CN108154425B - Offline merchant recommendation method combining social network and location - Google Patents
Offline merchant recommendation method combining social network and location Download PDFInfo
- Publication number
- CN108154425B CN108154425B CN201810053915.7A CN201810053915A CN108154425B CN 108154425 B CN108154425 B CN 108154425B CN 201810053915 A CN201810053915 A CN 201810053915A CN 108154425 B CN108154425 B CN 108154425B
- Authority
- CN
- China
- Prior art keywords
- user
- merchant
- preference
- candidate
- social relationship
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 239000011159 matrix material Substances 0.000 claims abstract description 18
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 33
- 238000004891 communication Methods 0.000 claims description 10
- 230000006854 communication Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000007175 bidirectional communication Effects 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000011664 signaling Effects 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 230000006399 behavior Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an offline merchant recommendation method combining social network and location, which is used for a telecommunication e-commerce public service comprehensive support platform and comprises the following steps: data preprocessing, namely processing telecommunication data to obtain a required data table; constructing a social relationship network according to the call detail list of the user; adopting a CNM community discovery algorithm to perform hierarchical clustering on the social network; based on the user position, screening merchants according to the distance threshold value to obtain a candidate merchant list; analyzing the user internet log information and constructing a two-dimensional preference matrix of a user merchant; and recommending by adopting a collaborative filtering algorithm based on the user's converged social relationship. The invention constructs a social relationship network by using the user call information, excavates the social network, finds out a user group with close contact, and carries out recommendation by using a collaborative filtering algorithm based on the user-fused social relationship in combination with the user position, thereby reducing the computational complexity and improving the recommendation accuracy.
Description
Technical Field
The invention relates to the field of accurate marketing, in particular to an offline merchant recommendation method combining social networks and positions.
Background
With the continuous expansion of the scale of mobile intelligent terminal users, the mobile internet market has entered the high-speed development stage, and simultaneously, a large amount of user data is generated, and personalized recommendation for mobile terminal users is also concerned by operators and merchants. Mining valuable information using mobile internet user data is becoming a popular business marketing tool. The existing personalized recommendation method for telecom operators mainly analyzes user behaviors, preferences and consumption conditions through the historical data by acquiring user personal data, voice flow package consumption data, user internet log data and user track data, matches the user behaviors, preferences and consumption conditions with marketing target characteristics, converts matching results into marketing recommendation information and pushes the marketing recommendation information to users.
The communication behavior of mobile terminal users also accumulates a large amount of social activity data. Under the mobile communication network, users establish contact through mutual communication to form a social relationship network. Information is spread through interaction among social individuals in a social relationship network, and the social relationship network comprises the potential behavior rules of users.
The existing personalized recommendation method of the telecom operator only considers the individual historical preference and consumption information of the user, but does not consider the influence of the social relationship behavior of the user on the consumption preference of the user. And when the traditional collaborative filtering algorithm is adopted for recommendation, under the condition that the user quantity and the project quantity are huge, the scoring matrix is very huge and sparse, and the calculation complexity is very high.
Disclosure of Invention
In order to solve the problems of the existing personalized recommendation method, the invention provides an offline merchant recommendation method combining the social network and the position.
The technical scheme adopted by the invention is as follows: an offline merchant recommendation method combining social networks and locations comprises the following steps:
step S1, data preprocessing, processing the signaling data and the user merchant data to obtain a user information table, a user track table, a user internet log table, a user call detail table and a merchant information table;
step S2, according to the user call detail list, quantifying the social relationship among users and constructing a social relationship network;
s3, performing hierarchical clustering on the constructed social relationship network, and discovering user groups closely related in the social relationship network;
step S4, based on the user position, screening merchants according to the distance threshold value to obtain a candidate merchant list;
step S5, analyzing user internet log information based on the clustering result in the step S3 and the candidate merchant list in the step S4, and constructing a user merchant two-dimensional preference matrix;
and S6, recommending the user and merchant two-dimensional preference matrix constructed in the step S5 by adopting a collaborative filtering algorithm based on the user social relationship fusion.
