a user real name, a user gender and a user age through wechat applet authorization;
acquiring MAC address information of user equipment through WIFI signal sending equipment;
and the user information automatically selects the public content according to the user requirement.
4. The method of claim 1, wherein the method comprises: the computing friend recommendation list comprises:
establishing a user-interest and hobby scoring matrix, which specifically comprises the following steps:
detecting and saving the action track of each user equipment in all the user equipment in real time,
the server is used for determining the behavior track of the user according to the position of the user equipment in the shop distribution map of the mall, the MAC address information of the user equipment and the time point information at the moment, the standard that the continuous stay time of the user equipment in the shop exceeds the preset time length is taken as the behavior track of the user, the position of the user equipment stays in a men's clothing shop within 15 minutes, the user is inferred to be interested in men's clothing, and the score of the user on the men's clothing is 1 point;
when the user equipment appears in a market within one month, the user goes to a shop for men's clothing every time, the stay time exceeds a preset threshold value, the fact that the user is very interested in the men's clothing is inferred, and the score of the user on the men's clothing is further 2 points and is 3 points;
at the time of the meal, the user equipment stays in the Sichuan-style hot pot restaurant for 30 minutes all the time, the user is inferred that the user has finished eating and is slightly interested in the Sichuan-style hot pot restaurant, the score of the user equipment in the Sichuan-style hot pot is 1 point, and the user equipment stays in the Sichuan-style hot pot restaurant for more than 30 minutes in one month
If the number of times exceeds 3 times, the user is inferred to be interested in Sichuan-style hot pot restaurants, and the score of the user in the Sichuan-style hot pot is added by 2 to be 3;
if the user device is not located in the restaurant for a period of time at the meal point time, such as 30 minutes of continuous stay, it is inferred that the user has not eaten;
and generating the friend recommendation list by combining pearson correlation coefficients according to the equipment list of the users in the presence and the user-interest and hobby scoring matrix.
5. The method of claim 1, wherein the method comprises: the information for making friends with meals comprises:
calculating the similarity between the interest scoring information corresponding to the single user equipment and the interest scoring information of the user equipment in the field, wherein the score of each kind of interest in the interest scoring information is increased or decreased according to a preset weight rule, and the method comprises the following steps: the scoring weight of the interests and hobbies of the catering is improved, and the scoring weight of other interests and hobbies is reduced; obtaining the similarity of the interest and hobby scoring information of the user equipment on the spot, and calculating the user equipment with the highest score and without having finished eating; the server sends the identity information and the interest scoring information corresponding to the single user equipment and the identity information and the interest scoring information of the on-site user equipment serving as dining friend-making information to the single user equipment and the on-site user equipment, wherein the dining friend-making information further comprises the following literal information: the system detects that two people have not eaten and the dining interests of the two people are very similar, and can consider the recommended languages of the two people who have eaten together;
or:
detecting whether the user equipment in the presence has a grouping requirement, if so, sequentially traversing each user equipment in the friend recommendation list according to the friend recommendation list to determine whether the user equipment in the presence has the grouping requirement which is the same as that of the user equipment in the presence;
the server sends the identity information and the interest scoring information corresponding to the user equipment with the same grouping requirement as the user equipment on the spot, and the identity information and the interest scoring information of the user equipment on the spot are used as the information for making friends with meals to the user equipment with the same grouping requirement as the user equipment on the spot and the user equipment on the spot, and the information for making friends with meals further comprises the following literal information: the system detects that you are both looking for a group and can consider the recommended languages to have a meal together.
6. The method of claim 1, wherein the method comprises: it is composed of
The obtaining of the training data of the friend-making machine learning model includes:
extracting communication information in friend making request information in the friend making information according to friend making information, detecting whether a communication text of ' making friends and ' seeing a shop bar ' is included, if yes, obtaining position information of a first user equipment in the friend making information and a second user equipment in the friend making information according to a wifi fingerprint positioning method, detecting whether the position information of the first user equipment and the second user equipment in the friend making information is synchronous within one month, and the time exceeds 30 minutes, if yes, presuming that the friend making is successful: taking the identity information and the interest scoring information of the first user equipment in the friend making information and the identity information and the interest scoring information of the second user equipment in the friend making information as data characteristic items, and storing the success of making friends as data labels into the server, wherein the data is used as positive sample training data for training a friend making machine learning model; on the contrary, if the position information of the two is kept synchronous within one month for no more than 30 minutes, the friend-making is presumed to fail: and taking the identity information and the interest scoring information of the first user equipment in the friend making information and the identity information and the interest scoring information of the second user equipment in the friend making information as data characteristic items, and storing the friend making failure as a data label into the server, wherein the data is used as negative sample training data for training a friend making machine learning model.
