CN112052380A - Intelligent terminal, and marketing recommendation method and system of intelligent household equipment - Google Patents
Intelligent terminal, and marketing recommendation method and system of intelligent household equipment Download PDFInfo
- Publication number
- CN112052380A CN112052380A CN202010067606.2A CN202010067606A CN112052380A CN 112052380 A CN112052380 A CN 112052380A CN 202010067606 A CN202010067606 A CN 202010067606A CN 112052380 A CN112052380 A CN 112052380A
- Authority
- CN
- China
- Prior art keywords
- user
- pool
- equipment
- host
- users
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 64
- 239000013598 vector Substances 0.000 claims abstract description 21
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 9
- 230000001276 controlling effect Effects 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 7
- 238000001914 filtration Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 3
- 239000010410 layer Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 239000002356 single layer Substances 0.000 description 2
- 238000005054 agglomeration Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
Abstract
The invention relates to an intelligent terminal, and a marketing recommendation method and system of intelligent household equipment. The method comprises the following steps: classifying all the hosts by adopting a clustering algorithm according to the equipment list bound by each host; counting users under each type of host to form a user pool of each type of host; for a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm; then, classifying the users in the user pool by adopting a clustering algorithm; counting an equipment list bound by a host where each type of user is located in the user pool to form an equipment pool of each type of user in the user pool; and recommending the equipment which is not purchased by the user in the corresponding equipment pool for a certain user under each type of users. According to the method, the used data range is narrowed when recommendation is carried out through double clustering, and the click operation habit of recent equipment of the user is considered, so that the similarity of similar users is higher, the recommendation accuracy is improved, and the calculation amount is effectively reduced.
Description
Technical Field
The invention relates to an intelligent terminal, and a marketing recommendation method and system of intelligent home equipment, and belongs to the technical field of intelligent home.
Background
With the advent of the big data era, data acquisition and storage equipment of various industries is continuously sound, and the data volume of users is rapidly increased from both time dimension and space dimension. The intelligent home industry is particularly a large-data-yield user, accurate marketing of intelligent home equipment can be realized through the data, and purchasable intelligent home products suitable for the user are recommended for the user.
Currently, the algorithms used for marketing in the prior art are the following:
1. and (4) clustering algorithm. The clustering algorithm for unsupervised learning is mainly classified into the following categories: a partition-based clustering algorithm, a hierarchy-based clustering algorithm, a density-based clustering algorithm, and the like. For example: the DBSCAN clustering algorithm can be used for clustering dense data sets in any shapes, and is strong in abnormal point interference resistance; however, when the spatial clustering density is not uniform and the clustering interval difference is large, the clustering quality is poor, so that the accuracy of the recommended result is reduced, and the physical examination effect of the user is poor.
2. Collaborative filtering algorithms. The collaborative filtering algorithm is used as a popular algorithm of a recommendation system and is widely applied to various large e-commerce platforms. Collaborative filtering algorithms are divided into two categories, namely user-based and article-based, and the article-based collaborative filtering algorithms are difficult to provide recommendation explanations for convincing users, so that the user experience is deteriorated; the collaborative filtering algorithm based on the users is suitable for occasions with few users, and if the number of the users is large, the cost of calculating the similarity matrix of the users is high, the recommendation efficiency is low, and the experience effect of the users is poor.
3. A time decay algorithm. When recommending commodities to users, the e-commerce industry analyzes interest preference of the users for the platform commodities, and meanwhile, the interest preference also changes along with the lapse of time. The time decay algorithm considers that the interest preference of the user is related to the time, however, the overall recommendation is not accurate due to the fact that the algorithm focuses on considering the recent interest preference of the user, and the physical examination effect of the user is poor.
In conclusion, the existing algorithm is inaccurate in recommendation, low in efficiency and poor in user experience effect.
Disclosure of Invention
The invention aims to provide a marketing recommendation method for intelligent household equipment, which is used for solving the problems of low recommendation accuracy, low efficiency and poor user experience effect of the existing marketing recommendation method; meanwhile, a marketing recommendation system of the intelligent household equipment is also provided, and the problems that the existing recommendation system is low in recommendation accuracy and efficiency and poor in user experience effect are solved; still provide an intelligent terminal simultaneously for solve current intelligent terminal recommendation degree of accuracy low, inefficiency, the not good problem of user experience effect.
In order to achieve the purpose, the invention provides a marketing recommendation method for intelligent household equipment, which comprises the following steps:
classifying all the hosts by adopting a clustering algorithm according to the equipment list bound by each host;
counting users under each type of host to form a user pool of each type of host;
for a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm according to the recent equipment clicking times of the user;
classifying the users in the user pool by adopting a clustering algorithm according to the recommended vector of each user in the user pool;
counting an equipment list bound by a host where each type of user is located in the user pool to form an equipment pool of each type of user in the user pool;
and for a certain user under each type of users in the user pool, recommending a recommendation list generated by the equipment which is not purchased by the user in the corresponding equipment pool to the user.
In addition, the invention also provides a marketing recommendation system of the intelligent household equipment, which comprises a mobile terminal and a control terminal, wherein the mobile terminal comprises input and output equipment and a wireless communication module used for communicating with the control terminal; the control terminal comprises a wireless communication device for communicating with the mobile terminal, a memory, a processor and a computer program which is stored in the memory and can run on the processor, and the processor realizes the marketing recommendation method of the intelligent household equipment when executing the computer program.
In addition, the invention also provides an intelligent terminal, which comprises an input/output device, a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes the marketing recommendation method of the intelligent household equipment when executing the computer program.
The beneficial effects are that: the method comprises the steps of firstly classifying hosts through a clustering algorithm to obtain a user pool of each type of host, ensuring that each user pool has similar equipment, then classifying users in the user pool through the clustering algorithm, before classifying the users in the user pool, considering the influence of recent behaviors of the users, obtaining a recommendation vector of each user through the recent behaviors of the users through a time attenuation algorithm, further classifying the users through the clustering algorithm according to the recommendation vector, enabling the users in each type of users to be more similar, and then recommending the users in each type of users in the user pool, wherein the equipment owned by the users is considered. According to the method, the used data range is narrowed down during recommendation through a double-layer clustering algorithm, and the recent behaviors of the user are taken into consideration, so that the recommendation accuracy is improved, the calculation amount is effectively reduced, the recommendation efficiency is improved, and the recommendation timeliness is improved.
Furthermore, in the marketing recommendation method and system for the intelligent terminal and the intelligent home device, in order to improve the accuracy of the time decay algorithm, not only the influence of the single-day click frequency is considered, but also the cumulative click days of the same device are further considered, and the time decay algorithm is as follows:
wherein Y is the recommended vector of the user, G is the time attenuation coefficient, x1,x2,...,xTThe number of clicks of each device of the user is 1 day, 2 days, … … days and T days from the recommended day, X is the accumulated number of clicks of each device of the user in the T day, and alpha is a weight coefficient for regulating and controlling the proportion of X.
Further, in the marketing recommendation method and system for the intelligent terminal and the intelligent home device, in order to improve the recommendation accuracy, the recommendation list is: recommendation lists generated by devices in the device pool that are clicked on frequently but not purchased by the user.
Furthermore, in the marketing recommendation method and system for the intelligent terminal and the intelligent home equipment, in order to reduce the calculation amount, users unbound by the host are filtered when counting the users under each type of host.
Furthermore, in the marketing recommendation method and system for the intelligent terminal and the intelligent home equipment, in order to improve the accuracy of host classification, a CURE clustering algorithm is adopted to classify the hosts.
Furthermore, in the marketing recommendation method and system for the intelligent terminal and the intelligent home equipment, in order to improve accuracy of user classification, a DBSCAN clustering algorithm is adopted to classify users in the user pool.
Drawings
Fig. 1 is a flowchart of a marketing recommendation method of smart home equipment according to the present invention.
Detailed Description
The embodiment of the marketing recommendation system of the intelligent household equipment comprises the following steps:
the marketing recommendation system of the smart home device provided by the embodiment comprises a mobile terminal and a control terminal, wherein the mobile terminal comprises an input/output device and a communication module for communicating with the control terminal; the control terminal comprises a communication device used for communicating with the mobile terminal, a memory and a processor, wherein a computer program capable of running on the processor is stored in the memory, and the processor realizes the marketing recommendation method of the intelligent household equipment when executing the computer program.
The mobile terminal is a terminal for providing an operation interface for a user, and can be terminal equipment in forms of android, ios and the like, wherein the terminal equipment comprises an intelligent home App based on electronic products such as a mobile phone and a pad, the input and output equipment is a touch screen, the control terminal is an intelligent home host (hereinafter referred to as a host), and the marketing recommendation method of the intelligent home equipment is realized by sending an obtained recommendation list to the mobile terminal for displaying. Of course, the control terminal may also be a host and a background (the control terminal in this embodiment), and the corresponding function of the control terminal is implemented. Moreover, the communication module may be a bluetooth module, a WIFI module, or 3G, 4G, GPRS, or even a wired communication module, and the present invention does not limit the specific implementation manner of the communication module.
The marketing recommendation method of the intelligent household equipment comprises the following steps:
classifying all the hosts by adopting a clustering algorithm according to the equipment list bound by each host;
counting users under each type of host to form a user pool of each type of host;
for a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm according to the recent equipment clicking times of the user; classifying the users in the user pool by adopting a clustering algorithm according to the recommended vector of each user in the user pool;
counting an equipment list bound by a host where each type of user is located in the user pool to form an equipment pool of each type of user in the user pool;
and for a certain user under each type of users in the user pool, recommending a recommendation list generated by the equipment which is not purchased by the user in the corresponding equipment pool to the user.
Specifically, the marketing recommendation method of the smart home device is shown in fig. 1, and includes the following steps:
1) first, user data is collected, and the subdivision categories (a categories) of all sold devices are counted, and the device categories are numbered.
The data that needs to be collected are: a list of devices bound under each host; all equipment lists and the divided subclasses of the intelligent home are obtained; clicking an operation record of the equipment by a user; records of binding and unbinding of a user to a host. The collected data is reported to a background (namely a cloud) at regular time through the host, the calculation of the recommendation method is carried out on the background, the recommendation result is stored in the background, and when the recommendation is needed, the background issues the recommendation result to the corresponding host, so that the recommendation of each host is realized.
2) And generating a characteristic vector Q of each host according to the equipment list bound by each host, wherein each component in the characteristic vector Q represents the purchase quantity of a types of equipment, if a certain equipment is not purchased, the value of the component is 0, and classifying the characteristic vectors generated by all the hosts by adopting a clustering algorithm to form n types of hosts.
The clustering algorithm is to divide a data set into different classes or clusters according to a certain criterion (such as a distance criterion), so that the similarity of objects in the same cluster is as large as possible, and the difference of objects in different clusters is as large as possible.
In the embodiment, the clustering algorithm adopted for classifying the hosts is a CURE clustering algorithm, belongs to an agglomeration method in a hierarchical clustering method, has good interpretability, can generate high-quality clusters, has strong abnormal value resistance, and can solve the non-spherical group. The clustering algorithm needs to provide 4 input data before initial iteration: the data set D, the classification number K, the contraction factor a is 0.2-0.7, and the representative point c is generally more than 10. As another embodiment, other clustering algorithms, such as WAVECLUSTER, ROCK, BICCH, K-PROTOTYPES, DENCLUE, OPTIRID, CLIQUE, DBSCAN, CLARANS, etc., may be used to classify the hosts, which is not limited by the invention.
3) And counting users under each type of host in the n types of hosts to form a user pool of each type of host, and summing up n user pools.
The users under the host comprise users bound with the host, so the users are filtered in the step, whether the users are unbound with the host is judged, and statistical recommendation is not carried out on the unbound users or the users with binding time less than T days, so that the calculation amount is reduced; and carrying out statistical recommendation on the users which are always bound or are bound again within T days. And the data of the users who are bound again within the T day records statistics from the binding day.
4) For a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm according to the recent T-day equipment click times (namely, operation records) of the user; and then classifying the users in the user pool by adopting a clustering algorithm according to the recommendation vector of each user in the user pool. The classification is m types.
In this embodiment, in order to improve the reliability of the time decay algorithm, the time decay algorithm is:
wherein, Y is the recommendation vector (namely the prediction vector of the recommendation day) of the user, G is the time attenuation coefficient, x1,x2,...,xTThe number of clicks of each device of the user is 1 day, 2 days, … … days and T days from the recommended day, X is the accumulated number of clicks of each device of the user in the T day, and alpha is a weight coefficient for regulating and controlling the proportion of X.
As another embodiment, a time decay algorithm in the prior art may also be used to obtain the recommendation list, where the current time decay algorithm is:
the interpretation of the characters in the formula is the same as the above formula, and the description thereof is omitted here.
The existing time attenuation algorithm focuses on the consideration of the influence of the single-day click frequency, however, according to the actual user operation condition, the accumulated click days of the same equipment are also particularly important, so that the time attenuation algorithm provided by the invention takes the accumulated click days of the same equipment into consideration, optimizes the prior art, determines the parameters G and alpha mainly according to the test data, the accuracy and the recall rate, and does not need to be repeated.
In the embodiment, after the recommended vector of each user is obtained, the users in the user pool are classified by adopting a density-based clustering algorithm, namely a DBSCAN clustering algorithm, the algorithm is high in clustering speed, can effectively process noise points and find spatial clusters with any shapes, does not need to specify the number of the classified points, and is self-determined. As another embodiment, other clustering algorithms may be used to classify users, such as WAVECLUSTER, ROCK, BICH, K-PROTOTYPES, DENCLUE, OPTIRID, CLIQUE, CURE, CLARANS, and the like.
The measurement formulas for the sample points used in the clustering algorithm mainly include manhattan distance, euclidean distance, minkowski distance, mahalanobis distance, etc., and if not specified, euclidean metric (euclidean metric) is generally used by default, and is also called euclidean distance. In mathematics, the euclidean distance or euclidean metric is the "normal" (i.e., straight line) distance between two points in euclidean space where, for two sample points in the n dimension, x ═ x1,x2,...,xn) And y ═ y1,y2,...,yn) The Euclidean distance Dist (x, y) is:
5) and counting the equipment list bound by the host where each type of user is located in the user pool to form the equipment pool of each type of user in the user pool, and generating a recommendation list for the equipment which is not purchased by the user in the corresponding equipment pool and recommending the recommendation list to the user for a certain user under each type of user in the user pool.
In this embodiment, in each device pool, the devices in the device pool are sorted in a descending order according to the number of times that the user clicks the devices, and a recommendation list recommended to the user is: and (4) recommendation lists generated by N devices which are clicked frequently but not purchased by the user in the device pool. As another embodiment, the purchase quantity of the device in the class of users may be used as a sorting criterion, and the device that the purchase quantity is correct but the user does not purchase is recommended, which is not limited in the present invention.
The invention adopts a double-layer clustering algorithm, which is obviously superior to single-layer clustering. The calculation cost is effectively saved, because single-layer clustering faces the operation records of the host data and the user at the same time, if the host data and the user data form a matrix, clustering is carried out by calculating the similarity between the matrices, the calculation amount of the similarity between phasors is obviously calculated twice and is higher than that of double-layer clustering, in order to make the classification more refined, the classification of the user is not only based on equipment under the host, but also considers the click operation habit of recent equipment of the user, the similarity of the similar user is higher, and therefore effective recommendation is achieved.
The embodiment of the intelligent terminal comprises:
the marketing recommendation system of the intelligent terminal and the intelligent home equipment provided by the embodiment is different in that in the marketing recommendation system of the intelligent home equipment, the control terminal processes the acquired data and generates a recommendation list, the recommendation list is displayed in the mobile terminal and is two independent devices capable of communicating, and the intelligent terminal integrates acquisition, processing and display.
The intelligent terminal comprises an input and output device, a memory, a processor and a computer program which is stored in the memory and can run on the processor, and the processor realizes the marketing recommendation method of the intelligent household equipment when executing the computer program.
The specific implementation process of the marketing recommendation method for the smart home devices is already introduced in the marketing recommendation system embodiment of the smart home devices, and is not described herein again.
The embodiment of the marketing recommendation method of the intelligent household equipment comprises the following steps:
the marketing recommendation method for the smart home equipment provided by the embodiment comprises the following steps:
classifying all the hosts by adopting a clustering algorithm according to the equipment list bound by each host;
counting users under each type of host to form a user pool of each type of host;
for a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm according to the recent equipment clicking times of the user; classifying the users in the user pool by adopting a clustering algorithm according to the recommended vector of each user in the user pool;
counting an equipment list bound by a host where each type of user is located in the user pool to form an equipment pool of each type of user in the user pool;
and for a certain user under each type of users in the user pool, recommending a recommendation list generated by the equipment which is not purchased by the user in the corresponding equipment pool to the user.
The specific implementation process of the marketing recommendation method for the smart home devices is already introduced in the marketing recommendation system embodiment of the smart home devices, and is not described herein again.
Claims (8)
1. A marketing recommendation method for intelligent household equipment is characterized by comprising the following steps:
classifying all the hosts by adopting a clustering algorithm according to the equipment list bound by each host;
counting users under each type of host to form a user pool of each type of host;
for a user pool of a certain type of host, obtaining a recommendation vector of each user in the user pool by adopting a time attenuation algorithm according to the recent equipment clicking times of the user;
classifying the users in the user pool by adopting a clustering algorithm according to the recommended vector of each user in the user pool;
counting an equipment list bound by a host where each type of user is located in the user pool to form an equipment pool of each type of user in the user pool;
and for a certain user under each type of users in the user pool, recommending a recommendation list generated by the equipment which is not purchased by the user in the corresponding equipment pool to the user.
2. The marketing recommendation method for smart home devices according to claim 1, wherein the time decay algorithm is as follows:
wherein Y is the recommended vector of the user, G is the time attenuation coefficient, x1,x2,...,xTThe number of clicks of each device of the user is 1 day, 2 days, … … days and T days from the recommended day, X is the accumulated number of clicks of each device of the user in the T day, and alpha is a weight coefficient for regulating and controlling the proportion of X.
3. The marketing recommendation method for the smart home devices according to claim 1 or 2, wherein the recommendation list is: recommendation lists generated by devices in the device pool that are clicked on frequently but not purchased by the user.
4. The marketing recommendation method for the smart home devices according to claim 1 or 2, wherein users unbound by the host are filtered out when counting users under each type of host.
5. The marketing recommendation method for the smart home devices according to claim 1 or 2, wherein the hosts are classified by using a CURE clustering algorithm.
6. The marketing recommendation method for the smart home devices according to claim 1 or 2, wherein the DBSCAN clustering algorithm is adopted to classify the users in the user pool.
7. A marketing recommendation system of intelligent household equipment comprises a mobile terminal and a control terminal, wherein the mobile terminal comprises input and output equipment and a communication module used for communicating with the control terminal; the control terminal comprises a communication device for communicating with the mobile terminal, a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the marketing recommendation method of the smart home device according to any one of claims 1 to 6 when executing the computer program.
8. An intelligent terminal comprising an input-output device, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the marketing recommendation method for smart home devices according to any one of claims 1 to 6 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010067606.2A CN112052380A (en) | 2020-01-20 | 2020-01-20 | Intelligent terminal, and marketing recommendation method and system of intelligent household equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010067606.2A CN112052380A (en) | 2020-01-20 | 2020-01-20 | Intelligent terminal, and marketing recommendation method and system of intelligent household equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112052380A true CN112052380A (en) | 2020-12-08 |
Family
ID=73609145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010067606.2A Pending CN112052380A (en) | 2020-01-20 | 2020-01-20 | Intelligent terminal, and marketing recommendation method and system of intelligent household equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112052380A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080243637A1 (en) * | 2007-03-30 | 2008-10-02 | Chan James D | Recommendation system with cluster-based filtering of recommendations |
| CN105426548A (en) * | 2015-12-29 | 2016-03-23 | 海信集团有限公司 | Video recommendation method and device based on multiple users |
| CN108052639A (en) * | 2017-12-21 | 2018-05-18 | 中国联合网络通信集团有限公司 | Industry user based on carrier data recommends method and device |
| CN110140135A (en) * | 2017-08-28 | 2019-08-16 | 北京嘀嘀无限科技发展有限公司 | Information processing method, information processing system and information processing unit |
| CN110517114A (en) * | 2019-08-21 | 2019-11-29 | 广州云徙科技有限公司 | A kind of information-pushing method and system based on community discovery algorithm |
-
2020
- 2020-01-20 CN CN202010067606.2A patent/CN112052380A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080243637A1 (en) * | 2007-03-30 | 2008-10-02 | Chan James D | Recommendation system with cluster-based filtering of recommendations |
| CN105426548A (en) * | 2015-12-29 | 2016-03-23 | 海信集团有限公司 | Video recommendation method and device based on multiple users |
| CN110140135A (en) * | 2017-08-28 | 2019-08-16 | 北京嘀嘀无限科技发展有限公司 | Information processing method, information processing system and information processing unit |
| CN108052639A (en) * | 2017-12-21 | 2018-05-18 | 中国联合网络通信集团有限公司 | Industry user based on carrier data recommends method and device |
| CN110517114A (en) * | 2019-08-21 | 2019-11-29 | 广州云徙科技有限公司 | A kind of information-pushing method and system based on community discovery algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245687B (en) | User classification method and device | |
| CN109165301A (en) | Video cover selection method, device and computer readable storage medium | |
| CN107305637B (en) | Data clustering method and device based on K-Means algorithm | |
| CN105005589A (en) | Text classification method and text classification device | |
| CN110909222B (en) | User portrait establishing method and device based on clustering, medium and electronic equipment | |
| CN111143697B (en) | Content recommendation method and related device | |
| CN103886068A (en) | Data processing method and device for Internet user behavior analysis | |
| CN105740480B (en) | Air ticket recommended method and system | |
| CN109903086A (en) | A kind of similar crowd's extended method, device and electronic equipment | |
| CN112488863A (en) | Dangerous seed recommendation method and related equipment in user cold start scene | |
| CN116205675B (en) | Data acquisition method and device based on thread division | |
| WO2018006631A1 (en) | User level automatic segmentation method and system | |
| CN111611496A (en) | Product recommendation method and device | |
| CN110895706B (en) | Method and device for acquiring target cluster number and computer system | |
| CN115631012A (en) | Target recommendation method and device | |
| CN113159213A (en) | Service distribution method, device and equipment | |
| CN108062415A (en) | Statistical method, device and the terminal device of user behavior | |
| CN112052380A (en) | Intelligent terminal, and marketing recommendation method and system of intelligent household equipment | |
| CN116701772A (en) | Data recommendation method and device, computer readable storage medium and electronic equipment | |
| CN111445280A (en) | Model generation method, restaurant ranking method, system, device and medium | |
| CN111309608B (en) | Test case selection method and device, electronic equipment and readable storage medium | |
| CN112540819B (en) | Method for automatically generating recommended detailed page and form page according to query page | |
| CN114461899A (en) | Collaborative filtering recommendation method and device for user, electronic equipment and storage medium | |
| CN107423759A (en) | Low-dimensional On Successive Projection Pursuit Clustering Model integrated evaluating method, device and application | |
| CN114168824A (en) | Cold and hot data separation method, system, equipment and medium based on machine learning |
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 |