Preferably, the social relationship network in step S2 is represented by a directed graph G ═ (V, E, W, P), where V is a set of vertices representing all calling and called user nodes; e is a directed edge set, the edge E of whichijRepresenting a node viCalling called node v as calling nodej(ii) a W is the edge E in the directed edge set EijA set of attribute vectors of (1), a one-dimensional vector W of WijIs an edge eijAttribute vector of (2), representing node viAnd node vjThe characteristics of the relationship between the nodes reflect the node viAnd node vjThe degree of tightness of the tube.
Preferably, the process of step S4 is as follows: in the user track table, the position (x) of the user position at the time point t is obtainedt,yt) Wherein x istRepresents longitude, ytRepresenting the latitude; setting a distance threshold D, R as the radius of the earth, and calculating the distance (x) from the user positiont,yt) The latitude and longitude range of D is not exceeded; screening out a part of candidate merchants meeting the conditions through the latitude and longitude range, and then calculating the position (x) of the merchantc,yc) And user location (x)t,yt) A distance d of; and then, removing the merchants with the distance D larger than D to obtain a final candidate merchant list.
Preferably, the step S5 of constructing the two-dimensional preference matrix of the user and merchant is as follows:
s5.1, taking Internet log records of the user in the latest T time period from a user Internet log table, and extracting keywords in a user search field to obtain a keyword list of the user;
s5.2, converting each word into a k-dimensional word vector for the keyword list, and then carrying out weighted average on the word vectors to obtain the keyword word vector of the user;
s5.3, for the merchants in the candidate merchant list, taking the shop description field, and calculating to obtain the keyword vector of the merchant according to the methods in the step S5.1 and the step S5.2;
s5.4, calculating preference values P for each user i and each merchant j in sequenceijObtaining a two-dimensional preference matrix of the user and the commercial tenant, wherein the preference value PijDefining the cosine similarity between the user keyword vector and the merchant keyword vector;
preferably, after hierarchical clustering in step S3, adding cluster label attributes to the users, wherein the user cluster labels belonging to the same cluster are the same; in step S5.4, the user i selects the user list with the same cluster mark, and the merchant j selects the candidate merchant list obtained in step S3.
Preferably, the step S6 is performed by using a collaborative filtering algorithm based on the user' S merged social relationship, and includes the following steps:
s6.1, regarding the two-dimensional preference matrix of the user commercial tenant, based on the preference vector of the user to the commercial tenant
Calculating and sequencing the similarity between the users, and taking K users before the similarity;
s6.2, calculating the relationship closeness between the user i and K users before the similarity according to the constructed social relationship network for K users before the similarity;
and S6.3, predicting the preference of the user i on the candidate commercial tenants, and selecting the commercial tenants with the maximum preference values to recommend the commercial tenants to the user.
Compared with the existing personalized recommendation method, the invention has the advantages that:
the social relationship network is established by using the user call information, the social relationship network is mined to find out a user group with close contact, and then the recommendation is performed by using a collaborative filtering algorithm based on the user-fused social relationship in combination with the user position, so that the calculation complexity is reduced, and the recommendation accuracy is improved.
Drawings
FIG. 1 is a flow chart of an offline merchant recommendation method incorporating social networking and location in accordance with the present invention;
FIG. 2 is a flowchart of step S5 of the present invention for constructing a two-dimensional preference matrix for user-merchants.
Detailed Description
The present invention will be further described with reference to the following drawings and examples, but the embodiments of the present invention are not limited thereto.
Examples
As shown in fig. 1, in this embodiment, the offline merchant recommendation method combining the social network and the location includes the following steps:
and step S1, data preprocessing, processing the signaling data and the user merchant data to obtain a user information table, a user track table, a user internet log table, a user call detail table and a merchant information table.
The data preprocessing comprises checking the correctness, consistency and integrity of the data and whether the data conforms to a business rule or not, and processing abnormal values, missing values, invalid values and redundant data; and carrying out feature processing on the data attributes, and converting the data attributes into data features to form a feature library.
The user information table comprises user identification, gender, age, attribution, network access time, VIP (very important person) and group users; the user track table comprises user identification, time, longitude and latitude; the user surfing log table comprises user identification, time, search keywords and url clicked by a user; the user call detail list comprises user identification, starting time, opposite terminal numbers, call types and call duration; the merchant data includes merchant identification, merchant name, industry category, longitude, latitude, address, contact phone, store description.
And step S2, quantifying social relationships among users according to the user call detail list, and constructing a social relationship network. The specific method comprises the following steps:
for a social relationship network formed by user communication behaviors, a directed graph G is represented by (V, E, W, P), wherein V is a vertex set and represents all calling and called user nodes; e is a set of directed edges, where edge EijRepresenting a node viCalling called node v as calling nodej(ii) a W is a set of directed edge attribute vectors, corresponding to the directed edges in E.
The directed edge attribute set W is the set of directed edges EijSet of attribute vectors. One-dimensional vector wijIs an edge eijAttribute vector of (2), representing node viAnd node vjThe characteristics of the relationship between the nodes reflect the node viAnd node vjThe degree of tightness of the tube.
Attribute vector wijThe properties of (A) are: bidirectional communication condition, communication times and call duration. Two-way communication situation, i.e. wijAnd wjiWhether or not they are present at the same time; number of communications, node viAs a caller and a node vjTotal number of communications of (1); duration of call, node viAs a caller and a node vjThe total duration of the call.
And step S3, performing hierarchical clustering on the social relationship network constructed in the step S2 by adopting a CNM community discovery algorithm, and discovering a user group with close contact in the social relationship network. After clustering, adding cluster mark attributes for the users, wherein the user cluster marks belonging to the same cluster are the same.
Step S4, based on the user position, screening merchants according to the distance threshold to obtain a candidate merchant list, wherein the specific method comprises the following steps:
in the user track table, the position (x) of the user position at the time point t is obtainedt,yt) Wherein x istRepresents longitude, ytRepresenting the latitude. Setting a distance threshold D, R as the radius of the earth, and calculating the distance (x) from the user positiont,yt) Longitude and latitude not exceeding KDegree range:
Δy＝2sin-1(sinD/(2R))/cosxt)
Δx＝D/R
and position (x)t,yt) The coarse latitude and longitude range not exceeding D is (x)t±Δx,yt± Δ y). Screening out a part of candidate merchants meeting the conditions through latitude and longitude ranges, and then calculating the position (x) of the merchant by utilizing a Haverine formulac,yc) And user location (x)t,yt) Distance d of (d):
Δx＝xc-xt
Δy＝yc-yt
and then, removing the merchants with the distance D larger than D to obtain a final candidate merchant list.
Step S5, analyzing the user internet log information based on the clustering result in the step S3 and the candidate merchant list in the step S4, and constructing a two-dimensional preference matrix of the user merchants, which specifically comprises the following steps:
s5.1, in a user Internet log table, taking Internet log records of the user in the latest T time period, and extracting keywords in a user search field by using TF-IDF:
and selecting words with TFIDF larger than a certain set threshold value as the keywords of the user to obtain a keyword list of the user.
Step S5.2, for the keyword list obtained in the step S5.1, converting each word into a k-dimensional word vector by using word2vec
Then, the word vectors are weighted and averaged to obtain the key word vector of the user
S5.3, for the commercial tenant in the candidate commercial tenant list, the shop description field is taken, and the keyword vector of the commercial tenant is calculated and obtained according to the method in the step S5.1 and the step S5.2
S5.4, for the user i and the commercial tenant j, the preference value P of the user i to the commercial tenant j in the two-dimensional preference matrix PijThe cosine similarity defined as the similarity between the user keyword vector and the merchant keyword vector is as follows:
selecting a user list userlist with the same cluster mark and the shoplist obtained in the step S3, and calculating P sequentially for each user i and each merchant jijAnd obtaining a two-dimensional preference matrix of the user and the merchant.
S6, recommending the user-merchant two-dimensional preference matrix constructed in the S5 by adopting a collaborative filtering algorithm based on the user-integrated social relationship, and comprising the following steps of:
s6.1, regarding the two-dimensional preference matrix of the user commercial tenant, based on the preference vector of the user to the commercial tenant
Calculating and sequencing the similarity among the users, and taking the first K users of the similarity, wherein the similarity is defined as:
wherein sim (i, j) is the similarity of users i, j,
and the preference vectors of the users i and j to the merchants are respectively.
Step S6.2, for K users before the similarity with the user i obtained by calculation in step S6.1, calculating the relationship closeness between the user i and the K users before the similarity according to the social relationship network constructed in step S2, where the relationship closeness is defined as:
c(i,j)＝t(i,j)*log(m(i,j))*factor
wherein t (i, j) is the number of times of call of the user i, j, and m (i, j) is the call duration of the user i, j; the factor is 1 when the user i, j has two-way communication, and the factor is 0.5 when there is no two-way communication.
S6.3, predicting the preference of the user i to the candidate commercial tenants, wherein the calculation formula is as follows:
obtaining the preference of user i to candidate merchants
And selecting the commercial tenant with the largest preference value to recommend to the user. Wherein c (i, j) is the relationship closeness of the user i, j,
a preference vector for candidate merchant for user j.
From the above, the offline merchant recommendation method combining the social network and the position is used for the telecommunication and electronic commerce public service comprehensive support platform, the method is based on telecommunication data analysis, the social relationship network is constructed by using the user call information, the social relationship network is mined to find out the user group with close contact, and then the user position is combined, and the collaborative filtering algorithm based on the user fused social relationship is used for recommendation, so that the calculation complexity is reduced, and the recommendation accuracy is improved.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include such modifications and variations.
Claims (4)
1. An offline merchant recommendation method combining a social network and a location is characterized by comprising the following steps:
step S1, data preprocessing, processing the signaling data and the user merchant data to obtain a user information table, a user track table, a user internet log table, a user call detail table and a merchant information table;
step S2, according to the user call detail list, quantifying the social relationship among users and constructing a social relationship network;
s3, performing hierarchical clustering on the constructed social relationship network by adopting a CNM community discovery algorithm, and exploring a user group with close contact in the social relationship network;
step S4, based on the user position, screening merchants according to the distance threshold value to obtain a candidate merchant list;
step S5, analyzing user internet log information based on the clustering result in the step S3 and the candidate merchant list in the step S4, and constructing a user merchant two-dimensional preference matrix;
s6, recommending the user-merchant two-dimensional preference matrix constructed in the S5 by adopting a collaborative filtering algorithm based on the user-fused social relationship;
the process of step S4 is as follows: in the user track table, the position (x) of the user position at the time point t is obtainedt，yt) Wherein x istRepresents longitude, ytRepresenting the latitude; setting a distance threshold D, R as the radius of the earth, and calculating the distance (x) from the user positiont，yt) The latitude and longitude range of D is not exceeded; screening out a part of candidate merchants meeting the conditions through the latitude and longitude range, and then calculating the position (x) of the merchantc，yc) And user location (x)t，yt) A distance d of; then, commercial tenant with the distance D larger than D is removed, and the final candidate commercial tenant column is obtainedTable;
step S6, recommending by adopting a collaborative filtering algorithm based on the user-fused social relationship, comprising the following steps:
s6.1, regarding the two-dimensional preference matrix of the user commercial tenant, based on the preference vector of the user to the commercial tenant
Calculating and sequencing the similarity between the users, and taking K users before the similarity;
s6.2, calculating the relationship closeness between the user i and K users before the similarity according to the constructed social relationship network for K users before the similarity;
s6.3, predicting the preference of the user i on the candidate commercial tenants, and selecting the commercial tenants with the largest preference values to recommend the commercial tenants to the user;
the closeness of relationship is defined as:
c(i，j)＝t(i，j)*log(m(i，j))*factor
wherein t (i, j) is the number of times of call of the user i, j, and m (i, j) is the call duration of the user i, j; when the user i, j has bidirectional communication, the factor is 1, and when the user i, j does not have bidirectional communication, the factor is 0.5;
predicting the preference of the user i to the candidate commercial tenants, wherein the calculation formula is as follows:
wherein c (i, j) is the relationship closeness of the users i, j, sim (i, j) is the similarity between the users i, j,
a preference vector of the user j to the candidate commercial tenant;
the step S5 of constructing the two-dimensional preference matrix of the user and the merchant is as follows:
s5.1, taking Internet log records of the user in the latest T time period from a user Internet log table, and extracting keywords in a user search field to obtain a keyword list of the user;
s5.2, converting each word into a k-dimensional word vector for the keyword list, and then carrying out weighted average on the word vectors to obtain the keyword word vector of the user;
s5.3, for the merchants in the candidate merchant list, taking the shop description field, and calculating to obtain the keyword vector of the merchant according to the methods in the step S5.1 and the step S5.2;
s5.4, calculating preference values P for each user i and each merchant j in sequenceijObtaining a two-dimensional preference matrix of the user and the commercial tenant, wherein the preference value PijDefined as the cosine similarity of the user keyword vector and the merchant keyword vector.
2. The offline merchant recommendation method according to claim 1, wherein the attribute vector wijThe attributes of (1) include: bidirectional communication condition, communication times and call duration.
3. The offline merchant recommending method according to claim 1, wherein after the step S3 hierarchical clustering, cluster label attributes are added to the users, and the user cluster labels belonging to the same cluster are the same; in step S5.4, the user i selects the user list with the same cluster mark, and the merchant j selects the candidate merchant list obtained in step S3.
4. The offline merchant recommendation method according to claim 1, wherein the data preprocessing comprises checking the correctness, consistency, completeness, and compliance with business rules of the data, and processing abnormal values, missing values, invalid values, and redundant data; and carrying out feature processing on the data attributes, and converting the data attributes into data features to form a feature library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810053915.7A CN108154425B (en) | 2018-01-19 | 2018-01-19 | Offline merchant recommendation method combining social network and location |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810053915.7A CN108154425B (en) | 2018-01-19 | 2018-01-19 | Offline merchant recommendation method combining social network and location |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108154425A CN108154425A (en) | 2018-06-12 |
| CN108154425B true CN108154425B (en) | 2021-03-26 |
Family
ID=62461712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810053915.7A Active CN108154425B (en) | 2018-01-19 | 2018-01-19 | Offline merchant recommendation method combining social network and location |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108154425B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110611689B (en) * | 2018-06-15 | 2022-06-28 | 中移动信息技术有限公司 | Information identification method and device and computer readable storage medium |
| CN109657145A (en) * | 2018-12-20 | 2019-04-19 | 拉扎斯网络科技（上海）有限公司 | Merchant searching method and device, electronic equipment and computer-readable storage medium |
| CN111491060B (en) * | 2019-01-28 | 2021-10-01 | 百度在线网络技术（北京）有限公司 | Information click log and ticket splicing method and device |
| CN111724183A (en) * | 2019-03-20 | 2020-09-29 | 中国银联股份有限公司 | Merchant recommendation method and merchant recommendation system |
| CN110413897B (en) * | 2019-06-14 | 2023-10-27 | 腾讯科技（深圳）有限公司 | User interest mining method and device, storage medium and computer equipment |
| CN110866180B (en) * | 2019-10-12 | 2022-07-29 | 平安国际智慧城市科技股份有限公司 | Resource recommendation method, server and storage medium |
| CN111506823A (en) * | 2020-03-11 | 2020-08-07 | 平安科技（深圳）有限公司 | Information recommendation method and device and computer equipment |
| CN111460332B (en) * | 2020-04-26 | 2020-10-30 | 成都信息工程大学 | Social relationship evaluation method based on user life track |
| CN111782620B (en) * | 2020-06-19 | 2024-07-05 | 庞茂林 | Credit link automatic tracking platform and method thereof |
| CN113792089B (en) * | 2021-09-16 | 2024-03-22 | 平安银行股份有限公司 | Illegal behavior detection method, device, equipment and medium based on artificial intelligence |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommendation method and system integrating attributes and structural similarity |
| CN103605791A (en) * | 2013-12-04 | 2014-02-26 | 深圳中兴网信科技有限公司 | Information pushing system and information pushing method |
| CN103995823A (en) * | 2014-03-25 | 2014-08-20 | 南京邮电大学 | Information recommending method based on social network |
| CN105721279A (en) * | 2016-01-15 | 2016-06-29 | 中国联合网络通信有限公司广东省分公司 | Relationship circle excavation method and system of telecommunication network users |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7792697B2 (en) * | 2004-12-28 | 2010-09-07 | American Express Travel Related Services Company, Inc. | System and method for predicting card member spending using collaborative filtering |
| CN102750647A (en) * | 2012-06-29 | 2012-10-24 | 南京大学 | Merchant recommendation method based on transaction network |
| SG10201502188PA (en) * | 2015-03-20 | 2016-10-28 | Mastercard Asia Pacific Pte Ltd | Method and system for recommending a competitor |
| CN105025091A (en) * | 2015-06-26 | 2015-11-04 | 南京邮电大学 | Shop recommendation method based on position of mobile user |
| CN105184603A (en) * | 2015-08-13 | 2015-12-23 | 湖南天冠电子信息技术有限公司 | Destination recommending method integrated with collaborative filtering algorithm and position information |
| CN107169052B (en) * | 2017-04-26 | 2019-03-05 | 北京小度信息科技有限公司 | Recommended method and device |
-
2018
- 2018-01-19 CN CN201810053915.7A patent/CN108154425B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommendation method and system integrating attributes and structural similarity |
| CN103605791A (en) * | 2013-12-04 | 2014-02-26 | 深圳中兴网信科技有限公司 | Information pushing system and information pushing method |
| CN103995823A (en) * | 2014-03-25 | 2014-08-20 | 南京邮电大学 | Information recommending method based on social network |
| CN105721279A (en) * | 2016-01-15 | 2016-06-29 | 中国联合网络通信有限公司广东省分公司 | Relationship circle excavation method and system of telecommunication network users |
Non-Patent Citations (1)
| Title |
|---|
| "基于用户网络社交信息的推荐算法研究与实现";付晶晶;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180115(第01期);正文第1-3章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108154425A (en) | 2018-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108154425B (en) | Offline merchant recommendation method combining social network and location | |
| US9336282B2 (en) | Systems and methods for identifying and analyzing internet users | |
| CN106778876B (en) | User classification method and system based on mobile user track similarity | |
| CN108153824B (en) | Method and device for determining target user group | |
| CN110119478B (en) | Similarity-based item recommendation method combining multiple user feedback data | |
| KR100792700B1 (en) | Method for targeting web advertisement clickers based on click pattern by using a collaborative filtering system with neural networks and system thereof | |
| CN115408618B (en) | Point-of-interest recommendation method based on social relation fusion position dynamic popularity and geographic features | |
| CN112241489A (en) | Information pushing method and device, readable storage medium and computer equipment | |
| KR101450453B1 (en) | Method and apparatus for recommending contents | |
| CN105786936A (en) | Search data processing method and device | |
| KR101910424B1 (en) | Method for movie ratings prediction using sentiment analysis of movie tags, recording medium and device for performing the method | |
| CN109299368B (en) | Method and system for intelligent and personalized recommendation of environmental information resources AI | |
| CN111931069A (en) | User interest determination method and device and computer equipment | |
| CN114282119B (en) | Scientific and technological information resource retrieval method and system based on heterogeneous information network | |
| Yang et al. | Personalized recommendation based on collaborative filtering in social network | |
| CN115455302A (en) | Knowledge graph recommendation method based on optimized graph attention network | |
| CN106066864B (en) | A kind of various dimensions mobile subscriber preference dynamic identifying method | |
| CN115098771A (en) | Recommendation model updating method, recommendation model training method and computing device | |
| CN114880572B (en) | Intelligent news client recommendation system | |
| Gupta et al. | Recommendation system for location-based services | |
| CN115687801B (en) | Position recommendation method based on position aging characteristics and time perception dynamic similarity | |
| Rohilla et al. | Random Forest with harmony search optimization for location based advertising | |
| CN116739641B (en) | Cross-border electronic commerce knowledge graph analysis method and system | |
| CN115878905B (en) | Individual service recommendation system of wisdom business district | |
| Rios et al. | Selecting and Weighting Users in Collaborative Filtering-based POI Recommendation |
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 |