7. The method of claim 1, wherein the method comprises: it is composed of
The training friend-making machine learning model, comprising:
training the friend-making machine learning model by adopting a Support Vector Machine (SVM) algorithm according to the obtained positive sample training data and negative sample training data;
the friend-making machine learning model is a binary classifier and is used for predicting the prediction results of 'recommending friend-making' and 'not recommending friend-making'.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910473832.8A CN110312222B (en) | 2019-06-01 | 2019-06-01 | Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910473832.8A CN110312222B (en) | 2019-06-01 | 2019-06-01 | Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110312222A CN110312222A (en) | 2019-10-08 |
| CN110312222B true CN110312222B (en) | 2022-01-04 |
Family
ID=68074977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910473832.8A Active CN110312222B (en) | 2019-06-01 | 2019-06-01 | Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110312222B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113204715B (en) * | 2021-04-13 | 2024-03-29 | Oppo广东移动通信有限公司 | User recommendation method and device, electronic equipment and storage medium |
| CN113205358A (en) * | 2021-04-13 | 2021-08-03 | Oppo广东移动通信有限公司 | Information recommendation method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
| CN104572937A (en) * | 2014-12-30 | 2015-04-29 | 杭州云象网络技术有限公司 | Offline friend recommendation method based on indoor living circle |
| CN105721574A (en) * | 2016-02-05 | 2016-06-29 | 北京奇虎科技有限公司 | Method, server, device and system for users making friends |
| CN106162872A (en) * | 2016-08-22 | 2016-11-23 | 京信通信技术(广州)有限公司 | A kind of indoor locating system, method and information transmission system based on it, method |
| CN107786431A (en) * | 2015-04-30 | 2018-03-09 | 广东欧珀移动通信有限公司 | Method for message interaction and relevant apparatus and communication system and medium product |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9222788B2 (en) * | 2012-06-27 | 2015-12-29 | Microsoft Technology Licensing, Llc | Proactive delivery of navigation options |
-
2019
- 2019-06-01 CN CN201910473832.8A patent/CN110312222B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021233A (en) * | 2014-06-30 | 2014-09-03 | 电子科技大学 | Social network friend recommendation method based on community discovery |
| CN104572937A (en) * | 2014-12-30 | 2015-04-29 | 杭州云象网络技术有限公司 | Offline friend recommendation method based on indoor living circle |
| CN107786431A (en) * | 2015-04-30 | 2018-03-09 | 广东欧珀移动通信有限公司 | Method for message interaction and relevant apparatus and communication system and medium product |
| CN105721574A (en) * | 2016-02-05 | 2016-06-29 | 北京奇虎科技有限公司 | Method, server, device and system for users making friends |
| CN106162872A (en) * | 2016-08-22 | 2016-11-23 | 京信通信技术(广州)有限公司 | A kind of indoor locating system, method and information transmission system based on it, method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110312222A (en) | 2019-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Point-of-interest recommendation in location based social networks with topic and location awareness | |
| CN109376313B (en) | Information recommendation method and device, terminal and storage medium | |
| US11657429B1 (en) | Public access hyperlocal information exchange | |
| WO2020148659A2 (en) | Augmented reality based reactions, actions, call-to-actions, survey, accessing query specific cameras | |
| CN106296281B (en) | A kind of user individual travel information method for pushing, device and system | |
| Wang et al. | The new landscape of travel: A comprehensive analysis of smartphone apps | |
| CN103425659B (en) | Information search method and server based on geographical position | |
| WO2019056661A1 (en) | Search term pushing method and device, and terminal | |
| Vallespín et al. | Segmentation and explanation of smartphone use for travel planning based on socio-demographic and behavioral variables | |
| Bojic et al. | Choosing the right home location definition method for the given dataset | |
| Chen et al. | Assessing destination satisfaction by social media: An innovative approach using Importance-Performance Analysis | |
| US20180032997A1 (en) | System, method, and computer program product for determining whether to prompt an action by a platform in connection with a mobile device | |
| TW200947976A (en) | System and method for delivery of augmented messages | |
| WO2019141109A1 (en) | Method and device for content recommendation | |
| Falcone et al. | What is this place? Inferring place categories through user patterns identification in geo-tagged tweets | |
| CN110312222B (en) | Shopping mall friend recommendation method based on WiFi (Wireless Fidelity) position fingerprint | |
| US20180189382A1 (en) | System for Forming Connections Between Users | |
| US11373219B2 (en) | System and method for providing a profiled video preview and recommendation portal | |
| CN105824799B (en) | A kind of information processing method, equipment and terminal device | |
| KR20070037665A (en) | Instant order system and method using instant messaging, and instant knowledge service system and method using instant messaging | |
| US20160110774A1 (en) | Method and apparatus for recommending affiliated store by using reverse auction | |
| CN112241489A (en) | Information pushing method and device, readable storage medium and computer equipment | |
| CN113590936A (en) | Information pushing method and device | |
| Vanegas López et al. | Service quality in Medellin hotels using perceptual maps | |
| CN107948312B (en) | Information classification and release method and system with position points as information access ports |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211216 Address after: 201620 zone a, No. 58, Qingyun street, Songjiang Science Park, Songjiang District, Shanghai Applicant after: Shanghai Shengye Information Technology Co.,Ltd. Address before: Room f101-12, No.1 incubation and production building, guanshao shuangchuang (equipment) center, Huake City, 42 Baiwang Avenue, Wujiang District, Shaoguan City, Guangdong Province, 512026 Applicant before: Shaoguan Qizhi Information Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |